Upgrade bmatcuk/doublestar to 1.3.1 (#586)

This commit is contained in:
bnkai
2020-06-03 04:46:12 +03:00
committed by GitHub
parent dcf7233e38
commit f255104951
65 changed files with 411 additions and 15209 deletions

3
go.mod
View File

@@ -2,9 +2,8 @@ module github.com/stashapp/stash
require (
github.com/99designs/gqlgen v0.9.0
github.com/PuerkitoBio/goquery v1.5.0
github.com/antchfx/htmlquery v1.2.3
github.com/bmatcuk/doublestar v1.1.5
github.com/bmatcuk/doublestar v1.3.1
github.com/disintegration/imaging v1.6.0
github.com/go-chi/chi v4.0.2+incompatible
github.com/gobuffalo/packr/v2 v2.0.2

13
go.sum
View File

@@ -16,8 +16,6 @@ github.com/Masterminds/semver v1.4.2/go.mod h1:MB6lktGJrhw8PrUyiEoblNEGEQ+RzHPF0
github.com/Microsoft/go-winio v0.4.11/go.mod h1:VhR8bwka0BXejwEJY73c50VrPtXAaKcyvVC4A4RozmA=
github.com/Nvveen/Gotty v0.0.0-20120604004816-cd527374f1e5/go.mod h1:lmUJ/7eu/Q8D7ML55dXQrVaamCz2vxCfdQBasLZfHKk=
github.com/OneOfOne/xxhash v1.2.2/go.mod h1:HSdplMjZKSmBqAxg5vPj2TmRDmfkzw+cTzAElWljhcU=
github.com/PuerkitoBio/goquery v1.5.0 h1:uGvmFXOA73IKluu/F84Xd1tt/z07GYm8X49XKHP7EJk=
github.com/PuerkitoBio/goquery v1.5.0/go.mod h1:qD2PgZ9lccMbQlc7eEOjaeRlFQON7xY8kdmcsrnKqMg=
github.com/Shopify/sarama v1.19.0/go.mod h1:FVkBWblsNy7DGZRfXLU0O9RCGt5g3g3yEuWXgklEdEo=
github.com/Shopify/toxiproxy v2.1.4+incompatible/go.mod h1:OXgGpZ6Cli1/URJOF1DMxUHB2q5Ap20/P/eIdh4G0pI=
github.com/agnivade/levenshtein v1.0.1 h1:3oJU7J3FGFmyhn8KHjmVaZCN5hxTr7GxgRue+sxIXdQ=
@@ -27,15 +25,9 @@ github.com/alecthomas/template v0.0.0-20160405071501-a0175ee3bccc/go.mod h1:LOuy
github.com/alecthomas/units v0.0.0-20151022065526-2efee857e7cf/go.mod h1:ybxpYRFXyAe+OPACYpWeL0wqObRcbAqCMya13uyzqw0=
github.com/andreyvit/diff v0.0.0-20170406064948-c7f18ee00883 h1:bvNMNQO63//z+xNgfBlViaCIJKLlCJ6/fmUseuG0wVQ=
github.com/andreyvit/diff v0.0.0-20170406064948-c7f18ee00883/go.mod h1:rCTlJbsFo29Kk6CurOXKm700vrz8f0KW0JNfpkRJY/8=
github.com/andybalholm/cascadia v1.0.0 h1:hOCXnnZ5A+3eVDX8pvgl4kofXv2ELss0bKcqRySc45o=
github.com/andybalholm/cascadia v1.0.0/go.mod h1:GsXiBklL0woXo1j/WYWtSYYC4ouU9PqHO0sqidkEA4Y=
github.com/anmitsu/go-shlex v0.0.0-20161002113705-648efa622239/go.mod h1:2FmKhYUyUczH0OGQWaF5ceTx0UBShxjsH6f8oGKYe2c=
github.com/antchfx/htmlquery v1.2.0 h1:oKShnsGlnOHX6t4uj5OHgLKkABcJoqnXpqnscoi9Lpw=
github.com/antchfx/htmlquery v1.2.0/go.mod h1:MS9yksVSQXls00iXkiMqXr0J+umL/AmxXKuP28SUJM8=
github.com/antchfx/htmlquery v1.2.3 h1:sP3NFDneHx2stfNXCKbhHFo8XgNjCACnU/4AO5gWz6M=
github.com/antchfx/htmlquery v1.2.3/go.mod h1:B0ABL+F5irhhMWg54ymEZinzMSi0Kt3I2if0BLYa3V0=
github.com/antchfx/xpath v1.1.2 h1:YziPrtM0gEJBnhdUGxYcIVYXZ8FXbtbovxOi+UW/yWQ=
github.com/antchfx/xpath v1.1.2/go.mod h1:Yee4kTMuNiPYJ7nSNorELQMr1J33uOpXDMByNYhvtNk=
github.com/antchfx/xpath v1.1.6 h1:6sVh6hB5T6phw1pFpHRQ+C4bd8sNI+O58flqtg7h0R0=
github.com/antchfx/xpath v1.1.6/go.mod h1:Yee4kTMuNiPYJ7nSNorELQMr1J33uOpXDMByNYhvtNk=
github.com/apache/thrift v0.0.0-20180902110319-2566ecd5d999/go.mod h1:cp2SuWMxlEZw2r+iP2GNCdIi4C1qmUzdZFSVb+bacwQ=
@@ -45,8 +37,8 @@ github.com/aws/aws-sdk-go v1.17.7/go.mod h1:KmX6BPdI08NWTb3/sm4ZGu5ShLoqVDhKgpiN
github.com/beorn7/perks v0.0.0-20180321164747-3a771d992973/go.mod h1:Dwedo/Wpr24TaqPxmxbtue+5NUziq4I4S80YR8gNf3Q=
github.com/beorn7/perks v1.0.0/go.mod h1:KWe93zE9D1o94FZ5RNwFwVgaQK1VOXiVxmqh+CedLV8=
github.com/bitly/go-hostpool v0.0.0-20171023180738-a3a6125de932/go.mod h1:NOuUCSz6Q9T7+igc/hlvDOUdtWKryOrtFyIVABv/p7k=
github.com/bmatcuk/doublestar v1.1.5 h1:2bNwBOmhyFEFcoB3tGvTD5xanq+4kyOZlB8wFYbMjkk=
github.com/bmatcuk/doublestar v1.1.5/go.mod h1:wiQtGV+rzVYxB7WIlirSN++5HPtPlXEo9MEoZQC/PmE=
github.com/bmatcuk/doublestar v1.3.1 h1:rT8rxDPsavp9G+4ZULzqhhUSaI/OPsTZNG88Z3i0xvY=
github.com/bmatcuk/doublestar v1.3.1/go.mod h1:wiQtGV+rzVYxB7WIlirSN++5HPtPlXEo9MEoZQC/PmE=
github.com/bmizerany/assert v0.0.0-20160611221934-b7ed37b82869/go.mod h1:Ekp36dRnpXw/yCqJaO+ZrUyxD+3VXMFFr56k5XYrpB4=
github.com/bradfitz/go-smtpd v0.0.0-20170404230938-deb6d6237625/go.mod h1:HYsPBTaaSFSlLx/70C2HPIMNZpVV8+vt/A+FMnYP11g=
github.com/cespare/xxhash v1.1.0/go.mod h1:XrSqR1VqqWfGrhpAt58auRo0WTKS1nRRg3ghfAqPWnc=
@@ -656,7 +648,6 @@ golang.org/x/lint v0.0.0-20181026193005-c67002cb31c3/go.mod h1:UVdnD1Gm6xHRNCYTk
golang.org/x/lint v0.0.0-20190227174305-5b3e6a55c961/go.mod h1:wehouNa3lNwaWXcvxsM5YxQ5yQlVC4a0KAMCusXpPoU=
golang.org/x/lint v0.0.0-20190301231843-5614ed5bae6f/go.mod h1:UVdnD1Gm6xHRNCYTkRU2/jEulfH38KcIWyp/GAMgvoE=
golang.org/x/lint v0.0.0-20190313153728-d0100b6bd8b3/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
golang.org/x/net v0.0.0-20180218175443-cbe0f9307d01/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
golang.org/x/net v0.0.0-20180724234803-3673e40ba225/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
golang.org/x/net v0.0.0-20180816102801-aaf60122140d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
golang.org/x/net v0.0.0-20180826012351-8a410e7b638d/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=

View File

@@ -1 +0,0 @@
testdata/* linguist-vendored

View File

@@ -1,16 +0,0 @@
# editor temporary files
*.sublime-*
.DS_Store
*.swp
#*.*#
tags
# direnv config
.env*
# test binaries
*.test
# coverage and profilte outputs
*.out

View File

@@ -1,16 +0,0 @@
language: go
go:
- 1.1
- 1.2.x
- 1.3.x
- 1.4.x
- 1.5.x
- 1.6.x
- 1.7.x
- 1.8.x
- 1.9.x
- "1.10.x"
- 1.11.x
- tip

View File

@@ -1,12 +0,0 @@
Copyright (c) 2012-2016, Martin Angers & Contributors
All rights reserved.
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
* Neither the name of the author nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

View File

@@ -1,179 +0,0 @@
# goquery - a little like that j-thing, only in Go
[![build status](https://secure.travis-ci.org/PuerkitoBio/goquery.svg?branch=master)](http://travis-ci.org/PuerkitoBio/goquery) [![GoDoc](https://godoc.org/github.com/PuerkitoBio/goquery?status.png)](http://godoc.org/github.com/PuerkitoBio/goquery) [![Sourcegraph Badge](https://sourcegraph.com/github.com/PuerkitoBio/goquery/-/badge.svg)](https://sourcegraph.com/github.com/PuerkitoBio/goquery?badge)
goquery brings a syntax and a set of features similar to [jQuery][] to the [Go language][go]. It is based on Go's [net/html package][html] and the CSS Selector library [cascadia][]. Since the net/html parser returns nodes, and not a full-featured DOM tree, jQuery's stateful manipulation functions (like height(), css(), detach()) have been left off.
Also, because the net/html parser requires UTF-8 encoding, so does goquery: it is the caller's responsibility to ensure that the source document provides UTF-8 encoded HTML. See the [wiki][] for various options to do this.
Syntax-wise, it is as close as possible to jQuery, with the same function names when possible, and that warm and fuzzy chainable interface. jQuery being the ultra-popular library that it is, I felt that writing a similar HTML-manipulating library was better to follow its API than to start anew (in the same spirit as Go's `fmt` package), even though some of its methods are less than intuitive (looking at you, [index()][index]...).
## Table of Contents
* [Installation](#installation)
* [Changelog](#changelog)
* [API](#api)
* [Examples](#examples)
* [Related Projects](#related-projects)
* [Support](#support)
* [License](#license)
## Installation
Please note that because of the net/html dependency, goquery requires Go1.1+.
$ go get github.com/PuerkitoBio/goquery
(optional) To run unit tests:
$ cd $GOPATH/src/github.com/PuerkitoBio/goquery
$ go test
(optional) To run benchmarks (warning: it runs for a few minutes):
$ cd $GOPATH/src/github.com/PuerkitoBio/goquery
$ go test -bench=".*"
## Changelog
**Note that goquery's API is now stable, and will not break.**
* **2018-11-15 (v1.5.0)** : Go module support (thanks @Zaba505).
* **2018-06-07 (v1.4.1)** : Add `NewDocumentFromReader` examples.
* **2018-03-24 (v1.4.0)** : Deprecate `NewDocument(url)` and `NewDocumentFromResponse(response)`.
* **2018-01-28 (v1.3.0)** : Add `ToEnd` constant to `Slice` until the end of the selection (thanks to @davidjwilkins for raising the issue).
* **2018-01-11 (v1.2.0)** : Add `AddBack*` and deprecate `AndSelf` (thanks to @davidjwilkins).
* **2017-02-12 (v1.1.0)** : Add `SetHtml` and `SetText` (thanks to @glebtv).
* **2016-12-29 (v1.0.2)** : Optimize allocations for `Selection.Text` (thanks to @radovskyb).
* **2016-08-28 (v1.0.1)** : Optimize performance for large documents.
* **2016-07-27 (v1.0.0)** : Tag version 1.0.0.
* **2016-06-15** : Invalid selector strings internally compile to a `Matcher` implementation that never matches any node (instead of a panic). So for example, `doc.Find("~")` returns an empty `*Selection` object.
* **2016-02-02** : Add `NodeName` utility function similar to the DOM's `nodeName` property. It returns the tag name of the first element in a selection, and other relevant values of non-element nodes (see godoc for details). Add `OuterHtml` utility function similar to the DOM's `outerHTML` property (named `OuterHtml` in small caps for consistency with the existing `Html` method on the `Selection`).
* **2015-04-20** : Add `AttrOr` helper method to return the attribute's value or a default value if absent. Thanks to [piotrkowalczuk][piotr].
* **2015-02-04** : Add more manipulation functions - Prepend* - thanks again to [Andrew Stone][thatguystone].
* **2014-11-28** : Add more manipulation functions - ReplaceWith*, Wrap* and Unwrap - thanks again to [Andrew Stone][thatguystone].
* **2014-11-07** : Add manipulation functions (thanks to [Andrew Stone][thatguystone]) and `*Matcher` functions, that receive compiled cascadia selectors instead of selector strings, thus avoiding potential panics thrown by goquery via `cascadia.MustCompile` calls. This results in better performance (selectors can be compiled once and reused) and more idiomatic error handling (you can handle cascadia's compilation errors, instead of recovering from panics, which had been bugging me for a long time). Note that the actual type expected is a `Matcher` interface, that `cascadia.Selector` implements. Other matcher implementations could be used.
* **2014-11-06** : Change import paths of net/html to golang.org/x/net/html (see https://groups.google.com/forum/#!topic/golang-nuts/eD8dh3T9yyA). Make sure to update your code to use the new import path too when you call goquery with `html.Node`s.
* **v0.3.2** : Add `NewDocumentFromReader()` (thanks jweir) which allows creating a goquery document from an io.Reader.
* **v0.3.1** : Add `NewDocumentFromResponse()` (thanks assassingj) which allows creating a goquery document from an http response.
* **v0.3.0** : Add `EachWithBreak()` which allows to break out of an `Each()` loop by returning false. This function was added instead of changing the existing `Each()` to avoid breaking compatibility.
* **v0.2.1** : Make go-getable, now that [go.net/html is Go1.0-compatible][gonet] (thanks to @matrixik for pointing this out).
* **v0.2.0** : Add support for negative indices in Slice(). **BREAKING CHANGE** `Document.Root` is removed, `Document` is now a `Selection` itself (a selection of one, the root element, just like `Document.Root` was before). Add jQuery's Closest() method.
* **v0.1.1** : Add benchmarks to use as baseline for refactorings, refactor Next...() and Prev...() methods to use the new html package's linked list features (Next/PrevSibling, FirstChild). Good performance boost (40+% in some cases).
* **v0.1.0** : Initial release.
## API
goquery exposes two structs, `Document` and `Selection`, and the `Matcher` interface. Unlike jQuery, which is loaded as part of a DOM document, and thus acts on its containing document, goquery doesn't know which HTML document to act upon. So it needs to be told, and that's what the `Document` type is for. It holds the root document node as the initial Selection value to manipulate.
jQuery often has many variants for the same function (no argument, a selector string argument, a jQuery object argument, a DOM element argument, ...). Instead of exposing the same features in goquery as a single method with variadic empty interface arguments, statically-typed signatures are used following this naming convention:
* When the jQuery equivalent can be called with no argument, it has the same name as jQuery for the no argument signature (e.g.: `Prev()`), and the version with a selector string argument is called `XxxFiltered()` (e.g.: `PrevFiltered()`)
* When the jQuery equivalent **requires** one argument, the same name as jQuery is used for the selector string version (e.g.: `Is()`)
* The signatures accepting a jQuery object as argument are defined in goquery as `XxxSelection()` and take a `*Selection` object as argument (e.g.: `FilterSelection()`)
* The signatures accepting a DOM element as argument in jQuery are defined in goquery as `XxxNodes()` and take a variadic argument of type `*html.Node` (e.g.: `FilterNodes()`)
* The signatures accepting a function as argument in jQuery are defined in goquery as `XxxFunction()` and take a function as argument (e.g.: `FilterFunction()`)
* The goquery methods that can be called with a selector string have a corresponding version that take a `Matcher` interface and are defined as `XxxMatcher()` (e.g.: `IsMatcher()`)
Utility functions that are not in jQuery but are useful in Go are implemented as functions (that take a `*Selection` as parameter), to avoid a potential naming clash on the `*Selection`'s methods (reserved for jQuery-equivalent behaviour).
The complete [godoc reference documentation can be found here][doc].
Please note that Cascadia's selectors do not necessarily match all supported selectors of jQuery (Sizzle). See the [cascadia project][cascadia] for details. Invalid selector strings compile to a `Matcher` that fails to match any node. Behaviour of the various functions that take a selector string as argument follows from that fact, e.g. (where `~` is an invalid selector string):
* `Find("~")` returns an empty selection because the selector string doesn't match anything.
* `Add("~")` returns a new selection that holds the same nodes as the original selection, because it didn't add any node (selector string didn't match anything).
* `ParentsFiltered("~")` returns an empty selection because the selector string doesn't match anything.
* `ParentsUntil("~")` returns all parents of the selection because the selector string didn't match any element to stop before the top element.
## Examples
See some tips and tricks in the [wiki][].
Adapted from example_test.go:
```Go
package main
import (
"fmt"
"log"
"net/http"
"github.com/PuerkitoBio/goquery"
)
func ExampleScrape() {
// Request the HTML page.
res, err := http.Get("http://metalsucks.net")
if err != nil {
log.Fatal(err)
}
defer res.Body.Close()
if res.StatusCode != 200 {
log.Fatalf("status code error: %d %s", res.StatusCode, res.Status)
}
// Load the HTML document
doc, err := goquery.NewDocumentFromReader(res.Body)
if err != nil {
log.Fatal(err)
}
// Find the review items
doc.Find(".sidebar-reviews article .content-block").Each(func(i int, s *goquery.Selection) {
// For each item found, get the band and title
band := s.Find("a").Text()
title := s.Find("i").Text()
fmt.Printf("Review %d: %s - %s\n", i, band, title)
})
}
func main() {
ExampleScrape()
}
```
## Related Projects
- [Goq][goq], an HTML deserialization and scraping library based on goquery and struct tags.
- [andybalholm/cascadia][cascadia], the CSS selector library used by goquery.
- [suntong/cascadia][cascadiacli], a command-line interface to the cascadia CSS selector library, useful to test selectors.
- [asciimoo/colly](https://github.com/asciimoo/colly), a lightning fast and elegant Scraping Framework
- [gnulnx/goperf](https://github.com/gnulnx/goperf), a website performance test tool that also fetches static assets.
- [MontFerret/ferret](https://github.com/MontFerret/ferret), declarative web scraping.
## Support
There are a number of ways you can support the project:
* Use it, star it, build something with it, spread the word!
- If you do build something open-source or otherwise publicly-visible, let me know so I can add it to the [Related Projects](#related-projects) section!
* Raise issues to improve the project (note: doc typos and clarifications are issues too!)
- Please search existing issues before opening a new one - it may have already been adressed.
* Pull requests: please discuss new code in an issue first, unless the fix is really trivial.
- Make sure new code is tested.
- Be mindful of existing code - PRs that break existing code have a high probability of being declined, unless it fixes a serious issue.
If you desperately want to send money my way, I have a BuyMeACoffee.com page:
<a href="https://www.buymeacoffee.com/mna" target="_blank"><img src="https://www.buymeacoffee.com/assets/img/custom_images/orange_img.png" alt="Buy Me A Coffee" style="height: 41px !important;width: 174px !important;box-shadow: 0px 3px 2px 0px rgba(190, 190, 190, 0.5) !important;-webkit-box-shadow: 0px 3px 2px 0px rgba(190, 190, 190, 0.5) !important;" ></a>
## License
The [BSD 3-Clause license][bsd], the same as the [Go language][golic]. Cascadia's license is [here][caslic].
[jquery]: http://jquery.com/
[go]: http://golang.org/
[cascadia]: https://github.com/andybalholm/cascadia
[cascadiacli]: https://github.com/suntong/cascadia
[bsd]: http://opensource.org/licenses/BSD-3-Clause
[golic]: http://golang.org/LICENSE
[caslic]: https://github.com/andybalholm/cascadia/blob/master/LICENSE
[doc]: http://godoc.org/github.com/PuerkitoBio/goquery
[index]: http://api.jquery.com/index/
[gonet]: https://github.com/golang/net/
[html]: http://godoc.org/golang.org/x/net/html
[wiki]: https://github.com/PuerkitoBio/goquery/wiki/Tips-and-tricks
[thatguystone]: https://github.com/thatguystone
[piotr]: https://github.com/piotrkowalczuk
[goq]: https://github.com/andrewstuart/goq

View File

@@ -1,124 +0,0 @@
package goquery
import (
"golang.org/x/net/html"
)
const (
maxUint = ^uint(0)
maxInt = int(maxUint >> 1)
// ToEnd is a special index value that can be used as end index in a call
// to Slice so that all elements are selected until the end of the Selection.
// It is equivalent to passing (*Selection).Length().
ToEnd = maxInt
)
// First reduces the set of matched elements to the first in the set.
// It returns a new Selection object, and an empty Selection object if the
// the selection is empty.
func (s *Selection) First() *Selection {
return s.Eq(0)
}
// Last reduces the set of matched elements to the last in the set.
// It returns a new Selection object, and an empty Selection object if
// the selection is empty.
func (s *Selection) Last() *Selection {
return s.Eq(-1)
}
// Eq reduces the set of matched elements to the one at the specified index.
// If a negative index is given, it counts backwards starting at the end of the
// set. It returns a new Selection object, and an empty Selection object if the
// index is invalid.
func (s *Selection) Eq(index int) *Selection {
if index < 0 {
index += len(s.Nodes)
}
if index >= len(s.Nodes) || index < 0 {
return newEmptySelection(s.document)
}
return s.Slice(index, index+1)
}
// Slice reduces the set of matched elements to a subset specified by a range
// of indices. The start index is 0-based and indicates the index of the first
// element to select. The end index is 0-based and indicates the index at which
// the elements stop being selected (the end index is not selected).
//
// The indices may be negative, in which case they represent an offset from the
// end of the selection.
//
// The special value ToEnd may be specified as end index, in which case all elements
// until the end are selected. This works both for a positive and negative start
// index.
func (s *Selection) Slice(start, end int) *Selection {
if start < 0 {
start += len(s.Nodes)
}
if end == ToEnd {
end = len(s.Nodes)
} else if end < 0 {
end += len(s.Nodes)
}
return pushStack(s, s.Nodes[start:end])
}
// Get retrieves the underlying node at the specified index.
// Get without parameter is not implemented, since the node array is available
// on the Selection object.
func (s *Selection) Get(index int) *html.Node {
if index < 0 {
index += len(s.Nodes) // Negative index gets from the end
}
return s.Nodes[index]
}
// Index returns the position of the first element within the Selection object
// relative to its sibling elements.
func (s *Selection) Index() int {
if len(s.Nodes) > 0 {
return newSingleSelection(s.Nodes[0], s.document).PrevAll().Length()
}
return -1
}
// IndexSelector returns the position of the first element within the
// Selection object relative to the elements matched by the selector, or -1 if
// not found.
func (s *Selection) IndexSelector(selector string) int {
if len(s.Nodes) > 0 {
sel := s.document.Find(selector)
return indexInSlice(sel.Nodes, s.Nodes[0])
}
return -1
}
// IndexMatcher returns the position of the first element within the
// Selection object relative to the elements matched by the matcher, or -1 if
// not found.
func (s *Selection) IndexMatcher(m Matcher) int {
if len(s.Nodes) > 0 {
sel := s.document.FindMatcher(m)
return indexInSlice(sel.Nodes, s.Nodes[0])
}
return -1
}
// IndexOfNode returns the position of the specified node within the Selection
// object, or -1 if not found.
func (s *Selection) IndexOfNode(node *html.Node) int {
return indexInSlice(s.Nodes, node)
}
// IndexOfSelection returns the position of the first node in the specified
// Selection object within this Selection object, or -1 if not found.
func (s *Selection) IndexOfSelection(sel *Selection) int {
if sel != nil && len(sel.Nodes) > 0 {
return indexInSlice(s.Nodes, sel.Nodes[0])
}
return -1
}

View File

@@ -1,123 +0,0 @@
// Copyright (c) 2012-2016, Martin Angers & Contributors
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// * Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation and/or
// other materials provided with the distribution.
// * Neither the name of the author nor the names of its contributors may be used to
// endorse or promote products derived from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS
// OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
// AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
// CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
// DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY
// WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
/*
Package goquery implements features similar to jQuery, including the chainable
syntax, to manipulate and query an HTML document.
It brings a syntax and a set of features similar to jQuery to the Go language.
It is based on Go's net/html package and the CSS Selector library cascadia.
Since the net/html parser returns nodes, and not a full-featured DOM
tree, jQuery's stateful manipulation functions (like height(), css(), detach())
have been left off.
Also, because the net/html parser requires UTF-8 encoding, so does goquery: it is
the caller's responsibility to ensure that the source document provides UTF-8 encoded HTML.
See the repository's wiki for various options on how to do this.
Syntax-wise, it is as close as possible to jQuery, with the same method names when
possible, and that warm and fuzzy chainable interface. jQuery being the
ultra-popular library that it is, writing a similar HTML-manipulating
library was better to follow its API than to start anew (in the same spirit as
Go's fmt package), even though some of its methods are less than intuitive (looking
at you, index()...).
It is hosted on GitHub, along with additional documentation in the README.md
file: https://github.com/puerkitobio/goquery
Please note that because of the net/html dependency, goquery requires Go1.1+.
The various methods are split into files based on the category of behavior.
The three dots (...) indicate that various "overloads" are available.
* array.go : array-like positional manipulation of the selection.
- Eq()
- First()
- Get()
- Index...()
- Last()
- Slice()
* expand.go : methods that expand or augment the selection's set.
- Add...()
- AndSelf()
- Union(), which is an alias for AddSelection()
* filter.go : filtering methods, that reduce the selection's set.
- End()
- Filter...()
- Has...()
- Intersection(), which is an alias of FilterSelection()
- Not...()
* iteration.go : methods to loop over the selection's nodes.
- Each()
- EachWithBreak()
- Map()
* manipulation.go : methods for modifying the document
- After...()
- Append...()
- Before...()
- Clone()
- Empty()
- Prepend...()
- Remove...()
- ReplaceWith...()
- Unwrap()
- Wrap...()
- WrapAll...()
- WrapInner...()
* property.go : methods that inspect and get the node's properties values.
- Attr*(), RemoveAttr(), SetAttr()
- AddClass(), HasClass(), RemoveClass(), ToggleClass()
- Html()
- Length()
- Size(), which is an alias for Length()
- Text()
* query.go : methods that query, or reflect, a node's identity.
- Contains()
- Is...()
* traversal.go : methods to traverse the HTML document tree.
- Children...()
- Contents()
- Find...()
- Next...()
- Parent[s]...()
- Prev...()
- Siblings...()
* type.go : definition of the types exposed by goquery.
- Document
- Selection
- Matcher
* utilities.go : definition of helper functions (and not methods on a *Selection)
that are not part of jQuery, but are useful to goquery.
- NodeName
- OuterHtml
*/
package goquery

View File

@@ -1,70 +0,0 @@
package goquery
import "golang.org/x/net/html"
// Add adds the selector string's matching nodes to those in the current
// selection and returns a new Selection object.
// The selector string is run in the context of the document of the current
// Selection object.
func (s *Selection) Add(selector string) *Selection {
return s.AddNodes(findWithMatcher([]*html.Node{s.document.rootNode}, compileMatcher(selector))...)
}
// AddMatcher adds the matcher's matching nodes to those in the current
// selection and returns a new Selection object.
// The matcher is run in the context of the document of the current
// Selection object.
func (s *Selection) AddMatcher(m Matcher) *Selection {
return s.AddNodes(findWithMatcher([]*html.Node{s.document.rootNode}, m)...)
}
// AddSelection adds the specified Selection object's nodes to those in the
// current selection and returns a new Selection object.
func (s *Selection) AddSelection(sel *Selection) *Selection {
if sel == nil {
return s.AddNodes()
}
return s.AddNodes(sel.Nodes...)
}
// Union is an alias for AddSelection.
func (s *Selection) Union(sel *Selection) *Selection {
return s.AddSelection(sel)
}
// AddNodes adds the specified nodes to those in the
// current selection and returns a new Selection object.
func (s *Selection) AddNodes(nodes ...*html.Node) *Selection {
return pushStack(s, appendWithoutDuplicates(s.Nodes, nodes, nil))
}
// AndSelf adds the previous set of elements on the stack to the current set.
// It returns a new Selection object containing the current Selection combined
// with the previous one.
// Deprecated: This function has been deprecated and is now an alias for AddBack().
func (s *Selection) AndSelf() *Selection {
return s.AddBack()
}
// AddBack adds the previous set of elements on the stack to the current set.
// It returns a new Selection object containing the current Selection combined
// with the previous one.
func (s *Selection) AddBack() *Selection {
return s.AddSelection(s.prevSel)
}
// AddBackFiltered reduces the previous set of elements on the stack to those that
// match the selector string, and adds them to the current set.
// It returns a new Selection object containing the current Selection combined
// with the filtered previous one
func (s *Selection) AddBackFiltered(selector string) *Selection {
return s.AddSelection(s.prevSel.Filter(selector))
}
// AddBackMatcher reduces the previous set of elements on the stack to those that match
// the mateher, and adds them to the curernt set.
// It returns a new Selection object containing the current Selection combined
// with the filtered previous one
func (s *Selection) AddBackMatcher(m Matcher) *Selection {
return s.AddSelection(s.prevSel.FilterMatcher(m))
}

View File

@@ -1,163 +0,0 @@
package goquery
import "golang.org/x/net/html"
// Filter reduces the set of matched elements to those that match the selector string.
// It returns a new Selection object for this subset of matching elements.
func (s *Selection) Filter(selector string) *Selection {
return s.FilterMatcher(compileMatcher(selector))
}
// FilterMatcher reduces the set of matched elements to those that match
// the given matcher. It returns a new Selection object for this subset
// of matching elements.
func (s *Selection) FilterMatcher(m Matcher) *Selection {
return pushStack(s, winnow(s, m, true))
}
// Not removes elements from the Selection that match the selector string.
// It returns a new Selection object with the matching elements removed.
func (s *Selection) Not(selector string) *Selection {
return s.NotMatcher(compileMatcher(selector))
}
// NotMatcher removes elements from the Selection that match the given matcher.
// It returns a new Selection object with the matching elements removed.
func (s *Selection) NotMatcher(m Matcher) *Selection {
return pushStack(s, winnow(s, m, false))
}
// FilterFunction reduces the set of matched elements to those that pass the function's test.
// It returns a new Selection object for this subset of elements.
func (s *Selection) FilterFunction(f func(int, *Selection) bool) *Selection {
return pushStack(s, winnowFunction(s, f, true))
}
// NotFunction removes elements from the Selection that pass the function's test.
// It returns a new Selection object with the matching elements removed.
func (s *Selection) NotFunction(f func(int, *Selection) bool) *Selection {
return pushStack(s, winnowFunction(s, f, false))
}
// FilterNodes reduces the set of matched elements to those that match the specified nodes.
// It returns a new Selection object for this subset of elements.
func (s *Selection) FilterNodes(nodes ...*html.Node) *Selection {
return pushStack(s, winnowNodes(s, nodes, true))
}
// NotNodes removes elements from the Selection that match the specified nodes.
// It returns a new Selection object with the matching elements removed.
func (s *Selection) NotNodes(nodes ...*html.Node) *Selection {
return pushStack(s, winnowNodes(s, nodes, false))
}
// FilterSelection reduces the set of matched elements to those that match a
// node in the specified Selection object.
// It returns a new Selection object for this subset of elements.
func (s *Selection) FilterSelection(sel *Selection) *Selection {
if sel == nil {
return pushStack(s, winnowNodes(s, nil, true))
}
return pushStack(s, winnowNodes(s, sel.Nodes, true))
}
// NotSelection removes elements from the Selection that match a node in the specified
// Selection object. It returns a new Selection object with the matching elements removed.
func (s *Selection) NotSelection(sel *Selection) *Selection {
if sel == nil {
return pushStack(s, winnowNodes(s, nil, false))
}
return pushStack(s, winnowNodes(s, sel.Nodes, false))
}
// Intersection is an alias for FilterSelection.
func (s *Selection) Intersection(sel *Selection) *Selection {
return s.FilterSelection(sel)
}
// Has reduces the set of matched elements to those that have a descendant
// that matches the selector.
// It returns a new Selection object with the matching elements.
func (s *Selection) Has(selector string) *Selection {
return s.HasSelection(s.document.Find(selector))
}
// HasMatcher reduces the set of matched elements to those that have a descendant
// that matches the matcher.
// It returns a new Selection object with the matching elements.
func (s *Selection) HasMatcher(m Matcher) *Selection {
return s.HasSelection(s.document.FindMatcher(m))
}
// HasNodes reduces the set of matched elements to those that have a
// descendant that matches one of the nodes.
// It returns a new Selection object with the matching elements.
func (s *Selection) HasNodes(nodes ...*html.Node) *Selection {
return s.FilterFunction(func(_ int, sel *Selection) bool {
// Add all nodes that contain one of the specified nodes
for _, n := range nodes {
if sel.Contains(n) {
return true
}
}
return false
})
}
// HasSelection reduces the set of matched elements to those that have a
// descendant that matches one of the nodes of the specified Selection object.
// It returns a new Selection object with the matching elements.
func (s *Selection) HasSelection(sel *Selection) *Selection {
if sel == nil {
return s.HasNodes()
}
return s.HasNodes(sel.Nodes...)
}
// End ends the most recent filtering operation in the current chain and
// returns the set of matched elements to its previous state.
func (s *Selection) End() *Selection {
if s.prevSel != nil {
return s.prevSel
}
return newEmptySelection(s.document)
}
// Filter based on the matcher, and the indicator to keep (Filter) or
// to get rid of (Not) the matching elements.
func winnow(sel *Selection, m Matcher, keep bool) []*html.Node {
// Optimize if keep is requested
if keep {
return m.Filter(sel.Nodes)
}
// Use grep
return grep(sel, func(i int, s *Selection) bool {
return !m.Match(s.Get(0))
})
}
// Filter based on an array of nodes, and the indicator to keep (Filter) or
// to get rid of (Not) the matching elements.
func winnowNodes(sel *Selection, nodes []*html.Node, keep bool) []*html.Node {
if len(nodes)+len(sel.Nodes) < minNodesForSet {
return grep(sel, func(i int, s *Selection) bool {
return isInSlice(nodes, s.Get(0)) == keep
})
}
set := make(map[*html.Node]bool)
for _, n := range nodes {
set[n] = true
}
return grep(sel, func(i int, s *Selection) bool {
return set[s.Get(0)] == keep
})
}
// Filter based on a function test, and the indicator to keep (Filter) or
// to get rid of (Not) the matching elements.
func winnowFunction(sel *Selection, f func(int, *Selection) bool, keep bool) []*html.Node {
return grep(sel, func(i int, s *Selection) bool {
return f(i, s) == keep
})
}

View File

@@ -1,6 +0,0 @@
module github.com/PuerkitoBio/goquery
require (
github.com/andybalholm/cascadia v1.0.0
golang.org/x/net v0.0.0-20181114220301-adae6a3d119a
)

View File

@@ -1,5 +0,0 @@
github.com/andybalholm/cascadia v1.0.0 h1:hOCXnnZ5A+3eVDX8pvgl4kofXv2ELss0bKcqRySc45o=
github.com/andybalholm/cascadia v1.0.0/go.mod h1:GsXiBklL0woXo1j/WYWtSYYC4ouU9PqHO0sqidkEA4Y=
golang.org/x/net v0.0.0-20180218175443-cbe0f9307d01/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=
golang.org/x/net v0.0.0-20181114220301-adae6a3d119a h1:gOpx8G595UYyvj8UK4+OFyY4rx037g3fmfhe5SasG3U=
golang.org/x/net v0.0.0-20181114220301-adae6a3d119a/go.mod h1:mL1N/T3taQHkDXs73rZJwtUhF3w3ftmwwsq0BUmARs4=

View File

@@ -1,39 +0,0 @@
package goquery
// Each iterates over a Selection object, executing a function for each
// matched element. It returns the current Selection object. The function
// f is called for each element in the selection with the index of the
// element in that selection starting at 0, and a *Selection that contains
// only that element.
func (s *Selection) Each(f func(int, *Selection)) *Selection {
for i, n := range s.Nodes {
f(i, newSingleSelection(n, s.document))
}
return s
}
// EachWithBreak iterates over a Selection object, executing a function for each
// matched element. It is identical to Each except that it is possible to break
// out of the loop by returning false in the callback function. It returns the
// current Selection object.
func (s *Selection) EachWithBreak(f func(int, *Selection) bool) *Selection {
for i, n := range s.Nodes {
if !f(i, newSingleSelection(n, s.document)) {
return s
}
}
return s
}
// Map passes each element in the current matched set through a function,
// producing a slice of string holding the returned values. The function
// f is called for each element in the selection with the index of the
// element in that selection starting at 0, and a *Selection that contains
// only that element.
func (s *Selection) Map(f func(int, *Selection) string) (result []string) {
for i, n := range s.Nodes {
result = append(result, f(i, newSingleSelection(n, s.document)))
}
return result
}

View File

@@ -1,574 +0,0 @@
package goquery
import (
"strings"
"golang.org/x/net/html"
)
// After applies the selector from the root document and inserts the matched elements
// after the elements in the set of matched elements.
//
// If one of the matched elements in the selection is not currently in the
// document, it's impossible to insert nodes after it, so it will be ignored.
//
// This follows the same rules as Selection.Append.
func (s *Selection) After(selector string) *Selection {
return s.AfterMatcher(compileMatcher(selector))
}
// AfterMatcher applies the matcher from the root document and inserts the matched elements
// after the elements in the set of matched elements.
//
// If one of the matched elements in the selection is not currently in the
// document, it's impossible to insert nodes after it, so it will be ignored.
//
// This follows the same rules as Selection.Append.
func (s *Selection) AfterMatcher(m Matcher) *Selection {
return s.AfterNodes(m.MatchAll(s.document.rootNode)...)
}
// AfterSelection inserts the elements in the selection after each element in the set of matched
// elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) AfterSelection(sel *Selection) *Selection {
return s.AfterNodes(sel.Nodes...)
}
// AfterHtml parses the html and inserts it after the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) AfterHtml(html string) *Selection {
return s.AfterNodes(parseHtml(html)...)
}
// AfterNodes inserts the nodes after each element in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) AfterNodes(ns ...*html.Node) *Selection {
return s.manipulateNodes(ns, true, func(sn *html.Node, n *html.Node) {
if sn.Parent != nil {
sn.Parent.InsertBefore(n, sn.NextSibling)
}
})
}
// Append appends the elements specified by the selector to the end of each element
// in the set of matched elements, following those rules:
//
// 1) The selector is applied to the root document.
//
// 2) Elements that are part of the document will be moved to the new location.
//
// 3) If there are multiple locations to append to, cloned nodes will be
// appended to all target locations except the last one, which will be moved
// as noted in (2).
func (s *Selection) Append(selector string) *Selection {
return s.AppendMatcher(compileMatcher(selector))
}
// AppendMatcher appends the elements specified by the matcher to the end of each element
// in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) AppendMatcher(m Matcher) *Selection {
return s.AppendNodes(m.MatchAll(s.document.rootNode)...)
}
// AppendSelection appends the elements in the selection to the end of each element
// in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) AppendSelection(sel *Selection) *Selection {
return s.AppendNodes(sel.Nodes...)
}
// AppendHtml parses the html and appends it to the set of matched elements.
func (s *Selection) AppendHtml(html string) *Selection {
return s.AppendNodes(parseHtml(html)...)
}
// AppendNodes appends the specified nodes to each node in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) AppendNodes(ns ...*html.Node) *Selection {
return s.manipulateNodes(ns, false, func(sn *html.Node, n *html.Node) {
sn.AppendChild(n)
})
}
// Before inserts the matched elements before each element in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) Before(selector string) *Selection {
return s.BeforeMatcher(compileMatcher(selector))
}
// BeforeMatcher inserts the matched elements before each element in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) BeforeMatcher(m Matcher) *Selection {
return s.BeforeNodes(m.MatchAll(s.document.rootNode)...)
}
// BeforeSelection inserts the elements in the selection before each element in the set of matched
// elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) BeforeSelection(sel *Selection) *Selection {
return s.BeforeNodes(sel.Nodes...)
}
// BeforeHtml parses the html and inserts it before the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) BeforeHtml(html string) *Selection {
return s.BeforeNodes(parseHtml(html)...)
}
// BeforeNodes inserts the nodes before each element in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) BeforeNodes(ns ...*html.Node) *Selection {
return s.manipulateNodes(ns, false, func(sn *html.Node, n *html.Node) {
if sn.Parent != nil {
sn.Parent.InsertBefore(n, sn)
}
})
}
// Clone creates a deep copy of the set of matched nodes. The new nodes will not be
// attached to the document.
func (s *Selection) Clone() *Selection {
ns := newEmptySelection(s.document)
ns.Nodes = cloneNodes(s.Nodes)
return ns
}
// Empty removes all children nodes from the set of matched elements.
// It returns the children nodes in a new Selection.
func (s *Selection) Empty() *Selection {
var nodes []*html.Node
for _, n := range s.Nodes {
for c := n.FirstChild; c != nil; c = n.FirstChild {
n.RemoveChild(c)
nodes = append(nodes, c)
}
}
return pushStack(s, nodes)
}
// Prepend prepends the elements specified by the selector to each element in
// the set of matched elements, following the same rules as Append.
func (s *Selection) Prepend(selector string) *Selection {
return s.PrependMatcher(compileMatcher(selector))
}
// PrependMatcher prepends the elements specified by the matcher to each
// element in the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) PrependMatcher(m Matcher) *Selection {
return s.PrependNodes(m.MatchAll(s.document.rootNode)...)
}
// PrependSelection prepends the elements in the selection to each element in
// the set of matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) PrependSelection(sel *Selection) *Selection {
return s.PrependNodes(sel.Nodes...)
}
// PrependHtml parses the html and prepends it to the set of matched elements.
func (s *Selection) PrependHtml(html string) *Selection {
return s.PrependNodes(parseHtml(html)...)
}
// PrependNodes prepends the specified nodes to each node in the set of
// matched elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) PrependNodes(ns ...*html.Node) *Selection {
return s.manipulateNodes(ns, true, func(sn *html.Node, n *html.Node) {
// sn.FirstChild may be nil, in which case this functions like
// sn.AppendChild()
sn.InsertBefore(n, sn.FirstChild)
})
}
// Remove removes the set of matched elements from the document.
// It returns the same selection, now consisting of nodes not in the document.
func (s *Selection) Remove() *Selection {
for _, n := range s.Nodes {
if n.Parent != nil {
n.Parent.RemoveChild(n)
}
}
return s
}
// RemoveFiltered removes the set of matched elements by selector.
// It returns the Selection of removed nodes.
func (s *Selection) RemoveFiltered(selector string) *Selection {
return s.RemoveMatcher(compileMatcher(selector))
}
// RemoveMatcher removes the set of matched elements.
// It returns the Selection of removed nodes.
func (s *Selection) RemoveMatcher(m Matcher) *Selection {
return s.FilterMatcher(m).Remove()
}
// ReplaceWith replaces each element in the set of matched elements with the
// nodes matched by the given selector.
// It returns the removed elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) ReplaceWith(selector string) *Selection {
return s.ReplaceWithMatcher(compileMatcher(selector))
}
// ReplaceWithMatcher replaces each element in the set of matched elements with
// the nodes matched by the given Matcher.
// It returns the removed elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) ReplaceWithMatcher(m Matcher) *Selection {
return s.ReplaceWithNodes(m.MatchAll(s.document.rootNode)...)
}
// ReplaceWithSelection replaces each element in the set of matched elements with
// the nodes from the given Selection.
// It returns the removed elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) ReplaceWithSelection(sel *Selection) *Selection {
return s.ReplaceWithNodes(sel.Nodes...)
}
// ReplaceWithHtml replaces each element in the set of matched elements with
// the parsed HTML.
// It returns the removed elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) ReplaceWithHtml(html string) *Selection {
return s.ReplaceWithNodes(parseHtml(html)...)
}
// ReplaceWithNodes replaces each element in the set of matched elements with
// the given nodes.
// It returns the removed elements.
//
// This follows the same rules as Selection.Append.
func (s *Selection) ReplaceWithNodes(ns ...*html.Node) *Selection {
s.AfterNodes(ns...)
return s.Remove()
}
// SetHtml sets the html content of each element in the selection to
// specified html string.
func (s *Selection) SetHtml(html string) *Selection {
return setHtmlNodes(s, parseHtml(html)...)
}
// SetText sets the content of each element in the selection to specified content.
// The provided text string is escaped.
func (s *Selection) SetText(text string) *Selection {
return s.SetHtml(html.EscapeString(text))
}
// Unwrap removes the parents of the set of matched elements, leaving the matched
// elements (and their siblings, if any) in their place.
// It returns the original selection.
func (s *Selection) Unwrap() *Selection {
s.Parent().Each(func(i int, ss *Selection) {
// For some reason, jquery allows unwrap to remove the <head> element, so
// allowing it here too. Same for <html>. Why it allows those elements to
// be unwrapped while not allowing body is a mystery to me.
if ss.Nodes[0].Data != "body" {
ss.ReplaceWithSelection(ss.Contents())
}
})
return s
}
// Wrap wraps each element in the set of matched elements inside the first
// element matched by the given selector. The matched child is cloned before
// being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) Wrap(selector string) *Selection {
return s.WrapMatcher(compileMatcher(selector))
}
// WrapMatcher wraps each element in the set of matched elements inside the
// first element matched by the given matcher. The matched child is cloned
// before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapMatcher(m Matcher) *Selection {
return s.wrapNodes(m.MatchAll(s.document.rootNode)...)
}
// WrapSelection wraps each element in the set of matched elements inside the
// first element in the given Selection. The element is cloned before being
// inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapSelection(sel *Selection) *Selection {
return s.wrapNodes(sel.Nodes...)
}
// WrapHtml wraps each element in the set of matched elements inside the inner-
// most child of the given HTML.
//
// It returns the original set of elements.
func (s *Selection) WrapHtml(html string) *Selection {
return s.wrapNodes(parseHtml(html)...)
}
// WrapNode wraps each element in the set of matched elements inside the inner-
// most child of the given node. The given node is copied before being inserted
// into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapNode(n *html.Node) *Selection {
return s.wrapNodes(n)
}
func (s *Selection) wrapNodes(ns ...*html.Node) *Selection {
s.Each(func(i int, ss *Selection) {
ss.wrapAllNodes(ns...)
})
return s
}
// WrapAll wraps a single HTML structure, matched by the given selector, around
// all elements in the set of matched elements. The matched child is cloned
// before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapAll(selector string) *Selection {
return s.WrapAllMatcher(compileMatcher(selector))
}
// WrapAllMatcher wraps a single HTML structure, matched by the given Matcher,
// around all elements in the set of matched elements. The matched child is
// cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapAllMatcher(m Matcher) *Selection {
return s.wrapAllNodes(m.MatchAll(s.document.rootNode)...)
}
// WrapAllSelection wraps a single HTML structure, the first node of the given
// Selection, around all elements in the set of matched elements. The matched
// child is cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapAllSelection(sel *Selection) *Selection {
return s.wrapAllNodes(sel.Nodes...)
}
// WrapAllHtml wraps the given HTML structure around all elements in the set of
// matched elements. The matched child is cloned before being inserted into the
// document.
//
// It returns the original set of elements.
func (s *Selection) WrapAllHtml(html string) *Selection {
return s.wrapAllNodes(parseHtml(html)...)
}
func (s *Selection) wrapAllNodes(ns ...*html.Node) *Selection {
if len(ns) > 0 {
return s.WrapAllNode(ns[0])
}
return s
}
// WrapAllNode wraps the given node around the first element in the Selection,
// making all other nodes in the Selection children of the given node. The node
// is cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapAllNode(n *html.Node) *Selection {
if s.Size() == 0 {
return s
}
wrap := cloneNode(n)
first := s.Nodes[0]
if first.Parent != nil {
first.Parent.InsertBefore(wrap, first)
first.Parent.RemoveChild(first)
}
for c := getFirstChildEl(wrap); c != nil; c = getFirstChildEl(wrap) {
wrap = c
}
newSingleSelection(wrap, s.document).AppendSelection(s)
return s
}
// WrapInner wraps an HTML structure, matched by the given selector, around the
// content of element in the set of matched elements. The matched child is
// cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapInner(selector string) *Selection {
return s.WrapInnerMatcher(compileMatcher(selector))
}
// WrapInnerMatcher wraps an HTML structure, matched by the given selector,
// around the content of element in the set of matched elements. The matched
// child is cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapInnerMatcher(m Matcher) *Selection {
return s.wrapInnerNodes(m.MatchAll(s.document.rootNode)...)
}
// WrapInnerSelection wraps an HTML structure, matched by the given selector,
// around the content of element in the set of matched elements. The matched
// child is cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapInnerSelection(sel *Selection) *Selection {
return s.wrapInnerNodes(sel.Nodes...)
}
// WrapInnerHtml wraps an HTML structure, matched by the given selector, around
// the content of element in the set of matched elements. The matched child is
// cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapInnerHtml(html string) *Selection {
return s.wrapInnerNodes(parseHtml(html)...)
}
// WrapInnerNode wraps an HTML structure, matched by the given selector, around
// the content of element in the set of matched elements. The matched child is
// cloned before being inserted into the document.
//
// It returns the original set of elements.
func (s *Selection) WrapInnerNode(n *html.Node) *Selection {
return s.wrapInnerNodes(n)
}
func (s *Selection) wrapInnerNodes(ns ...*html.Node) *Selection {
if len(ns) == 0 {
return s
}
s.Each(func(i int, s *Selection) {
contents := s.Contents()
if contents.Size() > 0 {
contents.wrapAllNodes(ns...)
} else {
s.AppendNodes(cloneNode(ns[0]))
}
})
return s
}
func parseHtml(h string) []*html.Node {
// Errors are only returned when the io.Reader returns any error besides
// EOF, but strings.Reader never will
nodes, err := html.ParseFragment(strings.NewReader(h), &html.Node{Type: html.ElementNode})
if err != nil {
panic("goquery: failed to parse HTML: " + err.Error())
}
return nodes
}
func setHtmlNodes(s *Selection, ns ...*html.Node) *Selection {
for _, n := range s.Nodes {
for c := n.FirstChild; c != nil; c = n.FirstChild {
n.RemoveChild(c)
}
for _, c := range ns {
n.AppendChild(cloneNode(c))
}
}
return s
}
// Get the first child that is an ElementNode
func getFirstChildEl(n *html.Node) *html.Node {
c := n.FirstChild
for c != nil && c.Type != html.ElementNode {
c = c.NextSibling
}
return c
}
// Deep copy a slice of nodes.
func cloneNodes(ns []*html.Node) []*html.Node {
cns := make([]*html.Node, 0, len(ns))
for _, n := range ns {
cns = append(cns, cloneNode(n))
}
return cns
}
// Deep copy a node. The new node has clones of all the original node's
// children but none of its parents or siblings.
func cloneNode(n *html.Node) *html.Node {
nn := &html.Node{
Type: n.Type,
DataAtom: n.DataAtom,
Data: n.Data,
Attr: make([]html.Attribute, len(n.Attr)),
}
copy(nn.Attr, n.Attr)
for c := n.FirstChild; c != nil; c = c.NextSibling {
nn.AppendChild(cloneNode(c))
}
return nn
}
func (s *Selection) manipulateNodes(ns []*html.Node, reverse bool,
f func(sn *html.Node, n *html.Node)) *Selection {
lasti := s.Size() - 1
// net.Html doesn't provide document fragments for insertion, so to get
// things in the correct order with After() and Prepend(), the callback
// needs to be called on the reverse of the nodes.
if reverse {
for i, j := 0, len(ns)-1; i < j; i, j = i+1, j-1 {
ns[i], ns[j] = ns[j], ns[i]
}
}
for i, sn := range s.Nodes {
for _, n := range ns {
if i != lasti {
f(sn, cloneNode(n))
} else {
if n.Parent != nil {
n.Parent.RemoveChild(n)
}
f(sn, n)
}
}
}
return s
}

View File

@@ -1,275 +0,0 @@
package goquery
import (
"bytes"
"regexp"
"strings"
"golang.org/x/net/html"
)
var rxClassTrim = regexp.MustCompile("[\t\r\n]")
// Attr gets the specified attribute's value for the first element in the
// Selection. To get the value for each element individually, use a looping
// construct such as Each or Map method.
func (s *Selection) Attr(attrName string) (val string, exists bool) {
if len(s.Nodes) == 0 {
return
}
return getAttributeValue(attrName, s.Nodes[0])
}
// AttrOr works like Attr but returns default value if attribute is not present.
func (s *Selection) AttrOr(attrName, defaultValue string) string {
if len(s.Nodes) == 0 {
return defaultValue
}
val, exists := getAttributeValue(attrName, s.Nodes[0])
if !exists {
return defaultValue
}
return val
}
// RemoveAttr removes the named attribute from each element in the set of matched elements.
func (s *Selection) RemoveAttr(attrName string) *Selection {
for _, n := range s.Nodes {
removeAttr(n, attrName)
}
return s
}
// SetAttr sets the given attribute on each element in the set of matched elements.
func (s *Selection) SetAttr(attrName, val string) *Selection {
for _, n := range s.Nodes {
attr := getAttributePtr(attrName, n)
if attr == nil {
n.Attr = append(n.Attr, html.Attribute{Key: attrName, Val: val})
} else {
attr.Val = val
}
}
return s
}
// Text gets the combined text contents of each element in the set of matched
// elements, including their descendants.
func (s *Selection) Text() string {
var buf bytes.Buffer
// Slightly optimized vs calling Each: no single selection object created
var f func(*html.Node)
f = func(n *html.Node) {
if n.Type == html.TextNode {
// Keep newlines and spaces, like jQuery
buf.WriteString(n.Data)
}
if n.FirstChild != nil {
for c := n.FirstChild; c != nil; c = c.NextSibling {
f(c)
}
}
}
for _, n := range s.Nodes {
f(n)
}
return buf.String()
}
// Size is an alias for Length.
func (s *Selection) Size() int {
return s.Length()
}
// Length returns the number of elements in the Selection object.
func (s *Selection) Length() int {
return len(s.Nodes)
}
// Html gets the HTML contents of the first element in the set of matched
// elements. It includes text and comment nodes.
func (s *Selection) Html() (ret string, e error) {
// Since there is no .innerHtml, the HTML content must be re-created from
// the nodes using html.Render.
var buf bytes.Buffer
if len(s.Nodes) > 0 {
for c := s.Nodes[0].FirstChild; c != nil; c = c.NextSibling {
e = html.Render(&buf, c)
if e != nil {
return
}
}
ret = buf.String()
}
return
}
// AddClass adds the given class(es) to each element in the set of matched elements.
// Multiple class names can be specified, separated by a space or via multiple arguments.
func (s *Selection) AddClass(class ...string) *Selection {
classStr := strings.TrimSpace(strings.Join(class, " "))
if classStr == "" {
return s
}
tcls := getClassesSlice(classStr)
for _, n := range s.Nodes {
curClasses, attr := getClassesAndAttr(n, true)
for _, newClass := range tcls {
if !strings.Contains(curClasses, " "+newClass+" ") {
curClasses += newClass + " "
}
}
setClasses(n, attr, curClasses)
}
return s
}
// HasClass determines whether any of the matched elements are assigned the
// given class.
func (s *Selection) HasClass(class string) bool {
class = " " + class + " "
for _, n := range s.Nodes {
classes, _ := getClassesAndAttr(n, false)
if strings.Contains(classes, class) {
return true
}
}
return false
}
// RemoveClass removes the given class(es) from each element in the set of matched elements.
// Multiple class names can be specified, separated by a space or via multiple arguments.
// If no class name is provided, all classes are removed.
func (s *Selection) RemoveClass(class ...string) *Selection {
var rclasses []string
classStr := strings.TrimSpace(strings.Join(class, " "))
remove := classStr == ""
if !remove {
rclasses = getClassesSlice(classStr)
}
for _, n := range s.Nodes {
if remove {
removeAttr(n, "class")
} else {
classes, attr := getClassesAndAttr(n, true)
for _, rcl := range rclasses {
classes = strings.Replace(classes, " "+rcl+" ", " ", -1)
}
setClasses(n, attr, classes)
}
}
return s
}
// ToggleClass adds or removes the given class(es) for each element in the set of matched elements.
// Multiple class names can be specified, separated by a space or via multiple arguments.
func (s *Selection) ToggleClass(class ...string) *Selection {
classStr := strings.TrimSpace(strings.Join(class, " "))
if classStr == "" {
return s
}
tcls := getClassesSlice(classStr)
for _, n := range s.Nodes {
classes, attr := getClassesAndAttr(n, true)
for _, tcl := range tcls {
if strings.Contains(classes, " "+tcl+" ") {
classes = strings.Replace(classes, " "+tcl+" ", " ", -1)
} else {
classes += tcl + " "
}
}
setClasses(n, attr, classes)
}
return s
}
func getAttributePtr(attrName string, n *html.Node) *html.Attribute {
if n == nil {
return nil
}
for i, a := range n.Attr {
if a.Key == attrName {
return &n.Attr[i]
}
}
return nil
}
// Private function to get the specified attribute's value from a node.
func getAttributeValue(attrName string, n *html.Node) (val string, exists bool) {
if a := getAttributePtr(attrName, n); a != nil {
val = a.Val
exists = true
}
return
}
// Get and normalize the "class" attribute from the node.
func getClassesAndAttr(n *html.Node, create bool) (classes string, attr *html.Attribute) {
// Applies only to element nodes
if n.Type == html.ElementNode {
attr = getAttributePtr("class", n)
if attr == nil && create {
n.Attr = append(n.Attr, html.Attribute{
Key: "class",
Val: "",
})
attr = &n.Attr[len(n.Attr)-1]
}
}
if attr == nil {
classes = " "
} else {
classes = rxClassTrim.ReplaceAllString(" "+attr.Val+" ", " ")
}
return
}
func getClassesSlice(classes string) []string {
return strings.Split(rxClassTrim.ReplaceAllString(" "+classes+" ", " "), " ")
}
func removeAttr(n *html.Node, attrName string) {
for i, a := range n.Attr {
if a.Key == attrName {
n.Attr[i], n.Attr[len(n.Attr)-1], n.Attr =
n.Attr[len(n.Attr)-1], html.Attribute{}, n.Attr[:len(n.Attr)-1]
return
}
}
}
func setClasses(n *html.Node, attr *html.Attribute, classes string) {
classes = strings.TrimSpace(classes)
if classes == "" {
removeAttr(n, "class")
return
}
attr.Val = classes
}

View File

@@ -1,49 +0,0 @@
package goquery
import "golang.org/x/net/html"
// Is checks the current matched set of elements against a selector and
// returns true if at least one of these elements matches.
func (s *Selection) Is(selector string) bool {
return s.IsMatcher(compileMatcher(selector))
}
// IsMatcher checks the current matched set of elements against a matcher and
// returns true if at least one of these elements matches.
func (s *Selection) IsMatcher(m Matcher) bool {
if len(s.Nodes) > 0 {
if len(s.Nodes) == 1 {
return m.Match(s.Nodes[0])
}
return len(m.Filter(s.Nodes)) > 0
}
return false
}
// IsFunction checks the current matched set of elements against a predicate and
// returns true if at least one of these elements matches.
func (s *Selection) IsFunction(f func(int, *Selection) bool) bool {
return s.FilterFunction(f).Length() > 0
}
// IsSelection checks the current matched set of elements against a Selection object
// and returns true if at least one of these elements matches.
func (s *Selection) IsSelection(sel *Selection) bool {
return s.FilterSelection(sel).Length() > 0
}
// IsNodes checks the current matched set of elements against the specified nodes
// and returns true if at least one of these elements matches.
func (s *Selection) IsNodes(nodes ...*html.Node) bool {
return s.FilterNodes(nodes...).Length() > 0
}
// Contains returns true if the specified Node is within,
// at any depth, one of the nodes in the Selection object.
// It is NOT inclusive, to behave like jQuery's implementation, and
// unlike Javascript's .contains, so if the contained
// node is itself in the selection, it returns false.
func (s *Selection) Contains(n *html.Node) bool {
return sliceContains(s.Nodes, n)
}

View File

@@ -1,698 +0,0 @@
package goquery
import "golang.org/x/net/html"
type siblingType int
// Sibling type, used internally when iterating over children at the same
// level (siblings) to specify which nodes are requested.
const (
siblingPrevUntil siblingType = iota - 3
siblingPrevAll
siblingPrev
siblingAll
siblingNext
siblingNextAll
siblingNextUntil
siblingAllIncludingNonElements
)
// Find gets the descendants of each element in the current set of matched
// elements, filtered by a selector. It returns a new Selection object
// containing these matched elements.
func (s *Selection) Find(selector string) *Selection {
return pushStack(s, findWithMatcher(s.Nodes, compileMatcher(selector)))
}
// FindMatcher gets the descendants of each element in the current set of matched
// elements, filtered by the matcher. It returns a new Selection object
// containing these matched elements.
func (s *Selection) FindMatcher(m Matcher) *Selection {
return pushStack(s, findWithMatcher(s.Nodes, m))
}
// FindSelection gets the descendants of each element in the current
// Selection, filtered by a Selection. It returns a new Selection object
// containing these matched elements.
func (s *Selection) FindSelection(sel *Selection) *Selection {
if sel == nil {
return pushStack(s, nil)
}
return s.FindNodes(sel.Nodes...)
}
// FindNodes gets the descendants of each element in the current
// Selection, filtered by some nodes. It returns a new Selection object
// containing these matched elements.
func (s *Selection) FindNodes(nodes ...*html.Node) *Selection {
return pushStack(s, mapNodes(nodes, func(i int, n *html.Node) []*html.Node {
if sliceContains(s.Nodes, n) {
return []*html.Node{n}
}
return nil
}))
}
// Contents gets the children of each element in the Selection,
// including text and comment nodes. It returns a new Selection object
// containing these elements.
func (s *Selection) Contents() *Selection {
return pushStack(s, getChildrenNodes(s.Nodes, siblingAllIncludingNonElements))
}
// ContentsFiltered gets the children of each element in the Selection,
// filtered by the specified selector. It returns a new Selection
// object containing these elements. Since selectors only act on Element nodes,
// this function is an alias to ChildrenFiltered unless the selector is empty,
// in which case it is an alias to Contents.
func (s *Selection) ContentsFiltered(selector string) *Selection {
if selector != "" {
return s.ChildrenFiltered(selector)
}
return s.Contents()
}
// ContentsMatcher gets the children of each element in the Selection,
// filtered by the specified matcher. It returns a new Selection
// object containing these elements. Since matchers only act on Element nodes,
// this function is an alias to ChildrenMatcher.
func (s *Selection) ContentsMatcher(m Matcher) *Selection {
return s.ChildrenMatcher(m)
}
// Children gets the child elements of each element in the Selection.
// It returns a new Selection object containing these elements.
func (s *Selection) Children() *Selection {
return pushStack(s, getChildrenNodes(s.Nodes, siblingAll))
}
// ChildrenFiltered gets the child elements of each element in the Selection,
// filtered by the specified selector. It returns a new
// Selection object containing these elements.
func (s *Selection) ChildrenFiltered(selector string) *Selection {
return filterAndPush(s, getChildrenNodes(s.Nodes, siblingAll), compileMatcher(selector))
}
// ChildrenMatcher gets the child elements of each element in the Selection,
// filtered by the specified matcher. It returns a new
// Selection object containing these elements.
func (s *Selection) ChildrenMatcher(m Matcher) *Selection {
return filterAndPush(s, getChildrenNodes(s.Nodes, siblingAll), m)
}
// Parent gets the parent of each element in the Selection. It returns a
// new Selection object containing the matched elements.
func (s *Selection) Parent() *Selection {
return pushStack(s, getParentNodes(s.Nodes))
}
// ParentFiltered gets the parent of each element in the Selection filtered by a
// selector. It returns a new Selection object containing the matched elements.
func (s *Selection) ParentFiltered(selector string) *Selection {
return filterAndPush(s, getParentNodes(s.Nodes), compileMatcher(selector))
}
// ParentMatcher gets the parent of each element in the Selection filtered by a
// matcher. It returns a new Selection object containing the matched elements.
func (s *Selection) ParentMatcher(m Matcher) *Selection {
return filterAndPush(s, getParentNodes(s.Nodes), m)
}
// Closest gets the first element that matches the selector by testing the
// element itself and traversing up through its ancestors in the DOM tree.
func (s *Selection) Closest(selector string) *Selection {
cs := compileMatcher(selector)
return s.ClosestMatcher(cs)
}
// ClosestMatcher gets the first element that matches the matcher by testing the
// element itself and traversing up through its ancestors in the DOM tree.
func (s *Selection) ClosestMatcher(m Matcher) *Selection {
return pushStack(s, mapNodes(s.Nodes, func(i int, n *html.Node) []*html.Node {
// For each node in the selection, test the node itself, then each parent
// until a match is found.
for ; n != nil; n = n.Parent {
if m.Match(n) {
return []*html.Node{n}
}
}
return nil
}))
}
// ClosestNodes gets the first element that matches one of the nodes by testing the
// element itself and traversing up through its ancestors in the DOM tree.
func (s *Selection) ClosestNodes(nodes ...*html.Node) *Selection {
set := make(map[*html.Node]bool)
for _, n := range nodes {
set[n] = true
}
return pushStack(s, mapNodes(s.Nodes, func(i int, n *html.Node) []*html.Node {
// For each node in the selection, test the node itself, then each parent
// until a match is found.
for ; n != nil; n = n.Parent {
if set[n] {
return []*html.Node{n}
}
}
return nil
}))
}
// ClosestSelection gets the first element that matches one of the nodes in the
// Selection by testing the element itself and traversing up through its ancestors
// in the DOM tree.
func (s *Selection) ClosestSelection(sel *Selection) *Selection {
if sel == nil {
return pushStack(s, nil)
}
return s.ClosestNodes(sel.Nodes...)
}
// Parents gets the ancestors of each element in the current Selection. It
// returns a new Selection object with the matched elements.
func (s *Selection) Parents() *Selection {
return pushStack(s, getParentsNodes(s.Nodes, nil, nil))
}
// ParentsFiltered gets the ancestors of each element in the current
// Selection. It returns a new Selection object with the matched elements.
func (s *Selection) ParentsFiltered(selector string) *Selection {
return filterAndPush(s, getParentsNodes(s.Nodes, nil, nil), compileMatcher(selector))
}
// ParentsMatcher gets the ancestors of each element in the current
// Selection. It returns a new Selection object with the matched elements.
func (s *Selection) ParentsMatcher(m Matcher) *Selection {
return filterAndPush(s, getParentsNodes(s.Nodes, nil, nil), m)
}
// ParentsUntil gets the ancestors of each element in the Selection, up to but
// not including the element matched by the selector. It returns a new Selection
// object containing the matched elements.
func (s *Selection) ParentsUntil(selector string) *Selection {
return pushStack(s, getParentsNodes(s.Nodes, compileMatcher(selector), nil))
}
// ParentsUntilMatcher gets the ancestors of each element in the Selection, up to but
// not including the element matched by the matcher. It returns a new Selection
// object containing the matched elements.
func (s *Selection) ParentsUntilMatcher(m Matcher) *Selection {
return pushStack(s, getParentsNodes(s.Nodes, m, nil))
}
// ParentsUntilSelection gets the ancestors of each element in the Selection,
// up to but not including the elements in the specified Selection. It returns a
// new Selection object containing the matched elements.
func (s *Selection) ParentsUntilSelection(sel *Selection) *Selection {
if sel == nil {
return s.Parents()
}
return s.ParentsUntilNodes(sel.Nodes...)
}
// ParentsUntilNodes gets the ancestors of each element in the Selection,
// up to but not including the specified nodes. It returns a
// new Selection object containing the matched elements.
func (s *Selection) ParentsUntilNodes(nodes ...*html.Node) *Selection {
return pushStack(s, getParentsNodes(s.Nodes, nil, nodes))
}
// ParentsFilteredUntil is like ParentsUntil, with the option to filter the
// results based on a selector string. It returns a new Selection
// object containing the matched elements.
func (s *Selection) ParentsFilteredUntil(filterSelector, untilSelector string) *Selection {
return filterAndPush(s, getParentsNodes(s.Nodes, compileMatcher(untilSelector), nil), compileMatcher(filterSelector))
}
// ParentsFilteredUntilMatcher is like ParentsUntilMatcher, with the option to filter the
// results based on a matcher. It returns a new Selection object containing the matched elements.
func (s *Selection) ParentsFilteredUntilMatcher(filter, until Matcher) *Selection {
return filterAndPush(s, getParentsNodes(s.Nodes, until, nil), filter)
}
// ParentsFilteredUntilSelection is like ParentsUntilSelection, with the
// option to filter the results based on a selector string. It returns a new
// Selection object containing the matched elements.
func (s *Selection) ParentsFilteredUntilSelection(filterSelector string, sel *Selection) *Selection {
return s.ParentsMatcherUntilSelection(compileMatcher(filterSelector), sel)
}
// ParentsMatcherUntilSelection is like ParentsUntilSelection, with the
// option to filter the results based on a matcher. It returns a new
// Selection object containing the matched elements.
func (s *Selection) ParentsMatcherUntilSelection(filter Matcher, sel *Selection) *Selection {
if sel == nil {
return s.ParentsMatcher(filter)
}
return s.ParentsMatcherUntilNodes(filter, sel.Nodes...)
}
// ParentsFilteredUntilNodes is like ParentsUntilNodes, with the
// option to filter the results based on a selector string. It returns a new
// Selection object containing the matched elements.
func (s *Selection) ParentsFilteredUntilNodes(filterSelector string, nodes ...*html.Node) *Selection {
return filterAndPush(s, getParentsNodes(s.Nodes, nil, nodes), compileMatcher(filterSelector))
}
// ParentsMatcherUntilNodes is like ParentsUntilNodes, with the
// option to filter the results based on a matcher. It returns a new
// Selection object containing the matched elements.
func (s *Selection) ParentsMatcherUntilNodes(filter Matcher, nodes ...*html.Node) *Selection {
return filterAndPush(s, getParentsNodes(s.Nodes, nil, nodes), filter)
}
// Siblings gets the siblings of each element in the Selection. It returns
// a new Selection object containing the matched elements.
func (s *Selection) Siblings() *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingAll, nil, nil))
}
// SiblingsFiltered gets the siblings of each element in the Selection
// filtered by a selector. It returns a new Selection object containing the
// matched elements.
func (s *Selection) SiblingsFiltered(selector string) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingAll, nil, nil), compileMatcher(selector))
}
// SiblingsMatcher gets the siblings of each element in the Selection
// filtered by a matcher. It returns a new Selection object containing the
// matched elements.
func (s *Selection) SiblingsMatcher(m Matcher) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingAll, nil, nil), m)
}
// Next gets the immediately following sibling of each element in the
// Selection. It returns a new Selection object containing the matched elements.
func (s *Selection) Next() *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingNext, nil, nil))
}
// NextFiltered gets the immediately following sibling of each element in the
// Selection filtered by a selector. It returns a new Selection object
// containing the matched elements.
func (s *Selection) NextFiltered(selector string) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNext, nil, nil), compileMatcher(selector))
}
// NextMatcher gets the immediately following sibling of each element in the
// Selection filtered by a matcher. It returns a new Selection object
// containing the matched elements.
func (s *Selection) NextMatcher(m Matcher) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNext, nil, nil), m)
}
// NextAll gets all the following siblings of each element in the
// Selection. It returns a new Selection object containing the matched elements.
func (s *Selection) NextAll() *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingNextAll, nil, nil))
}
// NextAllFiltered gets all the following siblings of each element in the
// Selection filtered by a selector. It returns a new Selection object
// containing the matched elements.
func (s *Selection) NextAllFiltered(selector string) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNextAll, nil, nil), compileMatcher(selector))
}
// NextAllMatcher gets all the following siblings of each element in the
// Selection filtered by a matcher. It returns a new Selection object
// containing the matched elements.
func (s *Selection) NextAllMatcher(m Matcher) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNextAll, nil, nil), m)
}
// Prev gets the immediately preceding sibling of each element in the
// Selection. It returns a new Selection object containing the matched elements.
func (s *Selection) Prev() *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingPrev, nil, nil))
}
// PrevFiltered gets the immediately preceding sibling of each element in the
// Selection filtered by a selector. It returns a new Selection object
// containing the matched elements.
func (s *Selection) PrevFiltered(selector string) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrev, nil, nil), compileMatcher(selector))
}
// PrevMatcher gets the immediately preceding sibling of each element in the
// Selection filtered by a matcher. It returns a new Selection object
// containing the matched elements.
func (s *Selection) PrevMatcher(m Matcher) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrev, nil, nil), m)
}
// PrevAll gets all the preceding siblings of each element in the
// Selection. It returns a new Selection object containing the matched elements.
func (s *Selection) PrevAll() *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingPrevAll, nil, nil))
}
// PrevAllFiltered gets all the preceding siblings of each element in the
// Selection filtered by a selector. It returns a new Selection object
// containing the matched elements.
func (s *Selection) PrevAllFiltered(selector string) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrevAll, nil, nil), compileMatcher(selector))
}
// PrevAllMatcher gets all the preceding siblings of each element in the
// Selection filtered by a matcher. It returns a new Selection object
// containing the matched elements.
func (s *Selection) PrevAllMatcher(m Matcher) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrevAll, nil, nil), m)
}
// NextUntil gets all following siblings of each element up to but not
// including the element matched by the selector. It returns a new Selection
// object containing the matched elements.
func (s *Selection) NextUntil(selector string) *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingNextUntil,
compileMatcher(selector), nil))
}
// NextUntilMatcher gets all following siblings of each element up to but not
// including the element matched by the matcher. It returns a new Selection
// object containing the matched elements.
func (s *Selection) NextUntilMatcher(m Matcher) *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingNextUntil,
m, nil))
}
// NextUntilSelection gets all following siblings of each element up to but not
// including the element matched by the Selection. It returns a new Selection
// object containing the matched elements.
func (s *Selection) NextUntilSelection(sel *Selection) *Selection {
if sel == nil {
return s.NextAll()
}
return s.NextUntilNodes(sel.Nodes...)
}
// NextUntilNodes gets all following siblings of each element up to but not
// including the element matched by the nodes. It returns a new Selection
// object containing the matched elements.
func (s *Selection) NextUntilNodes(nodes ...*html.Node) *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingNextUntil,
nil, nodes))
}
// PrevUntil gets all preceding siblings of each element up to but not
// including the element matched by the selector. It returns a new Selection
// object containing the matched elements.
func (s *Selection) PrevUntil(selector string) *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingPrevUntil,
compileMatcher(selector), nil))
}
// PrevUntilMatcher gets all preceding siblings of each element up to but not
// including the element matched by the matcher. It returns a new Selection
// object containing the matched elements.
func (s *Selection) PrevUntilMatcher(m Matcher) *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingPrevUntil,
m, nil))
}
// PrevUntilSelection gets all preceding siblings of each element up to but not
// including the element matched by the Selection. It returns a new Selection
// object containing the matched elements.
func (s *Selection) PrevUntilSelection(sel *Selection) *Selection {
if sel == nil {
return s.PrevAll()
}
return s.PrevUntilNodes(sel.Nodes...)
}
// PrevUntilNodes gets all preceding siblings of each element up to but not
// including the element matched by the nodes. It returns a new Selection
// object containing the matched elements.
func (s *Selection) PrevUntilNodes(nodes ...*html.Node) *Selection {
return pushStack(s, getSiblingNodes(s.Nodes, siblingPrevUntil,
nil, nodes))
}
// NextFilteredUntil is like NextUntil, with the option to filter
// the results based on a selector string.
// It returns a new Selection object containing the matched elements.
func (s *Selection) NextFilteredUntil(filterSelector, untilSelector string) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNextUntil,
compileMatcher(untilSelector), nil), compileMatcher(filterSelector))
}
// NextFilteredUntilMatcher is like NextUntilMatcher, with the option to filter
// the results based on a matcher.
// It returns a new Selection object containing the matched elements.
func (s *Selection) NextFilteredUntilMatcher(filter, until Matcher) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNextUntil,
until, nil), filter)
}
// NextFilteredUntilSelection is like NextUntilSelection, with the
// option to filter the results based on a selector string. It returns a new
// Selection object containing the matched elements.
func (s *Selection) NextFilteredUntilSelection(filterSelector string, sel *Selection) *Selection {
return s.NextMatcherUntilSelection(compileMatcher(filterSelector), sel)
}
// NextMatcherUntilSelection is like NextUntilSelection, with the
// option to filter the results based on a matcher. It returns a new
// Selection object containing the matched elements.
func (s *Selection) NextMatcherUntilSelection(filter Matcher, sel *Selection) *Selection {
if sel == nil {
return s.NextMatcher(filter)
}
return s.NextMatcherUntilNodes(filter, sel.Nodes...)
}
// NextFilteredUntilNodes is like NextUntilNodes, with the
// option to filter the results based on a selector string. It returns a new
// Selection object containing the matched elements.
func (s *Selection) NextFilteredUntilNodes(filterSelector string, nodes ...*html.Node) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNextUntil,
nil, nodes), compileMatcher(filterSelector))
}
// NextMatcherUntilNodes is like NextUntilNodes, with the
// option to filter the results based on a matcher. It returns a new
// Selection object containing the matched elements.
func (s *Selection) NextMatcherUntilNodes(filter Matcher, nodes ...*html.Node) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingNextUntil,
nil, nodes), filter)
}
// PrevFilteredUntil is like PrevUntil, with the option to filter
// the results based on a selector string.
// It returns a new Selection object containing the matched elements.
func (s *Selection) PrevFilteredUntil(filterSelector, untilSelector string) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrevUntil,
compileMatcher(untilSelector), nil), compileMatcher(filterSelector))
}
// PrevFilteredUntilMatcher is like PrevUntilMatcher, with the option to filter
// the results based on a matcher.
// It returns a new Selection object containing the matched elements.
func (s *Selection) PrevFilteredUntilMatcher(filter, until Matcher) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrevUntil,
until, nil), filter)
}
// PrevFilteredUntilSelection is like PrevUntilSelection, with the
// option to filter the results based on a selector string. It returns a new
// Selection object containing the matched elements.
func (s *Selection) PrevFilteredUntilSelection(filterSelector string, sel *Selection) *Selection {
return s.PrevMatcherUntilSelection(compileMatcher(filterSelector), sel)
}
// PrevMatcherUntilSelection is like PrevUntilSelection, with the
// option to filter the results based on a matcher. It returns a new
// Selection object containing the matched elements.
func (s *Selection) PrevMatcherUntilSelection(filter Matcher, sel *Selection) *Selection {
if sel == nil {
return s.PrevMatcher(filter)
}
return s.PrevMatcherUntilNodes(filter, sel.Nodes...)
}
// PrevFilteredUntilNodes is like PrevUntilNodes, with the
// option to filter the results based on a selector string. It returns a new
// Selection object containing the matched elements.
func (s *Selection) PrevFilteredUntilNodes(filterSelector string, nodes ...*html.Node) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrevUntil,
nil, nodes), compileMatcher(filterSelector))
}
// PrevMatcherUntilNodes is like PrevUntilNodes, with the
// option to filter the results based on a matcher. It returns a new
// Selection object containing the matched elements.
func (s *Selection) PrevMatcherUntilNodes(filter Matcher, nodes ...*html.Node) *Selection {
return filterAndPush(s, getSiblingNodes(s.Nodes, siblingPrevUntil,
nil, nodes), filter)
}
// Filter and push filters the nodes based on a matcher, and pushes the results
// on the stack, with the srcSel as previous selection.
func filterAndPush(srcSel *Selection, nodes []*html.Node, m Matcher) *Selection {
// Create a temporary Selection with the specified nodes to filter using winnow
sel := &Selection{nodes, srcSel.document, nil}
// Filter based on matcher and push on stack
return pushStack(srcSel, winnow(sel, m, true))
}
// Internal implementation of Find that return raw nodes.
func findWithMatcher(nodes []*html.Node, m Matcher) []*html.Node {
// Map nodes to find the matches within the children of each node
return mapNodes(nodes, func(i int, n *html.Node) (result []*html.Node) {
// Go down one level, becausejQuery's Find selects only within descendants
for c := n.FirstChild; c != nil; c = c.NextSibling {
if c.Type == html.ElementNode {
result = append(result, m.MatchAll(c)...)
}
}
return
})
}
// Internal implementation to get all parent nodes, stopping at the specified
// node (or nil if no stop).
func getParentsNodes(nodes []*html.Node, stopm Matcher, stopNodes []*html.Node) []*html.Node {
return mapNodes(nodes, func(i int, n *html.Node) (result []*html.Node) {
for p := n.Parent; p != nil; p = p.Parent {
sel := newSingleSelection(p, nil)
if stopm != nil {
if sel.IsMatcher(stopm) {
break
}
} else if len(stopNodes) > 0 {
if sel.IsNodes(stopNodes...) {
break
}
}
if p.Type == html.ElementNode {
result = append(result, p)
}
}
return
})
}
// Internal implementation of sibling nodes that return a raw slice of matches.
func getSiblingNodes(nodes []*html.Node, st siblingType, untilm Matcher, untilNodes []*html.Node) []*html.Node {
var f func(*html.Node) bool
// If the requested siblings are ...Until, create the test function to
// determine if the until condition is reached (returns true if it is)
if st == siblingNextUntil || st == siblingPrevUntil {
f = func(n *html.Node) bool {
if untilm != nil {
// Matcher-based condition
sel := newSingleSelection(n, nil)
return sel.IsMatcher(untilm)
} else if len(untilNodes) > 0 {
// Nodes-based condition
sel := newSingleSelection(n, nil)
return sel.IsNodes(untilNodes...)
}
return false
}
}
return mapNodes(nodes, func(i int, n *html.Node) []*html.Node {
return getChildrenWithSiblingType(n.Parent, st, n, f)
})
}
// Gets the children nodes of each node in the specified slice of nodes,
// based on the sibling type request.
func getChildrenNodes(nodes []*html.Node, st siblingType) []*html.Node {
return mapNodes(nodes, func(i int, n *html.Node) []*html.Node {
return getChildrenWithSiblingType(n, st, nil, nil)
})
}
// Gets the children of the specified parent, based on the requested sibling
// type, skipping a specified node if required.
func getChildrenWithSiblingType(parent *html.Node, st siblingType, skipNode *html.Node,
untilFunc func(*html.Node) bool) (result []*html.Node) {
// Create the iterator function
var iter = func(cur *html.Node) (ret *html.Node) {
// Based on the sibling type requested, iterate the right way
for {
switch st {
case siblingAll, siblingAllIncludingNonElements:
if cur == nil {
// First iteration, start with first child of parent
// Skip node if required
if ret = parent.FirstChild; ret == skipNode && skipNode != nil {
ret = skipNode.NextSibling
}
} else {
// Skip node if required
if ret = cur.NextSibling; ret == skipNode && skipNode != nil {
ret = skipNode.NextSibling
}
}
case siblingPrev, siblingPrevAll, siblingPrevUntil:
if cur == nil {
// Start with previous sibling of the skip node
ret = skipNode.PrevSibling
} else {
ret = cur.PrevSibling
}
case siblingNext, siblingNextAll, siblingNextUntil:
if cur == nil {
// Start with next sibling of the skip node
ret = skipNode.NextSibling
} else {
ret = cur.NextSibling
}
default:
panic("Invalid sibling type.")
}
if ret == nil || ret.Type == html.ElementNode || st == siblingAllIncludingNonElements {
return
}
// Not a valid node, try again from this one
cur = ret
}
}
for c := iter(nil); c != nil; c = iter(c) {
// If this is an ...Until case, test before append (returns true
// if the until condition is reached)
if st == siblingNextUntil || st == siblingPrevUntil {
if untilFunc(c) {
return
}
}
result = append(result, c)
if st == siblingNext || st == siblingPrev {
// Only one node was requested (immediate next or previous), so exit
return
}
}
return
}
// Internal implementation of parent nodes that return a raw slice of Nodes.
func getParentNodes(nodes []*html.Node) []*html.Node {
return mapNodes(nodes, func(i int, n *html.Node) []*html.Node {
if n.Parent != nil && n.Parent.Type == html.ElementNode {
return []*html.Node{n.Parent}
}
return nil
})
}
// Internal map function used by many traversing methods. Takes the source nodes
// to iterate on and the mapping function that returns an array of nodes.
// Returns an array of nodes mapped by calling the callback function once for
// each node in the source nodes.
func mapNodes(nodes []*html.Node, f func(int, *html.Node) []*html.Node) (result []*html.Node) {
set := make(map[*html.Node]bool)
for i, n := range nodes {
if vals := f(i, n); len(vals) > 0 {
result = appendWithoutDuplicates(result, vals, set)
}
}
return result
}

View File

@@ -1,141 +0,0 @@
package goquery
import (
"errors"
"io"
"net/http"
"net/url"
"github.com/andybalholm/cascadia"
"golang.org/x/net/html"
)
// Document represents an HTML document to be manipulated. Unlike jQuery, which
// is loaded as part of a DOM document, and thus acts upon its containing
// document, GoQuery doesn't know which HTML document to act upon. So it needs
// to be told, and that's what the Document class is for. It holds the root
// document node to manipulate, and can make selections on this document.
type Document struct {
*Selection
Url *url.URL
rootNode *html.Node
}
// NewDocumentFromNode is a Document constructor that takes a root html Node
// as argument.
func NewDocumentFromNode(root *html.Node) *Document {
return newDocument(root, nil)
}
// NewDocument is a Document constructor that takes a string URL as argument.
// It loads the specified document, parses it, and stores the root Document
// node, ready to be manipulated.
//
// Deprecated: Use the net/http standard library package to make the request
// and validate the response before calling goquery.NewDocumentFromReader
// with the response's body.
func NewDocument(url string) (*Document, error) {
// Load the URL
res, e := http.Get(url)
if e != nil {
return nil, e
}
return NewDocumentFromResponse(res)
}
// NewDocumentFromReader returns a Document from an io.Reader.
// It returns an error as second value if the reader's data cannot be parsed
// as html. It does not check if the reader is also an io.Closer, the
// provided reader is never closed by this call. It is the responsibility
// of the caller to close it if required.
func NewDocumentFromReader(r io.Reader) (*Document, error) {
root, e := html.Parse(r)
if e != nil {
return nil, e
}
return newDocument(root, nil), nil
}
// NewDocumentFromResponse is another Document constructor that takes an http response as argument.
// It loads the specified response's document, parses it, and stores the root Document
// node, ready to be manipulated. The response's body is closed on return.
//
// Deprecated: Use goquery.NewDocumentFromReader with the response's body.
func NewDocumentFromResponse(res *http.Response) (*Document, error) {
if res == nil {
return nil, errors.New("Response is nil")
}
defer res.Body.Close()
if res.Request == nil {
return nil, errors.New("Response.Request is nil")
}
// Parse the HTML into nodes
root, e := html.Parse(res.Body)
if e != nil {
return nil, e
}
// Create and fill the document
return newDocument(root, res.Request.URL), nil
}
// CloneDocument creates a deep-clone of a document.
func CloneDocument(doc *Document) *Document {
return newDocument(cloneNode(doc.rootNode), doc.Url)
}
// Private constructor, make sure all fields are correctly filled.
func newDocument(root *html.Node, url *url.URL) *Document {
// Create and fill the document
d := &Document{nil, url, root}
d.Selection = newSingleSelection(root, d)
return d
}
// Selection represents a collection of nodes matching some criteria. The
// initial Selection can be created by using Document.Find, and then
// manipulated using the jQuery-like chainable syntax and methods.
type Selection struct {
Nodes []*html.Node
document *Document
prevSel *Selection
}
// Helper constructor to create an empty selection
func newEmptySelection(doc *Document) *Selection {
return &Selection{nil, doc, nil}
}
// Helper constructor to create a selection of only one node
func newSingleSelection(node *html.Node, doc *Document) *Selection {
return &Selection{[]*html.Node{node}, doc, nil}
}
// Matcher is an interface that defines the methods to match
// HTML nodes against a compiled selector string. Cascadia's
// Selector implements this interface.
type Matcher interface {
Match(*html.Node) bool
MatchAll(*html.Node) []*html.Node
Filter([]*html.Node) []*html.Node
}
// compileMatcher compiles the selector string s and returns
// the corresponding Matcher. If s is an invalid selector string,
// it returns a Matcher that fails all matches.
func compileMatcher(s string) Matcher {
cs, err := cascadia.Compile(s)
if err != nil {
return invalidMatcher{}
}
return cs
}
// invalidMatcher is a Matcher that always fails to match.
type invalidMatcher struct{}
func (invalidMatcher) Match(n *html.Node) bool { return false }
func (invalidMatcher) MatchAll(n *html.Node) []*html.Node { return nil }
func (invalidMatcher) Filter(ns []*html.Node) []*html.Node { return nil }

View File

@@ -1,161 +0,0 @@
package goquery
import (
"bytes"
"golang.org/x/net/html"
)
// used to determine if a set (map[*html.Node]bool) should be used
// instead of iterating over a slice. The set uses more memory and
// is slower than slice iteration for small N.
const minNodesForSet = 1000
var nodeNames = []string{
html.ErrorNode: "#error",
html.TextNode: "#text",
html.DocumentNode: "#document",
html.CommentNode: "#comment",
}
// NodeName returns the node name of the first element in the selection.
// It tries to behave in a similar way as the DOM's nodeName property
// (https://developer.mozilla.org/en-US/docs/Web/API/Node/nodeName).
//
// Go's net/html package defines the following node types, listed with
// the corresponding returned value from this function:
//
// ErrorNode : #error
// TextNode : #text
// DocumentNode : #document
// ElementNode : the element's tag name
// CommentNode : #comment
// DoctypeNode : the name of the document type
//
func NodeName(s *Selection) string {
if s.Length() == 0 {
return ""
}
switch n := s.Get(0); n.Type {
case html.ElementNode, html.DoctypeNode:
return n.Data
default:
if n.Type >= 0 && int(n.Type) < len(nodeNames) {
return nodeNames[n.Type]
}
return ""
}
}
// OuterHtml returns the outer HTML rendering of the first item in
// the selection - that is, the HTML including the first element's
// tag and attributes.
//
// Unlike InnerHtml, this is a function and not a method on the Selection,
// because this is not a jQuery method (in javascript-land, this is
// a property provided by the DOM).
func OuterHtml(s *Selection) (string, error) {
var buf bytes.Buffer
if s.Length() == 0 {
return "", nil
}
n := s.Get(0)
if err := html.Render(&buf, n); err != nil {
return "", err
}
return buf.String(), nil
}
// Loop through all container nodes to search for the target node.
func sliceContains(container []*html.Node, contained *html.Node) bool {
for _, n := range container {
if nodeContains(n, contained) {
return true
}
}
return false
}
// Checks if the contained node is within the container node.
func nodeContains(container *html.Node, contained *html.Node) bool {
// Check if the parent of the contained node is the container node, traversing
// upward until the top is reached, or the container is found.
for contained = contained.Parent; contained != nil; contained = contained.Parent {
if container == contained {
return true
}
}
return false
}
// Checks if the target node is in the slice of nodes.
func isInSlice(slice []*html.Node, node *html.Node) bool {
return indexInSlice(slice, node) > -1
}
// Returns the index of the target node in the slice, or -1.
func indexInSlice(slice []*html.Node, node *html.Node) int {
if node != nil {
for i, n := range slice {
if n == node {
return i
}
}
}
return -1
}
// Appends the new nodes to the target slice, making sure no duplicate is added.
// There is no check to the original state of the target slice, so it may still
// contain duplicates. The target slice is returned because append() may create
// a new underlying array. If targetSet is nil, a local set is created with the
// target if len(target) + len(nodes) is greater than minNodesForSet.
func appendWithoutDuplicates(target []*html.Node, nodes []*html.Node, targetSet map[*html.Node]bool) []*html.Node {
// if there are not that many nodes, don't use the map, faster to just use nested loops
// (unless a non-nil targetSet is passed, in which case the caller knows better).
if targetSet == nil && len(target)+len(nodes) < minNodesForSet {
for _, n := range nodes {
if !isInSlice(target, n) {
target = append(target, n)
}
}
return target
}
// if a targetSet is passed, then assume it is reliable, otherwise create one
// and initialize it with the current target contents.
if targetSet == nil {
targetSet = make(map[*html.Node]bool, len(target))
for _, n := range target {
targetSet[n] = true
}
}
for _, n := range nodes {
if !targetSet[n] {
target = append(target, n)
targetSet[n] = true
}
}
return target
}
// Loop through a selection, returning only those nodes that pass the predicate
// function.
func grep(sel *Selection, predicate func(i int, s *Selection) bool) (result []*html.Node) {
for i, n := range sel.Nodes {
if predicate(i, newSingleSelection(n, sel.document)) {
result = append(result, n)
}
}
return result
}
// Creates a new Selection object based on the specified nodes, and keeps the
// source Selection object on the stack (linked list).
func pushStack(fromSel *Selection, nodes []*html.Node) *Selection {
result := &Selection{nodes, fromSel.document, fromSel}
return result
}

View File

@@ -1,14 +0,0 @@
language: go
go:
- 1.3
- 1.4
install:
- go get github.com/andybalholm/cascadia
script:
- go test -v
notifications:
email: false

View File

@@ -1,24 +0,0 @@
Copyright (c) 2011 Andy Balholm. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

View File

@@ -1,7 +0,0 @@
# cascadia
[![](https://travis-ci.org/andybalholm/cascadia.svg)](https://travis-ci.org/andybalholm/cascadia)
The Cascadia package implements CSS selectors for use with the parse trees produced by the html package.
To test CSS selectors without writing Go code, check out [cascadia](https://github.com/suntong/cascadia) the command line tool, a thin wrapper around this package.

View File

@@ -1,3 +0,0 @@
module "github.com/andybalholm/cascadia"
require "golang.org/x/net" v0.0.0-20180218175443-cbe0f9307d01

View File

@@ -1,835 +0,0 @@
// Package cascadia is an implementation of CSS selectors.
package cascadia
import (
"errors"
"fmt"
"regexp"
"strconv"
"strings"
"golang.org/x/net/html"
)
// a parser for CSS selectors
type parser struct {
s string // the source text
i int // the current position
}
// parseEscape parses a backslash escape.
func (p *parser) parseEscape() (result string, err error) {
if len(p.s) < p.i+2 || p.s[p.i] != '\\' {
return "", errors.New("invalid escape sequence")
}
start := p.i + 1
c := p.s[start]
switch {
case c == '\r' || c == '\n' || c == '\f':
return "", errors.New("escaped line ending outside string")
case hexDigit(c):
// unicode escape (hex)
var i int
for i = start; i < p.i+6 && i < len(p.s) && hexDigit(p.s[i]); i++ {
// empty
}
v, _ := strconv.ParseUint(p.s[start:i], 16, 21)
if len(p.s) > i {
switch p.s[i] {
case '\r':
i++
if len(p.s) > i && p.s[i] == '\n' {
i++
}
case ' ', '\t', '\n', '\f':
i++
}
}
p.i = i
return string(rune(v)), nil
}
// Return the literal character after the backslash.
result = p.s[start : start+1]
p.i += 2
return result, nil
}
func hexDigit(c byte) bool {
return '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F'
}
// nameStart returns whether c can be the first character of an identifier
// (not counting an initial hyphen, or an escape sequence).
func nameStart(c byte) bool {
return 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' || c == '_' || c > 127
}
// nameChar returns whether c can be a character within an identifier
// (not counting an escape sequence).
func nameChar(c byte) bool {
return 'a' <= c && c <= 'z' || 'A' <= c && c <= 'Z' || c == '_' || c > 127 ||
c == '-' || '0' <= c && c <= '9'
}
// parseIdentifier parses an identifier.
func (p *parser) parseIdentifier() (result string, err error) {
startingDash := false
if len(p.s) > p.i && p.s[p.i] == '-' {
startingDash = true
p.i++
}
if len(p.s) <= p.i {
return "", errors.New("expected identifier, found EOF instead")
}
if c := p.s[p.i]; !(nameStart(c) || c == '\\') {
return "", fmt.Errorf("expected identifier, found %c instead", c)
}
result, err = p.parseName()
if startingDash && err == nil {
result = "-" + result
}
return
}
// parseName parses a name (which is like an identifier, but doesn't have
// extra restrictions on the first character).
func (p *parser) parseName() (result string, err error) {
i := p.i
loop:
for i < len(p.s) {
c := p.s[i]
switch {
case nameChar(c):
start := i
for i < len(p.s) && nameChar(p.s[i]) {
i++
}
result += p.s[start:i]
case c == '\\':
p.i = i
val, err := p.parseEscape()
if err != nil {
return "", err
}
i = p.i
result += val
default:
break loop
}
}
if result == "" {
return "", errors.New("expected name, found EOF instead")
}
p.i = i
return result, nil
}
// parseString parses a single- or double-quoted string.
func (p *parser) parseString() (result string, err error) {
i := p.i
if len(p.s) < i+2 {
return "", errors.New("expected string, found EOF instead")
}
quote := p.s[i]
i++
loop:
for i < len(p.s) {
switch p.s[i] {
case '\\':
if len(p.s) > i+1 {
switch c := p.s[i+1]; c {
case '\r':
if len(p.s) > i+2 && p.s[i+2] == '\n' {
i += 3
continue loop
}
fallthrough
case '\n', '\f':
i += 2
continue loop
}
}
p.i = i
val, err := p.parseEscape()
if err != nil {
return "", err
}
i = p.i
result += val
case quote:
break loop
case '\r', '\n', '\f':
return "", errors.New("unexpected end of line in string")
default:
start := i
for i < len(p.s) {
if c := p.s[i]; c == quote || c == '\\' || c == '\r' || c == '\n' || c == '\f' {
break
}
i++
}
result += p.s[start:i]
}
}
if i >= len(p.s) {
return "", errors.New("EOF in string")
}
// Consume the final quote.
i++
p.i = i
return result, nil
}
// parseRegex parses a regular expression; the end is defined by encountering an
// unmatched closing ')' or ']' which is not consumed
func (p *parser) parseRegex() (rx *regexp.Regexp, err error) {
i := p.i
if len(p.s) < i+2 {
return nil, errors.New("expected regular expression, found EOF instead")
}
// number of open parens or brackets;
// when it becomes negative, finished parsing regex
open := 0
loop:
for i < len(p.s) {
switch p.s[i] {
case '(', '[':
open++
case ')', ']':
open--
if open < 0 {
break loop
}
}
i++
}
if i >= len(p.s) {
return nil, errors.New("EOF in regular expression")
}
rx, err = regexp.Compile(p.s[p.i:i])
p.i = i
return rx, err
}
// skipWhitespace consumes whitespace characters and comments.
// It returns true if there was actually anything to skip.
func (p *parser) skipWhitespace() bool {
i := p.i
for i < len(p.s) {
switch p.s[i] {
case ' ', '\t', '\r', '\n', '\f':
i++
continue
case '/':
if strings.HasPrefix(p.s[i:], "/*") {
end := strings.Index(p.s[i+len("/*"):], "*/")
if end != -1 {
i += end + len("/**/")
continue
}
}
}
break
}
if i > p.i {
p.i = i
return true
}
return false
}
// consumeParenthesis consumes an opening parenthesis and any following
// whitespace. It returns true if there was actually a parenthesis to skip.
func (p *parser) consumeParenthesis() bool {
if p.i < len(p.s) && p.s[p.i] == '(' {
p.i++
p.skipWhitespace()
return true
}
return false
}
// consumeClosingParenthesis consumes a closing parenthesis and any preceding
// whitespace. It returns true if there was actually a parenthesis to skip.
func (p *parser) consumeClosingParenthesis() bool {
i := p.i
p.skipWhitespace()
if p.i < len(p.s) && p.s[p.i] == ')' {
p.i++
return true
}
p.i = i
return false
}
// parseTypeSelector parses a type selector (one that matches by tag name).
func (p *parser) parseTypeSelector() (result Selector, err error) {
tag, err := p.parseIdentifier()
if err != nil {
return nil, err
}
return typeSelector(tag), nil
}
// parseIDSelector parses a selector that matches by id attribute.
func (p *parser) parseIDSelector() (Selector, error) {
if p.i >= len(p.s) {
return nil, fmt.Errorf("expected id selector (#id), found EOF instead")
}
if p.s[p.i] != '#' {
return nil, fmt.Errorf("expected id selector (#id), found '%c' instead", p.s[p.i])
}
p.i++
id, err := p.parseName()
if err != nil {
return nil, err
}
return attributeEqualsSelector("id", id), nil
}
// parseClassSelector parses a selector that matches by class attribute.
func (p *parser) parseClassSelector() (Selector, error) {
if p.i >= len(p.s) {
return nil, fmt.Errorf("expected class selector (.class), found EOF instead")
}
if p.s[p.i] != '.' {
return nil, fmt.Errorf("expected class selector (.class), found '%c' instead", p.s[p.i])
}
p.i++
class, err := p.parseIdentifier()
if err != nil {
return nil, err
}
return attributeIncludesSelector("class", class), nil
}
// parseAttributeSelector parses a selector that matches by attribute value.
func (p *parser) parseAttributeSelector() (Selector, error) {
if p.i >= len(p.s) {
return nil, fmt.Errorf("expected attribute selector ([attribute]), found EOF instead")
}
if p.s[p.i] != '[' {
return nil, fmt.Errorf("expected attribute selector ([attribute]), found '%c' instead", p.s[p.i])
}
p.i++
p.skipWhitespace()
key, err := p.parseIdentifier()
if err != nil {
return nil, err
}
p.skipWhitespace()
if p.i >= len(p.s) {
return nil, errors.New("unexpected EOF in attribute selector")
}
if p.s[p.i] == ']' {
p.i++
return attributeExistsSelector(key), nil
}
if p.i+2 >= len(p.s) {
return nil, errors.New("unexpected EOF in attribute selector")
}
op := p.s[p.i : p.i+2]
if op[0] == '=' {
op = "="
} else if op[1] != '=' {
return nil, fmt.Errorf(`expected equality operator, found "%s" instead`, op)
}
p.i += len(op)
p.skipWhitespace()
if p.i >= len(p.s) {
return nil, errors.New("unexpected EOF in attribute selector")
}
var val string
var rx *regexp.Regexp
if op == "#=" {
rx, err = p.parseRegex()
} else {
switch p.s[p.i] {
case '\'', '"':
val, err = p.parseString()
default:
val, err = p.parseIdentifier()
}
}
if err != nil {
return nil, err
}
p.skipWhitespace()
if p.i >= len(p.s) {
return nil, errors.New("unexpected EOF in attribute selector")
}
if p.s[p.i] != ']' {
return nil, fmt.Errorf("expected ']', found '%c' instead", p.s[p.i])
}
p.i++
switch op {
case "=":
return attributeEqualsSelector(key, val), nil
case "!=":
return attributeNotEqualSelector(key, val), nil
case "~=":
return attributeIncludesSelector(key, val), nil
case "|=":
return attributeDashmatchSelector(key, val), nil
case "^=":
return attributePrefixSelector(key, val), nil
case "$=":
return attributeSuffixSelector(key, val), nil
case "*=":
return attributeSubstringSelector(key, val), nil
case "#=":
return attributeRegexSelector(key, rx), nil
}
return nil, fmt.Errorf("attribute operator %q is not supported", op)
}
var errExpectedParenthesis = errors.New("expected '(' but didn't find it")
var errExpectedClosingParenthesis = errors.New("expected ')' but didn't find it")
var errUnmatchedParenthesis = errors.New("unmatched '('")
// parsePseudoclassSelector parses a pseudoclass selector like :not(p).
func (p *parser) parsePseudoclassSelector() (Selector, error) {
if p.i >= len(p.s) {
return nil, fmt.Errorf("expected pseudoclass selector (:pseudoclass), found EOF instead")
}
if p.s[p.i] != ':' {
return nil, fmt.Errorf("expected attribute selector (:pseudoclass), found '%c' instead", p.s[p.i])
}
p.i++
name, err := p.parseIdentifier()
if err != nil {
return nil, err
}
name = toLowerASCII(name)
switch name {
case "not", "has", "haschild":
if !p.consumeParenthesis() {
return nil, errExpectedParenthesis
}
sel, parseErr := p.parseSelectorGroup()
if parseErr != nil {
return nil, parseErr
}
if !p.consumeClosingParenthesis() {
return nil, errExpectedClosingParenthesis
}
switch name {
case "not":
return negatedSelector(sel), nil
case "has":
return hasDescendantSelector(sel), nil
case "haschild":
return hasChildSelector(sel), nil
}
case "contains", "containsown":
if !p.consumeParenthesis() {
return nil, errExpectedParenthesis
}
if p.i == len(p.s) {
return nil, errUnmatchedParenthesis
}
var val string
switch p.s[p.i] {
case '\'', '"':
val, err = p.parseString()
default:
val, err = p.parseIdentifier()
}
if err != nil {
return nil, err
}
val = strings.ToLower(val)
p.skipWhitespace()
if p.i >= len(p.s) {
return nil, errors.New("unexpected EOF in pseudo selector")
}
if !p.consumeClosingParenthesis() {
return nil, errExpectedClosingParenthesis
}
switch name {
case "contains":
return textSubstrSelector(val), nil
case "containsown":
return ownTextSubstrSelector(val), nil
}
case "matches", "matchesown":
if !p.consumeParenthesis() {
return nil, errExpectedParenthesis
}
rx, err := p.parseRegex()
if err != nil {
return nil, err
}
if p.i >= len(p.s) {
return nil, errors.New("unexpected EOF in pseudo selector")
}
if !p.consumeClosingParenthesis() {
return nil, errExpectedClosingParenthesis
}
switch name {
case "matches":
return textRegexSelector(rx), nil
case "matchesown":
return ownTextRegexSelector(rx), nil
}
case "nth-child", "nth-last-child", "nth-of-type", "nth-last-of-type":
if !p.consumeParenthesis() {
return nil, errExpectedParenthesis
}
a, b, err := p.parseNth()
if err != nil {
return nil, err
}
if !p.consumeClosingParenthesis() {
return nil, errExpectedClosingParenthesis
}
if a == 0 {
switch name {
case "nth-child":
return simpleNthChildSelector(b, false), nil
case "nth-of-type":
return simpleNthChildSelector(b, true), nil
case "nth-last-child":
return simpleNthLastChildSelector(b, false), nil
case "nth-last-of-type":
return simpleNthLastChildSelector(b, true), nil
}
}
return nthChildSelector(a, b,
name == "nth-last-child" || name == "nth-last-of-type",
name == "nth-of-type" || name == "nth-last-of-type"),
nil
case "first-child":
return simpleNthChildSelector(1, false), nil
case "last-child":
return simpleNthLastChildSelector(1, false), nil
case "first-of-type":
return simpleNthChildSelector(1, true), nil
case "last-of-type":
return simpleNthLastChildSelector(1, true), nil
case "only-child":
return onlyChildSelector(false), nil
case "only-of-type":
return onlyChildSelector(true), nil
case "input":
return inputSelector, nil
case "empty":
return emptyElementSelector, nil
case "root":
return rootSelector, nil
}
return nil, fmt.Errorf("unknown pseudoclass :%s", name)
}
// parseInteger parses a decimal integer.
func (p *parser) parseInteger() (int, error) {
i := p.i
start := i
for i < len(p.s) && '0' <= p.s[i] && p.s[i] <= '9' {
i++
}
if i == start {
return 0, errors.New("expected integer, but didn't find it")
}
p.i = i
val, err := strconv.Atoi(p.s[start:i])
if err != nil {
return 0, err
}
return val, nil
}
// parseNth parses the argument for :nth-child (normally of the form an+b).
func (p *parser) parseNth() (a, b int, err error) {
// initial state
if p.i >= len(p.s) {
goto eof
}
switch p.s[p.i] {
case '-':
p.i++
goto negativeA
case '+':
p.i++
goto positiveA
case '0', '1', '2', '3', '4', '5', '6', '7', '8', '9':
goto positiveA
case 'n', 'N':
a = 1
p.i++
goto readN
case 'o', 'O', 'e', 'E':
id, nameErr := p.parseName()
if nameErr != nil {
return 0, 0, nameErr
}
id = toLowerASCII(id)
if id == "odd" {
return 2, 1, nil
}
if id == "even" {
return 2, 0, nil
}
return 0, 0, fmt.Errorf("expected 'odd' or 'even', but found '%s' instead", id)
default:
goto invalid
}
positiveA:
if p.i >= len(p.s) {
goto eof
}
switch p.s[p.i] {
case '0', '1', '2', '3', '4', '5', '6', '7', '8', '9':
a, err = p.parseInteger()
if err != nil {
return 0, 0, err
}
goto readA
case 'n', 'N':
a = 1
p.i++
goto readN
default:
goto invalid
}
negativeA:
if p.i >= len(p.s) {
goto eof
}
switch p.s[p.i] {
case '0', '1', '2', '3', '4', '5', '6', '7', '8', '9':
a, err = p.parseInteger()
if err != nil {
return 0, 0, err
}
a = -a
goto readA
case 'n', 'N':
a = -1
p.i++
goto readN
default:
goto invalid
}
readA:
if p.i >= len(p.s) {
goto eof
}
switch p.s[p.i] {
case 'n', 'N':
p.i++
goto readN
default:
// The number we read as a is actually b.
return 0, a, nil
}
readN:
p.skipWhitespace()
if p.i >= len(p.s) {
goto eof
}
switch p.s[p.i] {
case '+':
p.i++
p.skipWhitespace()
b, err = p.parseInteger()
if err != nil {
return 0, 0, err
}
return a, b, nil
case '-':
p.i++
p.skipWhitespace()
b, err = p.parseInteger()
if err != nil {
return 0, 0, err
}
return a, -b, nil
default:
return a, 0, nil
}
eof:
return 0, 0, errors.New("unexpected EOF while attempting to parse expression of form an+b")
invalid:
return 0, 0, errors.New("unexpected character while attempting to parse expression of form an+b")
}
// parseSimpleSelectorSequence parses a selector sequence that applies to
// a single element.
func (p *parser) parseSimpleSelectorSequence() (Selector, error) {
var result Selector
if p.i >= len(p.s) {
return nil, errors.New("expected selector, found EOF instead")
}
switch p.s[p.i] {
case '*':
// It's the universal selector. Just skip over it, since it doesn't affect the meaning.
p.i++
case '#', '.', '[', ':':
// There's no type selector. Wait to process the other till the main loop.
default:
r, err := p.parseTypeSelector()
if err != nil {
return nil, err
}
result = r
}
loop:
for p.i < len(p.s) {
var ns Selector
var err error
switch p.s[p.i] {
case '#':
ns, err = p.parseIDSelector()
case '.':
ns, err = p.parseClassSelector()
case '[':
ns, err = p.parseAttributeSelector()
case ':':
ns, err = p.parsePseudoclassSelector()
default:
break loop
}
if err != nil {
return nil, err
}
if result == nil {
result = ns
} else {
result = intersectionSelector(result, ns)
}
}
if result == nil {
result = func(n *html.Node) bool {
return n.Type == html.ElementNode
}
}
return result, nil
}
// parseSelector parses a selector that may include combinators.
func (p *parser) parseSelector() (result Selector, err error) {
p.skipWhitespace()
result, err = p.parseSimpleSelectorSequence()
if err != nil {
return
}
for {
var combinator byte
if p.skipWhitespace() {
combinator = ' '
}
if p.i >= len(p.s) {
return
}
switch p.s[p.i] {
case '+', '>', '~':
combinator = p.s[p.i]
p.i++
p.skipWhitespace()
case ',', ')':
// These characters can't begin a selector, but they can legally occur after one.
return
}
if combinator == 0 {
return
}
c, err := p.parseSimpleSelectorSequence()
if err != nil {
return nil, err
}
switch combinator {
case ' ':
result = descendantSelector(result, c)
case '>':
result = childSelector(result, c)
case '+':
result = siblingSelector(result, c, true)
case '~':
result = siblingSelector(result, c, false)
}
}
panic("unreachable")
}
// parseSelectorGroup parses a group of selectors, separated by commas.
func (p *parser) parseSelectorGroup() (result Selector, err error) {
result, err = p.parseSelector()
if err != nil {
return
}
for p.i < len(p.s) {
if p.s[p.i] != ',' {
return result, nil
}
p.i++
c, err := p.parseSelector()
if err != nil {
return nil, err
}
result = unionSelector(result, c)
}
return
}

View File

@@ -1,622 +0,0 @@
package cascadia
import (
"bytes"
"fmt"
"regexp"
"strings"
"golang.org/x/net/html"
)
// the Selector type, and functions for creating them
// A Selector is a function which tells whether a node matches or not.
type Selector func(*html.Node) bool
// hasChildMatch returns whether n has any child that matches a.
func hasChildMatch(n *html.Node, a Selector) bool {
for c := n.FirstChild; c != nil; c = c.NextSibling {
if a(c) {
return true
}
}
return false
}
// hasDescendantMatch performs a depth-first search of n's descendants,
// testing whether any of them match a. It returns true as soon as a match is
// found, or false if no match is found.
func hasDescendantMatch(n *html.Node, a Selector) bool {
for c := n.FirstChild; c != nil; c = c.NextSibling {
if a(c) || (c.Type == html.ElementNode && hasDescendantMatch(c, a)) {
return true
}
}
return false
}
// Compile parses a selector and returns, if successful, a Selector object
// that can be used to match against html.Node objects.
func Compile(sel string) (Selector, error) {
p := &parser{s: sel}
compiled, err := p.parseSelectorGroup()
if err != nil {
return nil, err
}
if p.i < len(sel) {
return nil, fmt.Errorf("parsing %q: %d bytes left over", sel, len(sel)-p.i)
}
return compiled, nil
}
// MustCompile is like Compile, but panics instead of returning an error.
func MustCompile(sel string) Selector {
compiled, err := Compile(sel)
if err != nil {
panic(err)
}
return compiled
}
// MatchAll returns a slice of the nodes that match the selector,
// from n and its children.
func (s Selector) MatchAll(n *html.Node) []*html.Node {
return s.matchAllInto(n, nil)
}
func (s Selector) matchAllInto(n *html.Node, storage []*html.Node) []*html.Node {
if s(n) {
storage = append(storage, n)
}
for child := n.FirstChild; child != nil; child = child.NextSibling {
storage = s.matchAllInto(child, storage)
}
return storage
}
// Match returns true if the node matches the selector.
func (s Selector) Match(n *html.Node) bool {
return s(n)
}
// MatchFirst returns the first node that matches s, from n and its children.
func (s Selector) MatchFirst(n *html.Node) *html.Node {
if s.Match(n) {
return n
}
for c := n.FirstChild; c != nil; c = c.NextSibling {
m := s.MatchFirst(c)
if m != nil {
return m
}
}
return nil
}
// Filter returns the nodes in nodes that match the selector.
func (s Selector) Filter(nodes []*html.Node) (result []*html.Node) {
for _, n := range nodes {
if s(n) {
result = append(result, n)
}
}
return result
}
// typeSelector returns a Selector that matches elements with a given tag name.
func typeSelector(tag string) Selector {
tag = toLowerASCII(tag)
return func(n *html.Node) bool {
return n.Type == html.ElementNode && n.Data == tag
}
}
// toLowerASCII returns s with all ASCII capital letters lowercased.
func toLowerASCII(s string) string {
var b []byte
for i := 0; i < len(s); i++ {
if c := s[i]; 'A' <= c && c <= 'Z' {
if b == nil {
b = make([]byte, len(s))
copy(b, s)
}
b[i] = s[i] + ('a' - 'A')
}
}
if b == nil {
return s
}
return string(b)
}
// attributeSelector returns a Selector that matches elements
// where the attribute named key satisifes the function f.
func attributeSelector(key string, f func(string) bool) Selector {
key = toLowerASCII(key)
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
for _, a := range n.Attr {
if a.Key == key && f(a.Val) {
return true
}
}
return false
}
}
// attributeExistsSelector returns a Selector that matches elements that have
// an attribute named key.
func attributeExistsSelector(key string) Selector {
return attributeSelector(key, func(string) bool { return true })
}
// attributeEqualsSelector returns a Selector that matches elements where
// the attribute named key has the value val.
func attributeEqualsSelector(key, val string) Selector {
return attributeSelector(key,
func(s string) bool {
return s == val
})
}
// attributeNotEqualSelector returns a Selector that matches elements where
// the attribute named key does not have the value val.
func attributeNotEqualSelector(key, val string) Selector {
key = toLowerASCII(key)
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
for _, a := range n.Attr {
if a.Key == key && a.Val == val {
return false
}
}
return true
}
}
// attributeIncludesSelector returns a Selector that matches elements where
// the attribute named key is a whitespace-separated list that includes val.
func attributeIncludesSelector(key, val string) Selector {
return attributeSelector(key,
func(s string) bool {
for s != "" {
i := strings.IndexAny(s, " \t\r\n\f")
if i == -1 {
return s == val
}
if s[:i] == val {
return true
}
s = s[i+1:]
}
return false
})
}
// attributeDashmatchSelector returns a Selector that matches elements where
// the attribute named key equals val or starts with val plus a hyphen.
func attributeDashmatchSelector(key, val string) Selector {
return attributeSelector(key,
func(s string) bool {
if s == val {
return true
}
if len(s) <= len(val) {
return false
}
if s[:len(val)] == val && s[len(val)] == '-' {
return true
}
return false
})
}
// attributePrefixSelector returns a Selector that matches elements where
// the attribute named key starts with val.
func attributePrefixSelector(key, val string) Selector {
return attributeSelector(key,
func(s string) bool {
if strings.TrimSpace(s) == "" {
return false
}
return strings.HasPrefix(s, val)
})
}
// attributeSuffixSelector returns a Selector that matches elements where
// the attribute named key ends with val.
func attributeSuffixSelector(key, val string) Selector {
return attributeSelector(key,
func(s string) bool {
if strings.TrimSpace(s) == "" {
return false
}
return strings.HasSuffix(s, val)
})
}
// attributeSubstringSelector returns a Selector that matches nodes where
// the attribute named key contains val.
func attributeSubstringSelector(key, val string) Selector {
return attributeSelector(key,
func(s string) bool {
if strings.TrimSpace(s) == "" {
return false
}
return strings.Contains(s, val)
})
}
// attributeRegexSelector returns a Selector that matches nodes where
// the attribute named key matches the regular expression rx
func attributeRegexSelector(key string, rx *regexp.Regexp) Selector {
return attributeSelector(key,
func(s string) bool {
return rx.MatchString(s)
})
}
// intersectionSelector returns a selector that matches nodes that match
// both a and b.
func intersectionSelector(a, b Selector) Selector {
return func(n *html.Node) bool {
return a(n) && b(n)
}
}
// unionSelector returns a selector that matches elements that match
// either a or b.
func unionSelector(a, b Selector) Selector {
return func(n *html.Node) bool {
return a(n) || b(n)
}
}
// negatedSelector returns a selector that matches elements that do not match a.
func negatedSelector(a Selector) Selector {
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
return !a(n)
}
}
// writeNodeText writes the text contained in n and its descendants to b.
func writeNodeText(n *html.Node, b *bytes.Buffer) {
switch n.Type {
case html.TextNode:
b.WriteString(n.Data)
case html.ElementNode:
for c := n.FirstChild; c != nil; c = c.NextSibling {
writeNodeText(c, b)
}
}
}
// nodeText returns the text contained in n and its descendants.
func nodeText(n *html.Node) string {
var b bytes.Buffer
writeNodeText(n, &b)
return b.String()
}
// nodeOwnText returns the contents of the text nodes that are direct
// children of n.
func nodeOwnText(n *html.Node) string {
var b bytes.Buffer
for c := n.FirstChild; c != nil; c = c.NextSibling {
if c.Type == html.TextNode {
b.WriteString(c.Data)
}
}
return b.String()
}
// textSubstrSelector returns a selector that matches nodes that
// contain the given text.
func textSubstrSelector(val string) Selector {
return func(n *html.Node) bool {
text := strings.ToLower(nodeText(n))
return strings.Contains(text, val)
}
}
// ownTextSubstrSelector returns a selector that matches nodes that
// directly contain the given text
func ownTextSubstrSelector(val string) Selector {
return func(n *html.Node) bool {
text := strings.ToLower(nodeOwnText(n))
return strings.Contains(text, val)
}
}
// textRegexSelector returns a selector that matches nodes whose text matches
// the specified regular expression
func textRegexSelector(rx *regexp.Regexp) Selector {
return func(n *html.Node) bool {
return rx.MatchString(nodeText(n))
}
}
// ownTextRegexSelector returns a selector that matches nodes whose text
// directly matches the specified regular expression
func ownTextRegexSelector(rx *regexp.Regexp) Selector {
return func(n *html.Node) bool {
return rx.MatchString(nodeOwnText(n))
}
}
// hasChildSelector returns a selector that matches elements
// with a child that matches a.
func hasChildSelector(a Selector) Selector {
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
return hasChildMatch(n, a)
}
}
// hasDescendantSelector returns a selector that matches elements
// with any descendant that matches a.
func hasDescendantSelector(a Selector) Selector {
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
return hasDescendantMatch(n, a)
}
}
// nthChildSelector returns a selector that implements :nth-child(an+b).
// If last is true, implements :nth-last-child instead.
// If ofType is true, implements :nth-of-type instead.
func nthChildSelector(a, b int, last, ofType bool) Selector {
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
parent := n.Parent
if parent == nil {
return false
}
if parent.Type == html.DocumentNode {
return false
}
i := -1
count := 0
for c := parent.FirstChild; c != nil; c = c.NextSibling {
if (c.Type != html.ElementNode) || (ofType && c.Data != n.Data) {
continue
}
count++
if c == n {
i = count
if !last {
break
}
}
}
if i == -1 {
// This shouldn't happen, since n should always be one of its parent's children.
return false
}
if last {
i = count - i + 1
}
i -= b
if a == 0 {
return i == 0
}
return i%a == 0 && i/a >= 0
}
}
// simpleNthChildSelector returns a selector that implements :nth-child(b).
// If ofType is true, implements :nth-of-type instead.
func simpleNthChildSelector(b int, ofType bool) Selector {
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
parent := n.Parent
if parent == nil {
return false
}
if parent.Type == html.DocumentNode {
return false
}
count := 0
for c := parent.FirstChild; c != nil; c = c.NextSibling {
if c.Type != html.ElementNode || (ofType && c.Data != n.Data) {
continue
}
count++
if c == n {
return count == b
}
if count >= b {
return false
}
}
return false
}
}
// simpleNthLastChildSelector returns a selector that implements
// :nth-last-child(b). If ofType is true, implements :nth-last-of-type
// instead.
func simpleNthLastChildSelector(b int, ofType bool) Selector {
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
parent := n.Parent
if parent == nil {
return false
}
if parent.Type == html.DocumentNode {
return false
}
count := 0
for c := parent.LastChild; c != nil; c = c.PrevSibling {
if c.Type != html.ElementNode || (ofType && c.Data != n.Data) {
continue
}
count++
if c == n {
return count == b
}
if count >= b {
return false
}
}
return false
}
}
// onlyChildSelector returns a selector that implements :only-child.
// If ofType is true, it implements :only-of-type instead.
func onlyChildSelector(ofType bool) Selector {
return func(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
parent := n.Parent
if parent == nil {
return false
}
if parent.Type == html.DocumentNode {
return false
}
count := 0
for c := parent.FirstChild; c != nil; c = c.NextSibling {
if (c.Type != html.ElementNode) || (ofType && c.Data != n.Data) {
continue
}
count++
if count > 1 {
return false
}
}
return count == 1
}
}
// inputSelector is a Selector that matches input, select, textarea and button elements.
func inputSelector(n *html.Node) bool {
return n.Type == html.ElementNode && (n.Data == "input" || n.Data == "select" || n.Data == "textarea" || n.Data == "button")
}
// emptyElementSelector is a Selector that matches empty elements.
func emptyElementSelector(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
for c := n.FirstChild; c != nil; c = c.NextSibling {
switch c.Type {
case html.ElementNode, html.TextNode:
return false
}
}
return true
}
// descendantSelector returns a Selector that matches an element if
// it matches d and has an ancestor that matches a.
func descendantSelector(a, d Selector) Selector {
return func(n *html.Node) bool {
if !d(n) {
return false
}
for p := n.Parent; p != nil; p = p.Parent {
if a(p) {
return true
}
}
return false
}
}
// childSelector returns a Selector that matches an element if
// it matches d and its parent matches a.
func childSelector(a, d Selector) Selector {
return func(n *html.Node) bool {
return d(n) && n.Parent != nil && a(n.Parent)
}
}
// siblingSelector returns a Selector that matches an element
// if it matches s2 and in is preceded by an element that matches s1.
// If adjacent is true, the sibling must be immediately before the element.
func siblingSelector(s1, s2 Selector, adjacent bool) Selector {
return func(n *html.Node) bool {
if !s2(n) {
return false
}
if adjacent {
for n = n.PrevSibling; n != nil; n = n.PrevSibling {
if n.Type == html.TextNode || n.Type == html.CommentNode {
continue
}
return s1(n)
}
return false
}
// Walk backwards looking for element that matches s1
for c := n.PrevSibling; c != nil; c = c.PrevSibling {
if s1(c) {
return true
}
}
return false
}
}
// rootSelector implements :root
func rootSelector(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
if n.Parent == nil {
return false
}
return n.Parent.Type == html.DocumentNode
}

View File

@@ -1,8 +1,13 @@
language: go
go:
- 1.11
- 1.12
- 1.13
- 1.14
os:
- linux
- windows
before_install:
- go get -t -v ./...

View File

@@ -1,8 +1,12 @@
# doublestar
Path pattern matching and globbing supporting `doublestar` (`**`) patterns.
![Release](https://img.shields.io/github/release/bmatcuk/doublestar.svg?branch=master)
[![Build Status](https://travis-ci.org/bmatcuk/doublestar.svg?branch=master)](https://travis-ci.org/bmatcuk/doublestar)
[![codecov.io](https://img.shields.io/codecov/c/github/bmatcuk/doublestar.svg?branch=master)](https://codecov.io/github/bmatcuk/doublestar?branch=master)
# doublestar
## About
**doublestar** is a [golang](http://golang.org/) implementation of path pattern
matching and globbing with support for "doublestar" (aka globstar: `**`)
@@ -11,7 +15,7 @@ patterns.
doublestar patterns match files and directories recursively. For example, if
you had the following directory structure:
```
```bash
grandparent
`-- parent
|-- child1
@@ -43,9 +47,10 @@ To use it in your code, you must import it:
import "github.com/bmatcuk/doublestar"
```
## Functions
## Usage
### Match
```go
func Match(pattern, name string) (bool, error)
```
@@ -61,11 +66,12 @@ want to use `PathMatch()` (below) instead.
### PathMatch
```go
func PathMatch(pattern, name string) (bool, error)
```
PathMatch returns true if `name` matches the file name `pattern`
PathMatch returns true if `name` matches the file name `pattern`
([see below](#patterns)). The difference between Match and PathMatch is that
PathMatch will automatically use your system's path separator to split `name`
and `pattern`.
@@ -73,6 +79,7 @@ and `pattern`.
`PathMatch()` is meant to be a drop-in replacement for `filepath.Match()`.
### Glob
```go
func Glob(pattern string) ([]string, error)
```
@@ -83,7 +90,7 @@ directory), or absolute.
`Glob()` is meant to be a drop-in replacement for `filepath.Glob()`.
## Patterns
### Patterns
**doublestar** supports the following special terms in the patterns:
@@ -97,7 +104,7 @@ Special Terms | Meaning
Any character with a special meaning can be escaped with a backslash (`\`).
### Character Classes
#### Character Classes
Character classes support the following:
@@ -107,3 +114,27 @@ Class | Meaning
`[a-z]` | matches any single character in the range
`[^class]` | matches any single character which does *not* match the class
### Abstracting the `os` package
**doublestar** by default uses the `Open`, `Stat`, and `Lstat`, functions and
`PathSeparator` value from the standard library's `os` package. To abstract
this, for example to be able to perform tests of Windows paths on Linux, or to
interoperate with your own filesystem code, it includes the functions `GlobOS`
and `PathMatchOS` which are identical to `Glob` and `PathMatch` except that they
operate on an `OS` interface:
```go
type OS interface {
Lstat(name string) (os.FileInfo, error)
Open(name string) (*os.File, error)
PathSeparator() rune
Stat(name string) (os.FileInfo, error)
}
```
`StandardOS` is a value that implements this interface by calling functions in
the standard library's `os` package.
## License
[MIT License](LICENSE)

View File

@@ -5,10 +5,32 @@ import (
"os"
"path"
"path/filepath"
"sort"
"strings"
"unicode/utf8"
)
// An OS abstracts functions in the standard library's os package.
type OS interface {
Lstat(name string) (os.FileInfo, error)
Open(name string) (*os.File, error)
PathSeparator() rune
Stat(name string) (os.FileInfo, error)
}
// StandardOS is a value that implements the OS interface by calling functions
// in the standard libray's os package.
var StandardOS OS = standardOS{}
// A standardOS implements OS by calling functions in the standard library's os
// package.
type standardOS struct{}
func (standardOS) Lstat(name string) (os.FileInfo, error) { return os.Lstat(name) }
func (standardOS) Open(name string) (*os.File, error) { return os.Open(name) }
func (standardOS) PathSeparator() rune { return os.PathSeparator }
func (standardOS) Stat(name string) (os.FileInfo, error) { return os.Stat(name) }
// ErrBadPattern indicates a pattern was malformed.
var ErrBadPattern = path.ErrBadPattern
@@ -72,6 +94,101 @@ func indexRuneWithEscaping(s string, r rune) int {
return end
}
// Find the last index of a rune in a string,
// ignoring any times the rune is escaped using "\".
func lastIndexRuneWithEscaping(s string, r rune) int {
end := strings.LastIndex(s, string(r))
if end == -1 {
return -1
}
if end > 0 && s[end-1] == '\\' {
end = lastIndexRuneWithEscaping(s[:end-1], r)
}
return end
}
// Find the index of the first instance of one of the unicode characters in
// chars, ignoring any times those characters are escaped using "\".
func indexAnyWithEscaping(s, chars string) int {
end := strings.IndexAny(s, chars)
if end == -1 {
return -1
}
if end > 0 && s[end-1] == '\\' {
_, adj := utf8.DecodeRuneInString(s[end:])
start := end + adj
end = indexAnyWithEscaping(s[start:], chars)
if end != -1 {
end += start
}
}
return end
}
// Split a set of alternatives such as {alt1,alt2,...} and returns the index of
// the rune after the closing curly brace. Respects nested alternatives and
// escaped runes.
func splitAlternatives(s string) (ret []string, idx int) {
ret = make([]string, 0, 2)
idx = 0
slen := len(s)
braceCnt := 1
esc := false
start := 0
for braceCnt > 0 {
if idx >= slen {
return nil, -1
}
sRune, adj := utf8.DecodeRuneInString(s[idx:])
if esc {
esc = false
} else if sRune == '\\' {
esc = true
} else if sRune == '{' {
braceCnt++
} else if sRune == '}' {
braceCnt--
} else if sRune == ',' && braceCnt == 1 {
ret = append(ret, s[start:idx])
start = idx + adj
}
idx += adj
}
ret = append(ret, s[start:idx-1])
return
}
// Returns true if the pattern is "zero length", meaning
// it could match zero or more characters.
func isZeroLengthPattern(pattern string) (ret bool, err error) {
// * can match zero
if pattern == "" || pattern == "*" || pattern == "**" {
return true, nil
}
// an alternative with zero length can match zero, for example {,x} - the
// first alternative has zero length
r, adj := utf8.DecodeRuneInString(pattern)
if r == '{' {
options, endOptions := splitAlternatives(pattern[adj:])
if endOptions == -1 {
return false, ErrBadPattern
}
if ret, err = isZeroLengthPattern(pattern[adj+endOptions:]); !ret || err != nil {
return
}
for _, o := range options {
if ret, err = isZeroLengthPattern(o); ret || err != nil {
return
}
}
}
return false, nil
}
// Match returns true if name matches the shell file name pattern.
// The pattern syntax is:
//
@@ -114,7 +231,13 @@ func Match(pattern, name string) (bool, error) {
// Note: this is meant as a drop-in replacement for filepath.Match().
//
func PathMatch(pattern, name string) (bool, error) {
return matchWithSeparator(pattern, name, os.PathSeparator)
return PathMatchOS(StandardOS, pattern, name)
}
// PathMatchOS is like PathMatch except that it uses vos's path separator.
func PathMatchOS(vos OS, pattern, name string) (bool, error) {
pattern = filepath.ToSlash(pattern)
return matchWithSeparator(pattern, name, vos.PathSeparator())
}
// Match returns true if name matches the shell file name pattern.
@@ -143,49 +266,64 @@ func PathMatch(pattern, name string) (bool, error) {
// is malformed.
//
func matchWithSeparator(pattern, name string, separator rune) (bool, error) {
patternComponents := splitPathOnSeparator(pattern, separator)
nameComponents := splitPathOnSeparator(name, separator)
return doMatching(patternComponents, nameComponents)
return doMatching(pattern, nameComponents)
}
func doMatching(patternComponents, nameComponents []string) (matched bool, err error) {
func doMatching(pattern string, nameComponents []string) (matched bool, err error) {
// check for some base-cases
patternLen, nameLen := len(patternComponents), len(nameComponents)
patternLen, nameLen := len(pattern), len(nameComponents)
if patternLen == 0 && nameLen == 0 {
return true, nil
}
if patternLen == 0 || nameLen == 0 {
if patternLen == 0 {
if nameLen == 1 && nameComponents[0] == "" {
return true, nil
} else if nameLen == 0 {
return false, nil
}
}
slashIdx := indexRuneWithEscaping(pattern, '/')
lastComponent := slashIdx == -1
if lastComponent {
slashIdx = len(pattern)
}
if pattern[:slashIdx] == "**" {
// if our last pattern component is a doublestar, we're done -
// doublestar will match any remaining name components, if any.
if lastComponent {
return true, nil
}
// otherwise, try matching remaining components
for nameIdx := 0; nameIdx < nameLen; nameIdx++ {
if m, _ := doMatching(pattern[slashIdx+1:], nameComponents[nameIdx:]); m {
return true, nil
}
}
return false, nil
}
patIdx, nameIdx := 0, 0
for patIdx < patternLen && nameIdx < nameLen {
if patternComponents[patIdx] == "**" {
// if our last pattern component is a doublestar, we're done -
// doublestar will match any remaining name components, if any.
if patIdx++; patIdx >= patternLen {
return true, nil
}
// otherwise, try matching remaining components
for ; nameIdx < nameLen; nameIdx++ {
if m, _ := doMatching(patternComponents[patIdx:], nameComponents[nameIdx:]); m {
return true, nil
}
}
return false, nil
}
// try matching components
matched, err = matchComponent(patternComponents[patIdx], nameComponents[nameIdx])
if !matched || err != nil {
return
}
patIdx++
nameIdx++
var matches []string
matches, err = matchComponent(pattern, nameComponents[0])
if matches == nil || err != nil {
return
}
return patIdx >= patternLen && nameIdx >= nameLen, nil
if len(matches) == 0 && nameLen == 1 {
return true, nil
}
if nameLen > 1 {
for _, alt := range matches {
matched, err = doMatching(alt, nameComponents[1:])
if matched || err != nil {
return
}
}
}
return false, nil
}
// Glob returns the names of all files matching pattern or nil
@@ -204,58 +342,75 @@ func doMatching(patternComponents, nameComponents []string) (matched bool, err e
// Note: this is meant as a drop-in replacement for filepath.Glob().
//
func Glob(pattern string) (matches []string, err error) {
patternComponents := splitPathOnSeparator(filepath.ToSlash(pattern), '/')
if len(patternComponents) == 0 {
return GlobOS(StandardOS, pattern)
}
// GlobOS is like Glob except that it operates on vos.
func GlobOS(vos OS, pattern string) (matches []string, err error) {
if len(pattern) == 0 {
return nil, nil
}
// On Windows systems, this will return the drive name ('C:') for filesystem
// paths, or \\<server>\<share> for UNC paths. On other systems, it will
// return an empty string. Since absolute paths on non-Windows systems start
// with a slash, patternComponent[0] == volumeName will return true for both
// absolute Windows paths and absolute non-Windows paths, but we need a
// separate check for UNC paths.
volumeName := filepath.VolumeName(pattern)
isWindowsUNC := strings.HasPrefix(pattern, `\\`)
if isWindowsUNC || patternComponents[0] == volumeName {
startComponentIndex := 1
if isWindowsUNC {
startComponentIndex = 4
// if the pattern starts with alternatives, we need to handle that here - the
// alternatives may be a mix of relative and absolute
if pattern[0] == '{' {
options, endOptions := splitAlternatives(pattern[1:])
if endOptions == -1 {
return nil, ErrBadPattern
}
return doGlob(fmt.Sprintf("%s%s", volumeName, string(os.PathSeparator)), patternComponents[startComponentIndex:], matches)
for _, o := range options {
m, e := Glob(o + pattern[endOptions+1:])
if e != nil {
return nil, e
}
matches = append(matches, m...)
}
return matches, nil
}
// If the pattern is relative or absolute and we're on a non-Windows machine,
// volumeName will be an empty string. If it is absolute and we're on a
// Windows machine, volumeName will be a drive letter ("C:") for filesystem
// paths or \\<server>\<share> for UNC paths.
isAbs := filepath.IsAbs(pattern) || pattern[0] == '\\' || pattern[0] == '/'
volumeName := filepath.VolumeName(pattern)
isWindowsUNC := strings.HasPrefix(volumeName, `\\`)
if isWindowsUNC || isAbs {
startIdx := len(volumeName) + 1
return doGlob(vos, fmt.Sprintf("%s%s", volumeName, string(vos.PathSeparator())), filepath.ToSlash(pattern[startIdx:]), matches)
}
// otherwise, it's a relative pattern
return doGlob(".", patternComponents, matches)
return doGlob(vos, ".", filepath.ToSlash(pattern), matches)
}
// Perform a glob
func doGlob(basedir string, components, matches []string) (m []string, e error) {
func doGlob(vos OS, basedir, pattern string, matches []string) (m []string, e error) {
m = matches
e = nil
// figure out how many components we don't need to glob because they're
// just names without patterns - we'll use os.Lstat below to check if that
// path actually exists
patLen := len(components)
patIdx := 0
for ; patIdx < patLen; patIdx++ {
if strings.IndexAny(components[patIdx], "*?[{\\") >= 0 {
break
}
// if the pattern starts with any path components that aren't globbed (ie,
// `path/to/glob*`), we can skip over the un-globbed components (`path/to` in
// our example).
globIdx := indexAnyWithEscaping(pattern, "*?[{\\")
if globIdx > 0 {
globIdx = lastIndexRuneWithEscaping(pattern[:globIdx], '/')
} else if globIdx == -1 {
globIdx = lastIndexRuneWithEscaping(pattern, '/')
}
if patIdx > 0 {
basedir = filepath.Join(basedir, filepath.Join(components[0:patIdx]...))
if globIdx > 0 {
basedir = filepath.Join(basedir, pattern[:globIdx])
pattern = pattern[globIdx+1:]
}
// Lstat will return an error if the file/directory doesn't exist
fi, err := os.Lstat(basedir)
fi, err := vos.Lstat(basedir)
if err != nil {
return
}
// if there are no more components, we've found a match
if patIdx >= patLen {
// if the pattern is empty, we've found a match
if len(pattern) == 0 {
m = append(m, basedir)
return
}
@@ -263,7 +418,7 @@ func doGlob(basedir string, components, matches []string) (m []string, e error)
// otherwise, we need to check each item in the directory...
// first, if basedir is a symlink, follow it...
if (fi.Mode() & os.ModeSymlink) != 0 {
fi, err = os.Stat(basedir)
fi, err = vos.Stat(basedir)
if err != nil {
return
}
@@ -275,20 +430,33 @@ func doGlob(basedir string, components, matches []string) (m []string, e error)
}
// read directory
dir, err := os.Open(basedir)
dir, err := vos.Open(basedir)
if err != nil {
return
}
defer dir.Close()
defer func() {
if err := dir.Close(); e == nil {
e = err
}
}()
files, _ := dir.Readdir(-1)
lastComponent := (patIdx + 1) >= patLen
if components[patIdx] == "**" {
files, err := dir.Readdir(-1)
if err != nil {
return nil, err
}
sort.Slice(files, func(i, j int) bool { return files[i].Name() < files[j].Name() })
slashIdx := indexRuneWithEscaping(pattern, '/')
lastComponent := slashIdx == -1
if lastComponent {
slashIdx = len(pattern)
}
if pattern[:slashIdx] == "**" {
// if the current component is a doublestar, we'll try depth-first
for _, file := range files {
// if symlink, we may want to follow
if (file.Mode() & os.ModeSymlink) != 0 {
file, err = os.Stat(filepath.Join(basedir, file.Name()))
file, err = vos.Stat(filepath.Join(basedir, file.Name()))
if err != nil {
continue
}
@@ -299,7 +467,7 @@ func doGlob(basedir string, components, matches []string) (m []string, e error)
if lastComponent {
m = append(m, filepath.Join(basedir, file.Name()))
}
m, e = doGlob(filepath.Join(basedir, file.Name()), components[patIdx:], m)
m, e = doGlob(vos, filepath.Join(basedir, file.Name()), pattern, m)
} else if lastComponent {
// if the pattern's last component is a doublestar, we match filenames, too
m = append(m, filepath.Join(basedir, file.Name()))
@@ -308,80 +476,82 @@ func doGlob(basedir string, components, matches []string) (m []string, e error)
if lastComponent {
return // we're done
}
patIdx++
lastComponent = (patIdx + 1) >= patLen
pattern = pattern[slashIdx+1:]
}
// check items in current directory and recurse
var match bool
var match []string
for _, file := range files {
match, e = matchComponent(components[patIdx], file.Name())
match, e = matchComponent(pattern, file.Name())
if e != nil {
return
}
if match {
if lastComponent {
if match != nil {
if len(match) == 0 {
m = append(m, filepath.Join(basedir, file.Name()))
} else {
m, e = doGlob(filepath.Join(basedir, file.Name()), components[patIdx+1:], m)
for _, alt := range match {
m, e = doGlob(vos, filepath.Join(basedir, file.Name()), alt, m)
}
}
}
}
return
}
// Attempt to match a single pattern component with a path component
func matchComponent(pattern, name string) (bool, error) {
// check some base cases
patternLen, nameLen := len(pattern), len(name)
if patternLen == 0 && nameLen == 0 {
return true, nil
}
if patternLen == 0 {
return false, nil
}
if nameLen == 0 && pattern != "*" {
return false, nil
}
// Attempt to match a single path component with a pattern. Note that the
// pattern may include multiple components but that the "name" is just a single
// path component. The return value is a slice of patterns that should be
// checked against subsequent path components or nil, indicating that the
// pattern does not match this path. It is assumed that pattern components are
// separated by '/'
func matchComponent(pattern, name string) ([]string, error) {
// check for matches one rune at a time
patternLen, nameLen := len(pattern), len(name)
patIdx, nameIdx := 0, 0
for patIdx < patternLen && nameIdx < nameLen {
patRune, patAdj := utf8.DecodeRuneInString(pattern[patIdx:])
nameRune, nameAdj := utf8.DecodeRuneInString(name[nameIdx:])
if patRune == '\\' {
// handle escaped runes
if patRune == '/' {
patIdx++
break
} else if patRune == '\\' {
// handle escaped runes, only if separator isn't '\\'
patIdx += patAdj
patRune, patAdj = utf8.DecodeRuneInString(pattern[patIdx:])
if patRune == utf8.RuneError {
return false, ErrBadPattern
return nil, ErrBadPattern
} else if patRune == nameRune {
patIdx += patAdj
nameIdx += nameAdj
} else {
return false, nil
return nil, nil
}
} else if patRune == '*' {
// handle stars
// handle stars - a star at the end of the pattern or before a separator
// will always match the rest of the path component
if patIdx += patAdj; patIdx >= patternLen {
// a star at the end of a pattern will always
// match the rest of the path
return true, nil
return []string{}, nil
}
if patRune, patAdj = utf8.DecodeRuneInString(pattern[patIdx:]); patRune == '/' {
return []string{pattern[patIdx+patAdj:]}, nil
}
// check if we can make any matches
for ; nameIdx < nameLen; nameIdx += nameAdj {
if m, _ := matchComponent(pattern[patIdx:], name[nameIdx:]); m {
return true, nil
if m, e := matchComponent(pattern[patIdx:], name[nameIdx:]); m != nil || e != nil {
return m, e
}
_, nameAdj = utf8.DecodeRuneInString(name[nameIdx:])
}
return false, nil
return nil, nil
} else if patRune == '[' {
// handle character sets
patIdx += patAdj
endClass := indexRuneWithEscaping(pattern[patIdx:], ']')
if endClass == -1 {
return false, ErrBadPattern
return nil, ErrBadPattern
}
endClass += patIdx
classRunes := []rune(pattern[patIdx:endClass])
@@ -395,7 +565,7 @@ func matchComponent(pattern, name string) (bool, error) {
for classIdx < classRunesLen {
low := classRunes[classIdx]
if low == '-' {
return false, ErrBadPattern
return nil, ErrBadPattern
}
classIdx++
if low == '\\' {
@@ -403,18 +573,18 @@ func matchComponent(pattern, name string) (bool, error) {
low = classRunes[classIdx]
classIdx++
} else {
return false, ErrBadPattern
return nil, ErrBadPattern
}
}
high := low
if classIdx < classRunesLen && classRunes[classIdx] == '-' {
// we have a range of runes
if classIdx++; classIdx >= classRunesLen {
return false, ErrBadPattern
return nil, ErrBadPattern
}
high = classRunes[classIdx]
if high == '-' {
return false, ErrBadPattern
return nil, ErrBadPattern
}
classIdx++
if high == '\\' {
@@ -422,7 +592,7 @@ func matchComponent(pattern, name string) (bool, error) {
high = classRunes[classIdx]
classIdx++
} else {
return false, ErrBadPattern
return nil, ErrBadPattern
}
}
}
@@ -431,46 +601,74 @@ func matchComponent(pattern, name string) (bool, error) {
}
}
if matchClass == (classRunes[0] == '^') {
return false, nil
return nil, nil
}
} else {
return false, ErrBadPattern
return nil, ErrBadPattern
}
patIdx = endClass + 1
nameIdx += nameAdj
} else if patRune == '{' {
// handle alternatives such as {alt1,alt2,...}
patIdx += patAdj
endOptions := indexRuneWithEscaping(pattern[patIdx:], '}')
options, endOptions := splitAlternatives(pattern[patIdx:])
if endOptions == -1 {
return false, ErrBadPattern
return nil, ErrBadPattern
}
endOptions += patIdx
options := splitPathOnSeparator(pattern[patIdx:endOptions], ',')
patIdx = endOptions + 1
patIdx += endOptions
results := make([][]string, 0, len(options))
totalResults := 0
for _, o := range options {
m, e := matchComponent(o+pattern[patIdx:], name[nameIdx:])
if e != nil {
return false, e
return nil, e
}
if m {
return true, nil
if m != nil {
results = append(results, m)
totalResults += len(m)
}
}
return false, nil
if len(results) > 0 {
lst := make([]string, 0, totalResults)
for _, m := range results {
lst = append(lst, m...)
}
return lst, nil
}
return nil, nil
} else if patRune == '?' || patRune == nameRune {
// handle single-rune wildcard
patIdx += patAdj
nameIdx += nameAdj
} else {
return false, nil
return nil, nil
}
}
if patIdx >= patternLen && nameIdx >= nameLen {
return true, nil
if nameIdx >= nameLen {
if patIdx >= patternLen {
return []string{}, nil
}
pattern = pattern[patIdx:]
slashIdx := indexRuneWithEscaping(pattern, '/')
testPattern := pattern
if slashIdx >= 0 {
testPattern = pattern[:slashIdx]
}
zeroLength, err := isZeroLengthPattern(testPattern)
if err != nil {
return nil, err
}
if zeroLength {
if slashIdx == -1 {
return []string{}, nil
} else {
return []string{pattern[slashIdx+1:]}, nil
}
}
}
if nameIdx >= nameLen && pattern[patIdx:] == "*" || pattern[patIdx:] == "**" {
return true, nil
}
return false, nil
return nil, nil
}

View File

@@ -1,712 +0,0 @@
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
//go:generate go run gen.go
//go:generate go run gen.go -test
package main
import (
"bytes"
"flag"
"fmt"
"go/format"
"io/ioutil"
"math/rand"
"os"
"sort"
"strings"
)
// identifier converts s to a Go exported identifier.
// It converts "div" to "Div" and "accept-charset" to "AcceptCharset".
func identifier(s string) string {
b := make([]byte, 0, len(s))
cap := true
for _, c := range s {
if c == '-' {
cap = true
continue
}
if cap && 'a' <= c && c <= 'z' {
c -= 'a' - 'A'
}
cap = false
b = append(b, byte(c))
}
return string(b)
}
var test = flag.Bool("test", false, "generate table_test.go")
func genFile(name string, buf *bytes.Buffer) {
b, err := format.Source(buf.Bytes())
if err != nil {
fmt.Fprintln(os.Stderr, err)
os.Exit(1)
}
if err := ioutil.WriteFile(name, b, 0644); err != nil {
fmt.Fprintln(os.Stderr, err)
os.Exit(1)
}
}
func main() {
flag.Parse()
var all []string
all = append(all, elements...)
all = append(all, attributes...)
all = append(all, eventHandlers...)
all = append(all, extra...)
sort.Strings(all)
// uniq - lists have dups
w := 0
for _, s := range all {
if w == 0 || all[w-1] != s {
all[w] = s
w++
}
}
all = all[:w]
if *test {
var buf bytes.Buffer
fmt.Fprintln(&buf, "// Code generated by go generate gen.go; DO NOT EDIT.\n")
fmt.Fprintln(&buf, "//go:generate go run gen.go -test\n")
fmt.Fprintln(&buf, "package atom\n")
fmt.Fprintln(&buf, "var testAtomList = []string{")
for _, s := range all {
fmt.Fprintf(&buf, "\t%q,\n", s)
}
fmt.Fprintln(&buf, "}")
genFile("table_test.go", &buf)
return
}
// Find hash that minimizes table size.
var best *table
for i := 0; i < 1000000; i++ {
if best != nil && 1<<(best.k-1) < len(all) {
break
}
h := rand.Uint32()
for k := uint(0); k <= 16; k++ {
if best != nil && k >= best.k {
break
}
var t table
if t.init(h, k, all) {
best = &t
break
}
}
}
if best == nil {
fmt.Fprintf(os.Stderr, "failed to construct string table\n")
os.Exit(1)
}
// Lay out strings, using overlaps when possible.
layout := append([]string{}, all...)
// Remove strings that are substrings of other strings
for changed := true; changed; {
changed = false
for i, s := range layout {
if s == "" {
continue
}
for j, t := range layout {
if i != j && t != "" && strings.Contains(s, t) {
changed = true
layout[j] = ""
}
}
}
}
// Join strings where one suffix matches another prefix.
for {
// Find best i, j, k such that layout[i][len-k:] == layout[j][:k],
// maximizing overlap length k.
besti := -1
bestj := -1
bestk := 0
for i, s := range layout {
if s == "" {
continue
}
for j, t := range layout {
if i == j {
continue
}
for k := bestk + 1; k <= len(s) && k <= len(t); k++ {
if s[len(s)-k:] == t[:k] {
besti = i
bestj = j
bestk = k
}
}
}
}
if bestk > 0 {
layout[besti] += layout[bestj][bestk:]
layout[bestj] = ""
continue
}
break
}
text := strings.Join(layout, "")
atom := map[string]uint32{}
for _, s := range all {
off := strings.Index(text, s)
if off < 0 {
panic("lost string " + s)
}
atom[s] = uint32(off<<8 | len(s))
}
var buf bytes.Buffer
// Generate the Go code.
fmt.Fprintln(&buf, "// Code generated by go generate gen.go; DO NOT EDIT.\n")
fmt.Fprintln(&buf, "//go:generate go run gen.go\n")
fmt.Fprintln(&buf, "package atom\n\nconst (")
// compute max len
maxLen := 0
for _, s := range all {
if maxLen < len(s) {
maxLen = len(s)
}
fmt.Fprintf(&buf, "\t%s Atom = %#x\n", identifier(s), atom[s])
}
fmt.Fprintln(&buf, ")\n")
fmt.Fprintf(&buf, "const hash0 = %#x\n\n", best.h0)
fmt.Fprintf(&buf, "const maxAtomLen = %d\n\n", maxLen)
fmt.Fprintf(&buf, "var table = [1<<%d]Atom{\n", best.k)
for i, s := range best.tab {
if s == "" {
continue
}
fmt.Fprintf(&buf, "\t%#x: %#x, // %s\n", i, atom[s], s)
}
fmt.Fprintf(&buf, "}\n")
datasize := (1 << best.k) * 4
fmt.Fprintln(&buf, "const atomText =")
textsize := len(text)
for len(text) > 60 {
fmt.Fprintf(&buf, "\t%q +\n", text[:60])
text = text[60:]
}
fmt.Fprintf(&buf, "\t%q\n\n", text)
genFile("table.go", &buf)
fmt.Fprintf(os.Stdout, "%d atoms; %d string bytes + %d tables = %d total data\n", len(all), textsize, datasize, textsize+datasize)
}
type byLen []string
func (x byLen) Less(i, j int) bool { return len(x[i]) > len(x[j]) }
func (x byLen) Swap(i, j int) { x[i], x[j] = x[j], x[i] }
func (x byLen) Len() int { return len(x) }
// fnv computes the FNV hash with an arbitrary starting value h.
func fnv(h uint32, s string) uint32 {
for i := 0; i < len(s); i++ {
h ^= uint32(s[i])
h *= 16777619
}
return h
}
// A table represents an attempt at constructing the lookup table.
// The lookup table uses cuckoo hashing, meaning that each string
// can be found in one of two positions.
type table struct {
h0 uint32
k uint
mask uint32
tab []string
}
// hash returns the two hashes for s.
func (t *table) hash(s string) (h1, h2 uint32) {
h := fnv(t.h0, s)
h1 = h & t.mask
h2 = (h >> 16) & t.mask
return
}
// init initializes the table with the given parameters.
// h0 is the initial hash value,
// k is the number of bits of hash value to use, and
// x is the list of strings to store in the table.
// init returns false if the table cannot be constructed.
func (t *table) init(h0 uint32, k uint, x []string) bool {
t.h0 = h0
t.k = k
t.tab = make([]string, 1<<k)
t.mask = 1<<k - 1
for _, s := range x {
if !t.insert(s) {
return false
}
}
return true
}
// insert inserts s in the table.
func (t *table) insert(s string) bool {
h1, h2 := t.hash(s)
if t.tab[h1] == "" {
t.tab[h1] = s
return true
}
if t.tab[h2] == "" {
t.tab[h2] = s
return true
}
if t.push(h1, 0) {
t.tab[h1] = s
return true
}
if t.push(h2, 0) {
t.tab[h2] = s
return true
}
return false
}
// push attempts to push aside the entry in slot i.
func (t *table) push(i uint32, depth int) bool {
if depth > len(t.tab) {
return false
}
s := t.tab[i]
h1, h2 := t.hash(s)
j := h1 + h2 - i
if t.tab[j] != "" && !t.push(j, depth+1) {
return false
}
t.tab[j] = s
return true
}
// The lists of element names and attribute keys were taken from
// https://html.spec.whatwg.org/multipage/indices.html#index
// as of the "HTML Living Standard - Last Updated 16 April 2018" version.
// "command", "keygen" and "menuitem" have been removed from the spec,
// but are kept here for backwards compatibility.
var elements = []string{
"a",
"abbr",
"address",
"area",
"article",
"aside",
"audio",
"b",
"base",
"bdi",
"bdo",
"blockquote",
"body",
"br",
"button",
"canvas",
"caption",
"cite",
"code",
"col",
"colgroup",
"command",
"data",
"datalist",
"dd",
"del",
"details",
"dfn",
"dialog",
"div",
"dl",
"dt",
"em",
"embed",
"fieldset",
"figcaption",
"figure",
"footer",
"form",
"h1",
"h2",
"h3",
"h4",
"h5",
"h6",
"head",
"header",
"hgroup",
"hr",
"html",
"i",
"iframe",
"img",
"input",
"ins",
"kbd",
"keygen",
"label",
"legend",
"li",
"link",
"main",
"map",
"mark",
"menu",
"menuitem",
"meta",
"meter",
"nav",
"noscript",
"object",
"ol",
"optgroup",
"option",
"output",
"p",
"param",
"picture",
"pre",
"progress",
"q",
"rp",
"rt",
"ruby",
"s",
"samp",
"script",
"section",
"select",
"slot",
"small",
"source",
"span",
"strong",
"style",
"sub",
"summary",
"sup",
"table",
"tbody",
"td",
"template",
"textarea",
"tfoot",
"th",
"thead",
"time",
"title",
"tr",
"track",
"u",
"ul",
"var",
"video",
"wbr",
}
// https://html.spec.whatwg.org/multipage/indices.html#attributes-3
//
// "challenge", "command", "contextmenu", "dropzone", "icon", "keytype", "mediagroup",
// "radiogroup", "spellcheck", "scoped", "seamless", "sortable" and "sorted" have been removed from the spec,
// but are kept here for backwards compatibility.
var attributes = []string{
"abbr",
"accept",
"accept-charset",
"accesskey",
"action",
"allowfullscreen",
"allowpaymentrequest",
"allowusermedia",
"alt",
"as",
"async",
"autocomplete",
"autofocus",
"autoplay",
"challenge",
"charset",
"checked",
"cite",
"class",
"color",
"cols",
"colspan",
"command",
"content",
"contenteditable",
"contextmenu",
"controls",
"coords",
"crossorigin",
"data",
"datetime",
"default",
"defer",
"dir",
"dirname",
"disabled",
"download",
"draggable",
"dropzone",
"enctype",
"for",
"form",
"formaction",
"formenctype",
"formmethod",
"formnovalidate",
"formtarget",
"headers",
"height",
"hidden",
"high",
"href",
"hreflang",
"http-equiv",
"icon",
"id",
"inputmode",
"integrity",
"is",
"ismap",
"itemid",
"itemprop",
"itemref",
"itemscope",
"itemtype",
"keytype",
"kind",
"label",
"lang",
"list",
"loop",
"low",
"manifest",
"max",
"maxlength",
"media",
"mediagroup",
"method",
"min",
"minlength",
"multiple",
"muted",
"name",
"nomodule",
"nonce",
"novalidate",
"open",
"optimum",
"pattern",
"ping",
"placeholder",
"playsinline",
"poster",
"preload",
"radiogroup",
"readonly",
"referrerpolicy",
"rel",
"required",
"reversed",
"rows",
"rowspan",
"sandbox",
"spellcheck",
"scope",
"scoped",
"seamless",
"selected",
"shape",
"size",
"sizes",
"sortable",
"sorted",
"slot",
"span",
"spellcheck",
"src",
"srcdoc",
"srclang",
"srcset",
"start",
"step",
"style",
"tabindex",
"target",
"title",
"translate",
"type",
"typemustmatch",
"updateviacache",
"usemap",
"value",
"width",
"workertype",
"wrap",
}
// "onautocomplete", "onautocompleteerror", "onmousewheel",
// "onshow" and "onsort" have been removed from the spec,
// but are kept here for backwards compatibility.
var eventHandlers = []string{
"onabort",
"onautocomplete",
"onautocompleteerror",
"onauxclick",
"onafterprint",
"onbeforeprint",
"onbeforeunload",
"onblur",
"oncancel",
"oncanplay",
"oncanplaythrough",
"onchange",
"onclick",
"onclose",
"oncontextmenu",
"oncopy",
"oncuechange",
"oncut",
"ondblclick",
"ondrag",
"ondragend",
"ondragenter",
"ondragexit",
"ondragleave",
"ondragover",
"ondragstart",
"ondrop",
"ondurationchange",
"onemptied",
"onended",
"onerror",
"onfocus",
"onhashchange",
"oninput",
"oninvalid",
"onkeydown",
"onkeypress",
"onkeyup",
"onlanguagechange",
"onload",
"onloadeddata",
"onloadedmetadata",
"onloadend",
"onloadstart",
"onmessage",
"onmessageerror",
"onmousedown",
"onmouseenter",
"onmouseleave",
"onmousemove",
"onmouseout",
"onmouseover",
"onmouseup",
"onmousewheel",
"onwheel",
"onoffline",
"ononline",
"onpagehide",
"onpageshow",
"onpaste",
"onpause",
"onplay",
"onplaying",
"onpopstate",
"onprogress",
"onratechange",
"onreset",
"onresize",
"onrejectionhandled",
"onscroll",
"onsecuritypolicyviolation",
"onseeked",
"onseeking",
"onselect",
"onshow",
"onsort",
"onstalled",
"onstorage",
"onsubmit",
"onsuspend",
"ontimeupdate",
"ontoggle",
"onunhandledrejection",
"onunload",
"onvolumechange",
"onwaiting",
}
// extra are ad-hoc values not covered by any of the lists above.
var extra = []string{
"acronym",
"align",
"annotation",
"annotation-xml",
"applet",
"basefont",
"bgsound",
"big",
"blink",
"center",
"color",
"desc",
"face",
"font",
"foreignObject", // HTML is case-insensitive, but SVG-embedded-in-HTML is case-sensitive.
"foreignobject",
"frame",
"frameset",
"image",
"isindex",
"listing",
"malignmark",
"marquee",
"math",
"mglyph",
"mi",
"mn",
"mo",
"ms",
"mtext",
"nobr",
"noembed",
"noframes",
"plaintext",
"prompt",
"public",
"rb",
"rtc",
"spacer",
"strike",
"svg",
"system",
"tt",
"xmp",
}

View File

@@ -1,78 +0,0 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// mkasm_darwin.go generates assembly trampolines to call libSystem routines from Go.
//This program must be run after mksyscall.go.
package main
import (
"bytes"
"fmt"
"io/ioutil"
"log"
"os"
"strings"
)
func writeASMFile(in string, fileName string, buildTags string) {
trampolines := map[string]bool{}
var out bytes.Buffer
fmt.Fprintf(&out, "// go run mkasm_darwin.go %s\n", strings.Join(os.Args[1:], " "))
fmt.Fprintf(&out, "// Code generated by the command above; DO NOT EDIT.\n")
fmt.Fprintf(&out, "\n")
fmt.Fprintf(&out, "// +build %s\n", buildTags)
fmt.Fprintf(&out, "\n")
fmt.Fprintf(&out, "#include \"textflag.h\"\n")
for _, line := range strings.Split(in, "\n") {
if !strings.HasPrefix(line, "func ") || !strings.HasSuffix(line, "_trampoline()") {
continue
}
fn := line[5 : len(line)-13]
if !trampolines[fn] {
trampolines[fn] = true
fmt.Fprintf(&out, "TEXT ·%s_trampoline(SB),NOSPLIT,$0-0\n", fn)
fmt.Fprintf(&out, "\tJMP\t%s(SB)\n", fn)
}
}
err := ioutil.WriteFile(fileName, out.Bytes(), 0644)
if err != nil {
log.Fatalf("can't write %s: %s", fileName, err)
}
}
func main() {
in1, err := ioutil.ReadFile("syscall_darwin.go")
if err != nil {
log.Fatalf("can't open syscall_darwin.go: %s", err)
}
arch := os.Args[1]
in2, err := ioutil.ReadFile(fmt.Sprintf("syscall_darwin_%s.go", arch))
if err != nil {
log.Fatalf("can't open syscall_darwin_%s.go: %s", arch, err)
}
in3, err := ioutil.ReadFile(fmt.Sprintf("zsyscall_darwin_%s.go", arch))
if err != nil {
log.Fatalf("can't open zsyscall_darwin_%s.go: %s", arch, err)
}
in := string(in1) + string(in2) + string(in3)
writeASMFile(in, fmt.Sprintf("zsyscall_darwin_%s.s", arch), "go1.12")
in1, err = ioutil.ReadFile("syscall_darwin.1_13.go")
if err != nil {
log.Fatalf("can't open syscall_darwin.1_13.go: %s", err)
}
in2, err = ioutil.ReadFile(fmt.Sprintf("zsyscall_darwin_%s.1_13.go", arch))
if err != nil {
log.Fatalf("can't open zsyscall_darwin_%s.1_13.go: %s", arch, err)
}
in = string(in1) + string(in2)
writeASMFile(in, fmt.Sprintf("zsyscall_darwin_%s.1_13.s", arch), "go1.13")
}

View File

@@ -1,521 +0,0 @@
// Copyright 2020 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// mkmerge.go parses generated source files and merges common
// consts, funcs, and types into a common source file, per GOOS.
//
// Usage:
// $ go run mkmerge.go -out MERGED FILE [FILE ...]
//
// Example:
// # Remove all common consts, funcs, and types from zerrors_linux_*.go
// # and write the common code into zerrors_linux.go
// $ go run mkmerge.go -out zerrors_linux.go zerrors_linux_*.go
//
// mkmerge.go performs the merge in the following steps:
// 1. Construct the set of common code that is idential in all
// architecture-specific files.
// 2. Write this common code to the merged file.
// 3. Remove the common code from all architecture-specific files.
package main
import (
"bufio"
"bytes"
"flag"
"fmt"
"go/ast"
"go/format"
"go/parser"
"go/token"
"io"
"io/ioutil"
"log"
"os"
"path"
"path/filepath"
"regexp"
"strconv"
"strings"
)
const validGOOS = "aix|darwin|dragonfly|freebsd|linux|netbsd|openbsd|solaris"
// getValidGOOS returns GOOS, true if filename ends with a valid "_GOOS.go"
func getValidGOOS(filename string) (string, bool) {
matches := regexp.MustCompile(`_(` + validGOOS + `)\.go$`).FindStringSubmatch(filename)
if len(matches) != 2 {
return "", false
}
return matches[1], true
}
// codeElem represents an ast.Decl in a comparable way.
type codeElem struct {
tok token.Token // e.g. token.CONST, token.TYPE, or token.FUNC
src string // the declaration formatted as source code
}
// newCodeElem returns a codeElem based on tok and node, or an error is returned.
func newCodeElem(tok token.Token, node ast.Node) (codeElem, error) {
var b strings.Builder
err := format.Node(&b, token.NewFileSet(), node)
if err != nil {
return codeElem{}, err
}
return codeElem{tok, b.String()}, nil
}
// codeSet is a set of codeElems
type codeSet struct {
set map[codeElem]bool // true for all codeElems in the set
}
// newCodeSet returns a new codeSet
func newCodeSet() *codeSet { return &codeSet{make(map[codeElem]bool)} }
// add adds elem to c
func (c *codeSet) add(elem codeElem) { c.set[elem] = true }
// has returns true if elem is in c
func (c *codeSet) has(elem codeElem) bool { return c.set[elem] }
// isEmpty returns true if the set is empty
func (c *codeSet) isEmpty() bool { return len(c.set) == 0 }
// intersection returns a new set which is the intersection of c and a
func (c *codeSet) intersection(a *codeSet) *codeSet {
res := newCodeSet()
for elem := range c.set {
if a.has(elem) {
res.add(elem)
}
}
return res
}
// keepCommon is a filterFn for filtering the merged file with common declarations.
func (c *codeSet) keepCommon(elem codeElem) bool {
switch elem.tok {
case token.VAR:
// Remove all vars from the merged file
return false
case token.CONST, token.TYPE, token.FUNC, token.COMMENT:
// Remove arch-specific consts, types, functions, and file-level comments from the merged file
return c.has(elem)
case token.IMPORT:
// Keep imports, they are handled by filterImports
return true
}
log.Fatalf("keepCommon: invalid elem %v", elem)
return true
}
// keepArchSpecific is a filterFn for filtering the GOARC-specific files.
func (c *codeSet) keepArchSpecific(elem codeElem) bool {
switch elem.tok {
case token.CONST, token.TYPE, token.FUNC:
// Remove common consts, types, or functions from the arch-specific file
return !c.has(elem)
}
return true
}
// srcFile represents a source file
type srcFile struct {
name string
src []byte
}
// filterFn is a helper for filter
type filterFn func(codeElem) bool
// filter parses and filters Go source code from src, removing top
// level declarations using keep as predicate.
// For src parameter, please see docs for parser.ParseFile.
func filter(src interface{}, keep filterFn) ([]byte, error) {
// Parse the src into an ast
fset := token.NewFileSet()
f, err := parser.ParseFile(fset, "", src, parser.ParseComments)
if err != nil {
return nil, err
}
cmap := ast.NewCommentMap(fset, f, f.Comments)
// Group const/type specs on adjacent lines
var groups specGroups = make(map[string]int)
var groupID int
decls := f.Decls
f.Decls = f.Decls[:0]
for _, decl := range decls {
switch decl := decl.(type) {
case *ast.GenDecl:
// Filter imports, consts, types, vars
specs := decl.Specs
decl.Specs = decl.Specs[:0]
for i, spec := range specs {
elem, err := newCodeElem(decl.Tok, spec)
if err != nil {
return nil, err
}
// Create new group if there are empty lines between this and the previous spec
if i > 0 && fset.Position(specs[i-1].End()).Line < fset.Position(spec.Pos()).Line-1 {
groupID++
}
// Check if we should keep this spec
if keep(elem) {
decl.Specs = append(decl.Specs, spec)
groups.add(elem.src, groupID)
}
}
// Check if we should keep this decl
if len(decl.Specs) > 0 {
f.Decls = append(f.Decls, decl)
}
case *ast.FuncDecl:
// Filter funcs
elem, err := newCodeElem(token.FUNC, decl)
if err != nil {
return nil, err
}
if keep(elem) {
f.Decls = append(f.Decls, decl)
}
}
}
// Filter file level comments
if cmap[f] != nil {
commentGroups := cmap[f]
cmap[f] = cmap[f][:0]
for _, cGrp := range commentGroups {
if keep(codeElem{token.COMMENT, cGrp.Text()}) {
cmap[f] = append(cmap[f], cGrp)
}
}
}
f.Comments = cmap.Filter(f).Comments()
// Generate code for the filtered ast
var buf bytes.Buffer
if err = format.Node(&buf, fset, f); err != nil {
return nil, err
}
groupedSrc, err := groups.filterEmptyLines(&buf)
if err != nil {
return nil, err
}
return filterImports(groupedSrc)
}
// getCommonSet returns the set of consts, types, and funcs that are present in every file.
func getCommonSet(files []srcFile) (*codeSet, error) {
if len(files) == 0 {
return nil, fmt.Errorf("no files provided")
}
// Use the first architecture file as the baseline
baseSet, err := getCodeSet(files[0].src)
if err != nil {
return nil, err
}
// Compare baseline set with other architecture files: discard any element,
// that doesn't exist in other architecture files.
for _, f := range files[1:] {
set, err := getCodeSet(f.src)
if err != nil {
return nil, err
}
baseSet = baseSet.intersection(set)
}
return baseSet, nil
}
// getCodeSet returns the set of all top-level consts, types, and funcs from src.
// src must be string, []byte, or io.Reader (see go/parser.ParseFile docs)
func getCodeSet(src interface{}) (*codeSet, error) {
set := newCodeSet()
fset := token.NewFileSet()
f, err := parser.ParseFile(fset, "", src, parser.ParseComments)
if err != nil {
return nil, err
}
for _, decl := range f.Decls {
switch decl := decl.(type) {
case *ast.GenDecl:
// Add const, and type declarations
if !(decl.Tok == token.CONST || decl.Tok == token.TYPE) {
break
}
for _, spec := range decl.Specs {
elem, err := newCodeElem(decl.Tok, spec)
if err != nil {
return nil, err
}
set.add(elem)
}
case *ast.FuncDecl:
// Add func declarations
elem, err := newCodeElem(token.FUNC, decl)
if err != nil {
return nil, err
}
set.add(elem)
}
}
// Add file level comments
cmap := ast.NewCommentMap(fset, f, f.Comments)
for _, cGrp := range cmap[f] {
set.add(codeElem{token.COMMENT, cGrp.Text()})
}
return set, nil
}
// importName returns the identifier (PackageName) for an imported package
func importName(iSpec *ast.ImportSpec) (string, error) {
if iSpec.Name == nil {
name, err := strconv.Unquote(iSpec.Path.Value)
if err != nil {
return "", err
}
return path.Base(name), nil
}
return iSpec.Name.Name, nil
}
// specGroups tracks grouped const/type specs with a map of line: groupID pairs
type specGroups map[string]int
// add spec source to group
func (s specGroups) add(src string, groupID int) error {
srcBytes, err := format.Source(bytes.TrimSpace([]byte(src)))
if err != nil {
return err
}
s[string(srcBytes)] = groupID
return nil
}
// filterEmptyLines removes empty lines within groups of const/type specs.
// Returns the filtered source.
func (s specGroups) filterEmptyLines(src io.Reader) ([]byte, error) {
scanner := bufio.NewScanner(src)
var out bytes.Buffer
var emptyLines bytes.Buffer
prevGroupID := -1 // Initialize to invalid group
for scanner.Scan() {
line := bytes.TrimSpace(scanner.Bytes())
if len(line) == 0 {
fmt.Fprintf(&emptyLines, "%s\n", scanner.Bytes())
continue
}
// Discard emptyLines if previous non-empty line belonged to the same
// group as this line
if src, err := format.Source(line); err == nil {
groupID, ok := s[string(src)]
if ok && groupID == prevGroupID {
emptyLines.Reset()
}
prevGroupID = groupID
}
emptyLines.WriteTo(&out)
fmt.Fprintf(&out, "%s\n", scanner.Bytes())
}
if err := scanner.Err(); err != nil {
return nil, err
}
return out.Bytes(), nil
}
// filterImports removes unused imports from fileSrc, and returns a formatted src.
func filterImports(fileSrc []byte) ([]byte, error) {
fset := token.NewFileSet()
file, err := parser.ParseFile(fset, "", fileSrc, parser.ParseComments)
if err != nil {
return nil, err
}
cmap := ast.NewCommentMap(fset, file, file.Comments)
// create set of references to imported identifiers
keepImport := make(map[string]bool)
for _, u := range file.Unresolved {
keepImport[u.Name] = true
}
// filter import declarations
decls := file.Decls
file.Decls = file.Decls[:0]
for _, decl := range decls {
importDecl, ok := decl.(*ast.GenDecl)
// Keep non-import declarations
if !ok || importDecl.Tok != token.IMPORT {
file.Decls = append(file.Decls, decl)
continue
}
// Filter the import specs
specs := importDecl.Specs
importDecl.Specs = importDecl.Specs[:0]
for _, spec := range specs {
iSpec := spec.(*ast.ImportSpec)
name, err := importName(iSpec)
if err != nil {
return nil, err
}
if keepImport[name] {
importDecl.Specs = append(importDecl.Specs, iSpec)
}
}
if len(importDecl.Specs) > 0 {
file.Decls = append(file.Decls, importDecl)
}
}
// filter file.Imports
imports := file.Imports
file.Imports = file.Imports[:0]
for _, spec := range imports {
name, err := importName(spec)
if err != nil {
return nil, err
}
if keepImport[name] {
file.Imports = append(file.Imports, spec)
}
}
file.Comments = cmap.Filter(file).Comments()
var buf bytes.Buffer
err = format.Node(&buf, fset, file)
if err != nil {
return nil, err
}
return buf.Bytes(), nil
}
// merge extracts duplicate code from archFiles and merges it to mergeFile.
// 1. Construct commonSet: the set of code that is idential in all archFiles.
// 2. Write the code in commonSet to mergedFile.
// 3. Remove the commonSet code from all archFiles.
func merge(mergedFile string, archFiles ...string) error {
// extract and validate the GOOS part of the merged filename
goos, ok := getValidGOOS(mergedFile)
if !ok {
return fmt.Errorf("invalid GOOS in merged file name %s", mergedFile)
}
// Read architecture files
var inSrc []srcFile
for _, file := range archFiles {
src, err := ioutil.ReadFile(file)
if err != nil {
return fmt.Errorf("cannot read archfile %s: %w", file, err)
}
inSrc = append(inSrc, srcFile{file, src})
}
// 1. Construct the set of top-level declarations common for all files
commonSet, err := getCommonSet(inSrc)
if err != nil {
return err
}
if commonSet.isEmpty() {
// No common code => do not modify any files
return nil
}
// 2. Write the merged file
mergedSrc, err := filter(inSrc[0].src, commonSet.keepCommon)
if err != nil {
return err
}
f, err := os.Create(mergedFile)
if err != nil {
return err
}
buf := bufio.NewWriter(f)
fmt.Fprintln(buf, "// Code generated by mkmerge.go; DO NOT EDIT.")
fmt.Fprintln(buf)
fmt.Fprintf(buf, "// +build %s\n", goos)
fmt.Fprintln(buf)
buf.Write(mergedSrc)
err = buf.Flush()
if err != nil {
return err
}
err = f.Close()
if err != nil {
return err
}
// 3. Remove duplicate declarations from the architecture files
for _, inFile := range inSrc {
src, err := filter(inFile.src, commonSet.keepArchSpecific)
if err != nil {
return err
}
err = ioutil.WriteFile(inFile.name, src, 0644)
if err != nil {
return err
}
}
return nil
}
func main() {
var mergedFile string
flag.StringVar(&mergedFile, "out", "", "Write merged code to `FILE`")
flag.Parse()
// Expand wildcards
var filenames []string
for _, arg := range flag.Args() {
matches, err := filepath.Glob(arg)
if err != nil {
fmt.Fprintf(os.Stderr, "Invalid command line argument %q: %v\n", arg, err)
os.Exit(1)
}
filenames = append(filenames, matches...)
}
if len(filenames) < 2 {
// No need to merge
return
}
err := merge(mergedFile, filenames...)
if err != nil {
fmt.Fprintf(os.Stderr, "Merge failed with error: %v\n", err)
os.Exit(1)
}
}

View File

@@ -1,127 +0,0 @@
// Copyright 2016 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// mkpost processes the output of cgo -godefs to
// modify the generated types. It is used to clean up
// the sys API in an architecture specific manner.
//
// mkpost is run after cgo -godefs; see README.md.
package main
import (
"bytes"
"fmt"
"go/format"
"io/ioutil"
"log"
"os"
"regexp"
)
func main() {
// Get the OS and architecture (using GOARCH_TARGET if it exists)
goos := os.Getenv("GOOS")
goarch := os.Getenv("GOARCH_TARGET")
if goarch == "" {
goarch = os.Getenv("GOARCH")
}
// Check that we are using the Docker-based build system if we should be.
if goos == "linux" {
if os.Getenv("GOLANG_SYS_BUILD") != "docker" {
os.Stderr.WriteString("In the Docker-based build system, mkpost should not be called directly.\n")
os.Stderr.WriteString("See README.md\n")
os.Exit(1)
}
}
b, err := ioutil.ReadAll(os.Stdin)
if err != nil {
log.Fatal(err)
}
if goos == "aix" {
// Replace type of Atim, Mtim and Ctim by Timespec in Stat_t
// to avoid having both StTimespec and Timespec.
sttimespec := regexp.MustCompile(`_Ctype_struct_st_timespec`)
b = sttimespec.ReplaceAll(b, []byte("Timespec"))
}
// Intentionally export __val fields in Fsid and Sigset_t
valRegex := regexp.MustCompile(`type (Fsid|Sigset_t) struct {(\s+)X__(bits|val)(\s+\S+\s+)}`)
b = valRegex.ReplaceAll(b, []byte("type $1 struct {${2}Val$4}"))
// Intentionally export __fds_bits field in FdSet
fdSetRegex := regexp.MustCompile(`type (FdSet) struct {(\s+)X__fds_bits(\s+\S+\s+)}`)
b = fdSetRegex.ReplaceAll(b, []byte("type $1 struct {${2}Bits$3}"))
// If we have empty Ptrace structs, we should delete them. Only s390x emits
// nonempty Ptrace structs.
ptraceRexexp := regexp.MustCompile(`type Ptrace((Psw|Fpregs|Per) struct {\s*})`)
b = ptraceRexexp.ReplaceAll(b, nil)
// Replace the control_regs union with a blank identifier for now.
controlRegsRegex := regexp.MustCompile(`(Control_regs)\s+\[0\]uint64`)
b = controlRegsRegex.ReplaceAll(b, []byte("_ [0]uint64"))
// Remove fields that are added by glibc
// Note that this is unstable as the identifers are private.
removeFieldsRegex := regexp.MustCompile(`X__glibc\S*`)
b = removeFieldsRegex.ReplaceAll(b, []byte("_"))
// Convert [65]int8 to [65]byte in Utsname members to simplify
// conversion to string; see golang.org/issue/20753
convertUtsnameRegex := regexp.MustCompile(`((Sys|Node|Domain)name|Release|Version|Machine)(\s+)\[(\d+)\]u?int8`)
b = convertUtsnameRegex.ReplaceAll(b, []byte("$1$3[$4]byte"))
// Convert [n]int8 to [n]byte in Statvfs_t members to simplify
// conversion to string.
convertStatvfsRegex := regexp.MustCompile(`((Fstype|Mnton|Mntfrom)name)(\s+)\[(\d+)\]int8`)
b = convertStatvfsRegex.ReplaceAll(b, []byte("$1$3[$4]byte"))
// Convert [1024]int8 to [1024]byte in Ptmget members
convertPtmget := regexp.MustCompile(`([SC]n)(\s+)\[(\d+)\]u?int8`)
b = convertPtmget.ReplaceAll(b, []byte("$1[$3]byte"))
// Remove spare fields (e.g. in Statx_t)
spareFieldsRegex := regexp.MustCompile(`X__spare\S*`)
b = spareFieldsRegex.ReplaceAll(b, []byte("_"))
// Remove cgo padding fields
removePaddingFieldsRegex := regexp.MustCompile(`Pad_cgo_\d+`)
b = removePaddingFieldsRegex.ReplaceAll(b, []byte("_"))
// Remove padding, hidden, or unused fields
removeFieldsRegex = regexp.MustCompile(`\b(X_\S+|Padding)`)
b = removeFieldsRegex.ReplaceAll(b, []byte("_"))
// Remove the first line of warning from cgo
b = b[bytes.IndexByte(b, '\n')+1:]
// Modify the command in the header to include:
// mkpost, our own warning, and a build tag.
replacement := fmt.Sprintf(`$1 | go run mkpost.go
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s,%s`, goarch, goos)
cgoCommandRegex := regexp.MustCompile(`(cgo -godefs .*)`)
b = cgoCommandRegex.ReplaceAll(b, []byte(replacement))
// Rename Stat_t time fields
if goos == "freebsd" && goarch == "386" {
// Hide Stat_t.[AMCB]tim_ext fields
renameStatTimeExtFieldsRegex := regexp.MustCompile(`[AMCB]tim_ext`)
b = renameStatTimeExtFieldsRegex.ReplaceAll(b, []byte("_"))
}
renameStatTimeFieldsRegex := regexp.MustCompile(`([AMCB])(?:irth)?time?(?:spec)?\s+(Timespec|StTimespec)`)
b = renameStatTimeFieldsRegex.ReplaceAll(b, []byte("${1}tim ${2}"))
// gofmt
b, err = format.Source(b)
if err != nil {
log.Fatal(err)
}
os.Stdout.Write(b)
}

View File

@@ -1,402 +0,0 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
This program reads a file containing function prototypes
(like syscall_darwin.go) and generates system call bodies.
The prototypes are marked by lines beginning with "//sys"
and read like func declarations if //sys is replaced by func, but:
* The parameter lists must give a name for each argument.
This includes return parameters.
* The parameter lists must give a type for each argument:
the (x, y, z int) shorthand is not allowed.
* If the return parameter is an error number, it must be named errno.
A line beginning with //sysnb is like //sys, except that the
goroutine will not be suspended during the execution of the system
call. This must only be used for system calls which can never
block, as otherwise the system call could cause all goroutines to
hang.
*/
package main
import (
"bufio"
"flag"
"fmt"
"os"
"regexp"
"strings"
)
var (
b32 = flag.Bool("b32", false, "32bit big-endian")
l32 = flag.Bool("l32", false, "32bit little-endian")
plan9 = flag.Bool("plan9", false, "plan9")
openbsd = flag.Bool("openbsd", false, "openbsd")
netbsd = flag.Bool("netbsd", false, "netbsd")
dragonfly = flag.Bool("dragonfly", false, "dragonfly")
arm = flag.Bool("arm", false, "arm") // 64-bit value should use (even, odd)-pair
tags = flag.String("tags", "", "build tags")
filename = flag.String("output", "", "output file name (standard output if omitted)")
)
// cmdLine returns this programs's commandline arguments
func cmdLine() string {
return "go run mksyscall.go " + strings.Join(os.Args[1:], " ")
}
// buildTags returns build tags
func buildTags() string {
return *tags
}
// Param is function parameter
type Param struct {
Name string
Type string
}
// usage prints the program usage
func usage() {
fmt.Fprintf(os.Stderr, "usage: go run mksyscall.go [-b32 | -l32] [-tags x,y] [file ...]\n")
os.Exit(1)
}
// parseParamList parses parameter list and returns a slice of parameters
func parseParamList(list string) []string {
list = strings.TrimSpace(list)
if list == "" {
return []string{}
}
return regexp.MustCompile(`\s*,\s*`).Split(list, -1)
}
// parseParam splits a parameter into name and type
func parseParam(p string) Param {
ps := regexp.MustCompile(`^(\S*) (\S*)$`).FindStringSubmatch(p)
if ps == nil {
fmt.Fprintf(os.Stderr, "malformed parameter: %s\n", p)
os.Exit(1)
}
return Param{ps[1], ps[2]}
}
func main() {
// Get the OS and architecture (using GOARCH_TARGET if it exists)
goos := os.Getenv("GOOS")
if goos == "" {
fmt.Fprintln(os.Stderr, "GOOS not defined in environment")
os.Exit(1)
}
goarch := os.Getenv("GOARCH_TARGET")
if goarch == "" {
goarch = os.Getenv("GOARCH")
}
// Check that we are using the Docker-based build system if we should
if goos == "linux" {
if os.Getenv("GOLANG_SYS_BUILD") != "docker" {
fmt.Fprintf(os.Stderr, "In the Docker-based build system, mksyscall should not be called directly.\n")
fmt.Fprintf(os.Stderr, "See README.md\n")
os.Exit(1)
}
}
flag.Usage = usage
flag.Parse()
if len(flag.Args()) <= 0 {
fmt.Fprintf(os.Stderr, "no files to parse provided\n")
usage()
}
endianness := ""
if *b32 {
endianness = "big-endian"
} else if *l32 {
endianness = "little-endian"
}
libc := false
if goos == "darwin" && (strings.Contains(buildTags(), ",go1.12") || strings.Contains(buildTags(), ",go1.13")) {
libc = true
}
trampolines := map[string]bool{}
text := ""
for _, path := range flag.Args() {
file, err := os.Open(path)
if err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
s := bufio.NewScanner(file)
for s.Scan() {
t := s.Text()
t = strings.TrimSpace(t)
t = regexp.MustCompile(`\s+`).ReplaceAllString(t, ` `)
nonblock := regexp.MustCompile(`^\/\/sysnb `).FindStringSubmatch(t)
if regexp.MustCompile(`^\/\/sys `).FindStringSubmatch(t) == nil && nonblock == nil {
continue
}
// Line must be of the form
// func Open(path string, mode int, perm int) (fd int, errno error)
// Split into name, in params, out params.
f := regexp.MustCompile(`^\/\/sys(nb)? (\w+)\(([^()]*)\)\s*(?:\(([^()]+)\))?\s*(?:=\s*((?i)SYS_[A-Z0-9_]+))?$`).FindStringSubmatch(t)
if f == nil {
fmt.Fprintf(os.Stderr, "%s:%s\nmalformed //sys declaration\n", path, t)
os.Exit(1)
}
funct, inps, outps, sysname := f[2], f[3], f[4], f[5]
// ClockGettime doesn't have a syscall number on Darwin, only generate libc wrappers.
if goos == "darwin" && !libc && funct == "ClockGettime" {
continue
}
// Split argument lists on comma.
in := parseParamList(inps)
out := parseParamList(outps)
// Try in vain to keep people from editing this file.
// The theory is that they jump into the middle of the file
// without reading the header.
text += "// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT\n\n"
// Go function header.
outDecl := ""
if len(out) > 0 {
outDecl = fmt.Sprintf(" (%s)", strings.Join(out, ", "))
}
text += fmt.Sprintf("func %s(%s)%s {\n", funct, strings.Join(in, ", "), outDecl)
// Check if err return available
errvar := ""
for _, param := range out {
p := parseParam(param)
if p.Type == "error" {
errvar = p.Name
break
}
}
// Prepare arguments to Syscall.
var args []string
n := 0
for _, param := range in {
p := parseParam(param)
if regexp.MustCompile(`^\*`).FindStringSubmatch(p.Type) != nil {
args = append(args, "uintptr(unsafe.Pointer("+p.Name+"))")
} else if p.Type == "string" && errvar != "" {
text += fmt.Sprintf("\tvar _p%d *byte\n", n)
text += fmt.Sprintf("\t_p%d, %s = BytePtrFromString(%s)\n", n, errvar, p.Name)
text += fmt.Sprintf("\tif %s != nil {\n\t\treturn\n\t}\n", errvar)
args = append(args, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n))
n++
} else if p.Type == "string" {
fmt.Fprintf(os.Stderr, path+":"+funct+" uses string arguments, but has no error return\n")
text += fmt.Sprintf("\tvar _p%d *byte\n", n)
text += fmt.Sprintf("\t_p%d, _ = BytePtrFromString(%s)\n", n, p.Name)
args = append(args, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n))
n++
} else if regexp.MustCompile(`^\[\](.*)`).FindStringSubmatch(p.Type) != nil {
// Convert slice into pointer, length.
// Have to be careful not to take address of &a[0] if len == 0:
// pass dummy pointer in that case.
// Used to pass nil, but some OSes or simulators reject write(fd, nil, 0).
text += fmt.Sprintf("\tvar _p%d unsafe.Pointer\n", n)
text += fmt.Sprintf("\tif len(%s) > 0 {\n\t\t_p%d = unsafe.Pointer(&%s[0])\n\t}", p.Name, n, p.Name)
text += fmt.Sprintf(" else {\n\t\t_p%d = unsafe.Pointer(&_zero)\n\t}\n", n)
args = append(args, fmt.Sprintf("uintptr(_p%d)", n), fmt.Sprintf("uintptr(len(%s))", p.Name))
n++
} else if p.Type == "int64" && (*openbsd || *netbsd) {
args = append(args, "0")
if endianness == "big-endian" {
args = append(args, fmt.Sprintf("uintptr(%s>>32)", p.Name), fmt.Sprintf("uintptr(%s)", p.Name))
} else if endianness == "little-endian" {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name), fmt.Sprintf("uintptr(%s>>32)", p.Name))
} else {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name))
}
} else if p.Type == "int64" && *dragonfly {
if regexp.MustCompile(`^(?i)extp(read|write)`).FindStringSubmatch(funct) == nil {
args = append(args, "0")
}
if endianness == "big-endian" {
args = append(args, fmt.Sprintf("uintptr(%s>>32)", p.Name), fmt.Sprintf("uintptr(%s)", p.Name))
} else if endianness == "little-endian" {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name), fmt.Sprintf("uintptr(%s>>32)", p.Name))
} else {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name))
}
} else if (p.Type == "int64" || p.Type == "uint64") && endianness != "" {
if len(args)%2 == 1 && *arm {
// arm abi specifies 64-bit argument uses
// (even, odd) pair
args = append(args, "0")
}
if endianness == "big-endian" {
args = append(args, fmt.Sprintf("uintptr(%s>>32)", p.Name), fmt.Sprintf("uintptr(%s)", p.Name))
} else {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name), fmt.Sprintf("uintptr(%s>>32)", p.Name))
}
} else {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name))
}
}
// Determine which form to use; pad args with zeros.
asm := "Syscall"
if nonblock != nil {
if errvar == "" && goos == "linux" {
asm = "RawSyscallNoError"
} else {
asm = "RawSyscall"
}
} else {
if errvar == "" && goos == "linux" {
asm = "SyscallNoError"
}
}
if len(args) <= 3 {
for len(args) < 3 {
args = append(args, "0")
}
} else if len(args) <= 6 {
asm += "6"
for len(args) < 6 {
args = append(args, "0")
}
} else if len(args) <= 9 {
asm += "9"
for len(args) < 9 {
args = append(args, "0")
}
} else {
fmt.Fprintf(os.Stderr, "%s:%s too many arguments to system call\n", path, funct)
}
// System call number.
if sysname == "" {
sysname = "SYS_" + funct
sysname = regexp.MustCompile(`([a-z])([A-Z])`).ReplaceAllString(sysname, `${1}_$2`)
sysname = strings.ToUpper(sysname)
}
var libcFn string
if libc {
asm = "syscall_" + strings.ToLower(asm[:1]) + asm[1:] // internal syscall call
sysname = strings.TrimPrefix(sysname, "SYS_") // remove SYS_
sysname = strings.ToLower(sysname) // lowercase
libcFn = sysname
sysname = "funcPC(libc_" + sysname + "_trampoline)"
}
// Actual call.
arglist := strings.Join(args, ", ")
call := fmt.Sprintf("%s(%s, %s)", asm, sysname, arglist)
// Assign return values.
body := ""
ret := []string{"_", "_", "_"}
doErrno := false
for i := 0; i < len(out); i++ {
p := parseParam(out[i])
reg := ""
if p.Name == "err" && !*plan9 {
reg = "e1"
ret[2] = reg
doErrno = true
} else if p.Name == "err" && *plan9 {
ret[0] = "r0"
ret[2] = "e1"
break
} else {
reg = fmt.Sprintf("r%d", i)
ret[i] = reg
}
if p.Type == "bool" {
reg = fmt.Sprintf("%s != 0", reg)
}
if p.Type == "int64" && endianness != "" {
// 64-bit number in r1:r0 or r0:r1.
if i+2 > len(out) {
fmt.Fprintf(os.Stderr, "%s:%s not enough registers for int64 return\n", path, funct)
}
if endianness == "big-endian" {
reg = fmt.Sprintf("int64(r%d)<<32 | int64(r%d)", i, i+1)
} else {
reg = fmt.Sprintf("int64(r%d)<<32 | int64(r%d)", i+1, i)
}
ret[i] = fmt.Sprintf("r%d", i)
ret[i+1] = fmt.Sprintf("r%d", i+1)
}
if reg != "e1" || *plan9 {
body += fmt.Sprintf("\t%s = %s(%s)\n", p.Name, p.Type, reg)
}
}
if ret[0] == "_" && ret[1] == "_" && ret[2] == "_" {
text += fmt.Sprintf("\t%s\n", call)
} else {
if errvar == "" && goos == "linux" {
// raw syscall without error on Linux, see golang.org/issue/22924
text += fmt.Sprintf("\t%s, %s := %s\n", ret[0], ret[1], call)
} else {
text += fmt.Sprintf("\t%s, %s, %s := %s\n", ret[0], ret[1], ret[2], call)
}
}
text += body
if *plan9 && ret[2] == "e1" {
text += "\tif int32(r0) == -1 {\n"
text += "\t\terr = e1\n"
text += "\t}\n"
} else if doErrno {
text += "\tif e1 != 0 {\n"
text += "\t\terr = errnoErr(e1)\n"
text += "\t}\n"
}
text += "\treturn\n"
text += "}\n\n"
if libc && !trampolines[libcFn] {
// some system calls share a trampoline, like read and readlen.
trampolines[libcFn] = true
// Declare assembly trampoline.
text += fmt.Sprintf("func libc_%s_trampoline()\n", libcFn)
// Assembly trampoline calls the libc_* function, which this magic
// redirects to use the function from libSystem.
text += fmt.Sprintf("//go:linkname libc_%s libc_%s\n", libcFn, libcFn)
text += fmt.Sprintf("//go:cgo_import_dynamic libc_%s %s \"/usr/lib/libSystem.B.dylib\"\n", libcFn, libcFn)
text += "\n"
}
}
if err := s.Err(); err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
file.Close()
}
fmt.Printf(srcTemplate, cmdLine(), buildTags(), text)
}
const srcTemplate = `// %s
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s
package unix
import (
"syscall"
"unsafe"
)
var _ syscall.Errno
%s
`

View File

@@ -1,415 +0,0 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
This program reads a file containing function prototypes
(like syscall_aix.go) and generates system call bodies.
The prototypes are marked by lines beginning with "//sys"
and read like func declarations if //sys is replaced by func, but:
* The parameter lists must give a name for each argument.
This includes return parameters.
* The parameter lists must give a type for each argument:
the (x, y, z int) shorthand is not allowed.
* If the return parameter is an error number, it must be named err.
* If go func name needs to be different than its libc name,
* or the function is not in libc, name could be specified
* at the end, after "=" sign, like
//sys getsockopt(s int, level int, name int, val uintptr, vallen *_Socklen) (err error) = libsocket.getsockopt
*/
package main
import (
"bufio"
"flag"
"fmt"
"os"
"regexp"
"strings"
)
var (
b32 = flag.Bool("b32", false, "32bit big-endian")
l32 = flag.Bool("l32", false, "32bit little-endian")
aix = flag.Bool("aix", false, "aix")
tags = flag.String("tags", "", "build tags")
)
// cmdLine returns this programs's commandline arguments
func cmdLine() string {
return "go run mksyscall_aix_ppc.go " + strings.Join(os.Args[1:], " ")
}
// buildTags returns build tags
func buildTags() string {
return *tags
}
// Param is function parameter
type Param struct {
Name string
Type string
}
// usage prints the program usage
func usage() {
fmt.Fprintf(os.Stderr, "usage: go run mksyscall_aix_ppc.go [-b32 | -l32] [-tags x,y] [file ...]\n")
os.Exit(1)
}
// parseParamList parses parameter list and returns a slice of parameters
func parseParamList(list string) []string {
list = strings.TrimSpace(list)
if list == "" {
return []string{}
}
return regexp.MustCompile(`\s*,\s*`).Split(list, -1)
}
// parseParam splits a parameter into name and type
func parseParam(p string) Param {
ps := regexp.MustCompile(`^(\S*) (\S*)$`).FindStringSubmatch(p)
if ps == nil {
fmt.Fprintf(os.Stderr, "malformed parameter: %s\n", p)
os.Exit(1)
}
return Param{ps[1], ps[2]}
}
func main() {
flag.Usage = usage
flag.Parse()
if len(flag.Args()) <= 0 {
fmt.Fprintf(os.Stderr, "no files to parse provided\n")
usage()
}
endianness := ""
if *b32 {
endianness = "big-endian"
} else if *l32 {
endianness = "little-endian"
}
pack := ""
text := ""
cExtern := "/*\n#include <stdint.h>\n#include <stddef.h>\n"
for _, path := range flag.Args() {
file, err := os.Open(path)
if err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
s := bufio.NewScanner(file)
for s.Scan() {
t := s.Text()
t = strings.TrimSpace(t)
t = regexp.MustCompile(`\s+`).ReplaceAllString(t, ` `)
if p := regexp.MustCompile(`^package (\S+)$`).FindStringSubmatch(t); p != nil && pack == "" {
pack = p[1]
}
nonblock := regexp.MustCompile(`^\/\/sysnb `).FindStringSubmatch(t)
if regexp.MustCompile(`^\/\/sys `).FindStringSubmatch(t) == nil && nonblock == nil {
continue
}
// Line must be of the form
// func Open(path string, mode int, perm int) (fd int, err error)
// Split into name, in params, out params.
f := regexp.MustCompile(`^\/\/sys(nb)? (\w+)\(([^()]*)\)\s*(?:\(([^()]+)\))?\s*(?:=\s*(?:(\w*)\.)?(\w*))?$`).FindStringSubmatch(t)
if f == nil {
fmt.Fprintf(os.Stderr, "%s:%s\nmalformed //sys declaration\n", path, t)
os.Exit(1)
}
funct, inps, outps, modname, sysname := f[2], f[3], f[4], f[5], f[6]
// Split argument lists on comma.
in := parseParamList(inps)
out := parseParamList(outps)
inps = strings.Join(in, ", ")
outps = strings.Join(out, ", ")
// Try in vain to keep people from editing this file.
// The theory is that they jump into the middle of the file
// without reading the header.
text += "// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT\n\n"
// Check if value return, err return available
errvar := ""
retvar := ""
rettype := ""
for _, param := range out {
p := parseParam(param)
if p.Type == "error" {
errvar = p.Name
} else {
retvar = p.Name
rettype = p.Type
}
}
// System call name.
if sysname == "" {
sysname = funct
}
sysname = regexp.MustCompile(`([a-z])([A-Z])`).ReplaceAllString(sysname, `${1}_$2`)
sysname = strings.ToLower(sysname) // All libc functions are lowercase.
cRettype := ""
if rettype == "unsafe.Pointer" {
cRettype = "uintptr_t"
} else if rettype == "uintptr" {
cRettype = "uintptr_t"
} else if regexp.MustCompile(`^_`).FindStringSubmatch(rettype) != nil {
cRettype = "uintptr_t"
} else if rettype == "int" {
cRettype = "int"
} else if rettype == "int32" {
cRettype = "int"
} else if rettype == "int64" {
cRettype = "long long"
} else if rettype == "uint32" {
cRettype = "unsigned int"
} else if rettype == "uint64" {
cRettype = "unsigned long long"
} else {
cRettype = "int"
}
if sysname == "exit" {
cRettype = "void"
}
// Change p.Types to c
var cIn []string
for _, param := range in {
p := parseParam(param)
if regexp.MustCompile(`^\*`).FindStringSubmatch(p.Type) != nil {
cIn = append(cIn, "uintptr_t")
} else if p.Type == "string" {
cIn = append(cIn, "uintptr_t")
} else if regexp.MustCompile(`^\[\](.*)`).FindStringSubmatch(p.Type) != nil {
cIn = append(cIn, "uintptr_t", "size_t")
} else if p.Type == "unsafe.Pointer" {
cIn = append(cIn, "uintptr_t")
} else if p.Type == "uintptr" {
cIn = append(cIn, "uintptr_t")
} else if regexp.MustCompile(`^_`).FindStringSubmatch(p.Type) != nil {
cIn = append(cIn, "uintptr_t")
} else if p.Type == "int" {
cIn = append(cIn, "int")
} else if p.Type == "int32" {
cIn = append(cIn, "int")
} else if p.Type == "int64" {
cIn = append(cIn, "long long")
} else if p.Type == "uint32" {
cIn = append(cIn, "unsigned int")
} else if p.Type == "uint64" {
cIn = append(cIn, "unsigned long long")
} else {
cIn = append(cIn, "int")
}
}
if funct != "fcntl" && funct != "FcntlInt" && funct != "readlen" && funct != "writelen" {
if sysname == "select" {
// select is a keyword of Go. Its name is
// changed to c_select.
cExtern += "#define c_select select\n"
}
// Imports of system calls from libc
cExtern += fmt.Sprintf("%s %s", cRettype, sysname)
cIn := strings.Join(cIn, ", ")
cExtern += fmt.Sprintf("(%s);\n", cIn)
}
// So file name.
if *aix {
if modname == "" {
modname = "libc.a/shr_64.o"
} else {
fmt.Fprintf(os.Stderr, "%s: only syscall using libc are available\n", funct)
os.Exit(1)
}
}
strconvfunc := "C.CString"
// Go function header.
if outps != "" {
outps = fmt.Sprintf(" (%s)", outps)
}
if text != "" {
text += "\n"
}
text += fmt.Sprintf("func %s(%s)%s {\n", funct, strings.Join(in, ", "), outps)
// Prepare arguments to Syscall.
var args []string
n := 0
argN := 0
for _, param := range in {
p := parseParam(param)
if regexp.MustCompile(`^\*`).FindStringSubmatch(p.Type) != nil {
args = append(args, "C.uintptr_t(uintptr(unsafe.Pointer("+p.Name+")))")
} else if p.Type == "string" && errvar != "" {
text += fmt.Sprintf("\t_p%d := uintptr(unsafe.Pointer(%s(%s)))\n", n, strconvfunc, p.Name)
args = append(args, fmt.Sprintf("C.uintptr_t(_p%d)", n))
n++
} else if p.Type == "string" {
fmt.Fprintf(os.Stderr, path+":"+funct+" uses string arguments, but has no error return\n")
text += fmt.Sprintf("\t_p%d := uintptr(unsafe.Pointer(%s(%s)))\n", n, strconvfunc, p.Name)
args = append(args, fmt.Sprintf("C.uintptr_t(_p%d)", n))
n++
} else if m := regexp.MustCompile(`^\[\](.*)`).FindStringSubmatch(p.Type); m != nil {
// Convert slice into pointer, length.
// Have to be careful not to take address of &a[0] if len == 0:
// pass nil in that case.
text += fmt.Sprintf("\tvar _p%d *%s\n", n, m[1])
text += fmt.Sprintf("\tif len(%s) > 0 {\n\t\t_p%d = &%s[0]\n\t}\n", p.Name, n, p.Name)
args = append(args, fmt.Sprintf("C.uintptr_t(uintptr(unsafe.Pointer(_p%d)))", n))
n++
text += fmt.Sprintf("\tvar _p%d int\n", n)
text += fmt.Sprintf("\t_p%d = len(%s)\n", n, p.Name)
args = append(args, fmt.Sprintf("C.size_t(_p%d)", n))
n++
} else if p.Type == "int64" && endianness != "" {
if endianness == "big-endian" {
args = append(args, fmt.Sprintf("uintptr(%s>>32)", p.Name), fmt.Sprintf("uintptr(%s)", p.Name))
} else {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name), fmt.Sprintf("uintptr(%s>>32)", p.Name))
}
n++
} else if p.Type == "bool" {
text += fmt.Sprintf("\tvar _p%d uint32\n", n)
text += fmt.Sprintf("\tif %s {\n\t\t_p%d = 1\n\t} else {\n\t\t_p%d = 0\n\t}\n", p.Name, n, n)
args = append(args, fmt.Sprintf("_p%d", n))
} else if regexp.MustCompile(`^_`).FindStringSubmatch(p.Type) != nil {
args = append(args, fmt.Sprintf("C.uintptr_t(uintptr(%s))", p.Name))
} else if p.Type == "unsafe.Pointer" {
args = append(args, fmt.Sprintf("C.uintptr_t(uintptr(%s))", p.Name))
} else if p.Type == "int" {
if (argN == 2) && ((funct == "readlen") || (funct == "writelen")) {
args = append(args, fmt.Sprintf("C.size_t(%s)", p.Name))
} else if argN == 0 && funct == "fcntl" {
args = append(args, fmt.Sprintf("C.uintptr_t(%s)", p.Name))
} else if (argN == 2) && ((funct == "fcntl") || (funct == "FcntlInt")) {
args = append(args, fmt.Sprintf("C.uintptr_t(%s)", p.Name))
} else {
args = append(args, fmt.Sprintf("C.int(%s)", p.Name))
}
} else if p.Type == "int32" {
args = append(args, fmt.Sprintf("C.int(%s)", p.Name))
} else if p.Type == "int64" {
args = append(args, fmt.Sprintf("C.longlong(%s)", p.Name))
} else if p.Type == "uint32" {
args = append(args, fmt.Sprintf("C.uint(%s)", p.Name))
} else if p.Type == "uint64" {
args = append(args, fmt.Sprintf("C.ulonglong(%s)", p.Name))
} else if p.Type == "uintptr" {
args = append(args, fmt.Sprintf("C.uintptr_t(%s)", p.Name))
} else {
args = append(args, fmt.Sprintf("C.int(%s)", p.Name))
}
argN++
}
// Actual call.
arglist := strings.Join(args, ", ")
call := ""
if sysname == "exit" {
if errvar != "" {
call += "er :="
} else {
call += ""
}
} else if errvar != "" {
call += "r0,er :="
} else if retvar != "" {
call += "r0,_ :="
} else {
call += ""
}
if sysname == "select" {
// select is a keyword of Go. Its name is
// changed to c_select.
call += fmt.Sprintf("C.c_%s(%s)", sysname, arglist)
} else {
call += fmt.Sprintf("C.%s(%s)", sysname, arglist)
}
// Assign return values.
body := ""
for i := 0; i < len(out); i++ {
p := parseParam(out[i])
reg := ""
if p.Name == "err" {
reg = "e1"
} else {
reg = "r0"
}
if reg != "e1" {
body += fmt.Sprintf("\t%s = %s(%s)\n", p.Name, p.Type, reg)
}
}
// verify return
if sysname != "exit" && errvar != "" {
if regexp.MustCompile(`^uintptr`).FindStringSubmatch(cRettype) != nil {
body += "\tif (uintptr(r0) ==^uintptr(0) && er != nil) {\n"
body += fmt.Sprintf("\t\t%s = er\n", errvar)
body += "\t}\n"
} else {
body += "\tif (r0 ==-1 && er != nil) {\n"
body += fmt.Sprintf("\t\t%s = er\n", errvar)
body += "\t}\n"
}
} else if errvar != "" {
body += "\tif (er != nil) {\n"
body += fmt.Sprintf("\t\t%s = er\n", errvar)
body += "\t}\n"
}
text += fmt.Sprintf("\t%s\n", call)
text += body
text += "\treturn\n"
text += "}\n"
}
if err := s.Err(); err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
file.Close()
}
imp := ""
if pack != "unix" {
imp = "import \"golang.org/x/sys/unix\"\n"
}
fmt.Printf(srcTemplate, cmdLine(), buildTags(), pack, cExtern, imp, text)
}
const srcTemplate = `// %s
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s
package %s
%s
*/
import "C"
import (
"unsafe"
)
%s
%s
`

View File

@@ -1,614 +0,0 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
This program reads a file containing function prototypes
(like syscall_aix.go) and generates system call bodies.
The prototypes are marked by lines beginning with "//sys"
and read like func declarations if //sys is replaced by func, but:
* The parameter lists must give a name for each argument.
This includes return parameters.
* The parameter lists must give a type for each argument:
the (x, y, z int) shorthand is not allowed.
* If the return parameter is an error number, it must be named err.
* If go func name needs to be different than its libc name,
* or the function is not in libc, name could be specified
* at the end, after "=" sign, like
//sys getsockopt(s int, level int, name int, val uintptr, vallen *_Socklen) (err error) = libsocket.getsockopt
This program will generate three files and handle both gc and gccgo implementation:
- zsyscall_aix_ppc64.go: the common part of each implementation (error handler, pointer creation)
- zsyscall_aix_ppc64_gc.go: gc part with //go_cgo_import_dynamic and a call to syscall6
- zsyscall_aix_ppc64_gccgo.go: gccgo part with C function and conversion to C type.
The generated code looks like this
zsyscall_aix_ppc64.go
func asyscall(...) (n int, err error) {
// Pointer Creation
r1, e1 := callasyscall(...)
// Type Conversion
// Error Handler
return
}
zsyscall_aix_ppc64_gc.go
//go:cgo_import_dynamic libc_asyscall asyscall "libc.a/shr_64.o"
//go:linkname libc_asyscall libc_asyscall
var asyscall syscallFunc
func callasyscall(...) (r1 uintptr, e1 Errno) {
r1, _, e1 = syscall6(uintptr(unsafe.Pointer(&libc_asyscall)), "nb_args", ... )
return
}
zsyscall_aix_ppc64_ggcgo.go
// int asyscall(...)
import "C"
func callasyscall(...) (r1 uintptr, e1 Errno) {
r1 = uintptr(C.asyscall(...))
e1 = syscall.GetErrno()
return
}
*/
package main
import (
"bufio"
"flag"
"fmt"
"io/ioutil"
"os"
"regexp"
"strings"
)
var (
b32 = flag.Bool("b32", false, "32bit big-endian")
l32 = flag.Bool("l32", false, "32bit little-endian")
aix = flag.Bool("aix", false, "aix")
tags = flag.String("tags", "", "build tags")
)
// cmdLine returns this programs's commandline arguments
func cmdLine() string {
return "go run mksyscall_aix_ppc64.go " + strings.Join(os.Args[1:], " ")
}
// buildTags returns build tags
func buildTags() string {
return *tags
}
// Param is function parameter
type Param struct {
Name string
Type string
}
// usage prints the program usage
func usage() {
fmt.Fprintf(os.Stderr, "usage: go run mksyscall_aix_ppc64.go [-b32 | -l32] [-tags x,y] [file ...]\n")
os.Exit(1)
}
// parseParamList parses parameter list and returns a slice of parameters
func parseParamList(list string) []string {
list = strings.TrimSpace(list)
if list == "" {
return []string{}
}
return regexp.MustCompile(`\s*,\s*`).Split(list, -1)
}
// parseParam splits a parameter into name and type
func parseParam(p string) Param {
ps := regexp.MustCompile(`^(\S*) (\S*)$`).FindStringSubmatch(p)
if ps == nil {
fmt.Fprintf(os.Stderr, "malformed parameter: %s\n", p)
os.Exit(1)
}
return Param{ps[1], ps[2]}
}
func main() {
flag.Usage = usage
flag.Parse()
if len(flag.Args()) <= 0 {
fmt.Fprintf(os.Stderr, "no files to parse provided\n")
usage()
}
endianness := ""
if *b32 {
endianness = "big-endian"
} else if *l32 {
endianness = "little-endian"
}
pack := ""
// GCCGO
textgccgo := ""
cExtern := "/*\n#include <stdint.h>\n"
// GC
textgc := ""
dynimports := ""
linknames := ""
var vars []string
// COMMON
textcommon := ""
for _, path := range flag.Args() {
file, err := os.Open(path)
if err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
s := bufio.NewScanner(file)
for s.Scan() {
t := s.Text()
t = strings.TrimSpace(t)
t = regexp.MustCompile(`\s+`).ReplaceAllString(t, ` `)
if p := regexp.MustCompile(`^package (\S+)$`).FindStringSubmatch(t); p != nil && pack == "" {
pack = p[1]
}
nonblock := regexp.MustCompile(`^\/\/sysnb `).FindStringSubmatch(t)
if regexp.MustCompile(`^\/\/sys `).FindStringSubmatch(t) == nil && nonblock == nil {
continue
}
// Line must be of the form
// func Open(path string, mode int, perm int) (fd int, err error)
// Split into name, in params, out params.
f := regexp.MustCompile(`^\/\/sys(nb)? (\w+)\(([^()]*)\)\s*(?:\(([^()]+)\))?\s*(?:=\s*(?:(\w*)\.)?(\w*))?$`).FindStringSubmatch(t)
if f == nil {
fmt.Fprintf(os.Stderr, "%s:%s\nmalformed //sys declaration\n", path, t)
os.Exit(1)
}
funct, inps, outps, modname, sysname := f[2], f[3], f[4], f[5], f[6]
// Split argument lists on comma.
in := parseParamList(inps)
out := parseParamList(outps)
inps = strings.Join(in, ", ")
outps = strings.Join(out, ", ")
if sysname == "" {
sysname = funct
}
onlyCommon := false
if funct == "readlen" || funct == "writelen" || funct == "FcntlInt" || funct == "FcntlFlock" {
// This function call another syscall which is already implemented.
// Therefore, the gc and gccgo part must not be generated.
onlyCommon = true
}
// Try in vain to keep people from editing this file.
// The theory is that they jump into the middle of the file
// without reading the header.
textcommon += "// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT\n\n"
if !onlyCommon {
textgccgo += "// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT\n\n"
textgc += "// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT\n\n"
}
// Check if value return, err return available
errvar := ""
rettype := ""
for _, param := range out {
p := parseParam(param)
if p.Type == "error" {
errvar = p.Name
} else {
rettype = p.Type
}
}
sysname = regexp.MustCompile(`([a-z])([A-Z])`).ReplaceAllString(sysname, `${1}_$2`)
sysname = strings.ToLower(sysname) // All libc functions are lowercase.
// GCCGO Prototype return type
cRettype := ""
if rettype == "unsafe.Pointer" {
cRettype = "uintptr_t"
} else if rettype == "uintptr" {
cRettype = "uintptr_t"
} else if regexp.MustCompile(`^_`).FindStringSubmatch(rettype) != nil {
cRettype = "uintptr_t"
} else if rettype == "int" {
cRettype = "int"
} else if rettype == "int32" {
cRettype = "int"
} else if rettype == "int64" {
cRettype = "long long"
} else if rettype == "uint32" {
cRettype = "unsigned int"
} else if rettype == "uint64" {
cRettype = "unsigned long long"
} else {
cRettype = "int"
}
if sysname == "exit" {
cRettype = "void"
}
// GCCGO Prototype arguments type
var cIn []string
for i, param := range in {
p := parseParam(param)
if regexp.MustCompile(`^\*`).FindStringSubmatch(p.Type) != nil {
cIn = append(cIn, "uintptr_t")
} else if p.Type == "string" {
cIn = append(cIn, "uintptr_t")
} else if regexp.MustCompile(`^\[\](.*)`).FindStringSubmatch(p.Type) != nil {
cIn = append(cIn, "uintptr_t", "size_t")
} else if p.Type == "unsafe.Pointer" {
cIn = append(cIn, "uintptr_t")
} else if p.Type == "uintptr" {
cIn = append(cIn, "uintptr_t")
} else if regexp.MustCompile(`^_`).FindStringSubmatch(p.Type) != nil {
cIn = append(cIn, "uintptr_t")
} else if p.Type == "int" {
if (i == 0 || i == 2) && funct == "fcntl" {
// These fcntl arguments needs to be uintptr to be able to call FcntlInt and FcntlFlock
cIn = append(cIn, "uintptr_t")
} else {
cIn = append(cIn, "int")
}
} else if p.Type == "int32" {
cIn = append(cIn, "int")
} else if p.Type == "int64" {
cIn = append(cIn, "long long")
} else if p.Type == "uint32" {
cIn = append(cIn, "unsigned int")
} else if p.Type == "uint64" {
cIn = append(cIn, "unsigned long long")
} else {
cIn = append(cIn, "int")
}
}
if !onlyCommon {
// GCCGO Prototype Generation
// Imports of system calls from libc
if sysname == "select" {
// select is a keyword of Go. Its name is
// changed to c_select.
cExtern += "#define c_select select\n"
}
cExtern += fmt.Sprintf("%s %s", cRettype, sysname)
cIn := strings.Join(cIn, ", ")
cExtern += fmt.Sprintf("(%s);\n", cIn)
}
// GC Library name
if modname == "" {
modname = "libc.a/shr_64.o"
} else {
fmt.Fprintf(os.Stderr, "%s: only syscall using libc are available\n", funct)
os.Exit(1)
}
sysvarname := fmt.Sprintf("libc_%s", sysname)
if !onlyCommon {
// GC Runtime import of function to allow cross-platform builds.
dynimports += fmt.Sprintf("//go:cgo_import_dynamic %s %s \"%s\"\n", sysvarname, sysname, modname)
// GC Link symbol to proc address variable.
linknames += fmt.Sprintf("//go:linkname %s %s\n", sysvarname, sysvarname)
// GC Library proc address variable.
vars = append(vars, sysvarname)
}
strconvfunc := "BytePtrFromString"
strconvtype := "*byte"
// Go function header.
if outps != "" {
outps = fmt.Sprintf(" (%s)", outps)
}
if textcommon != "" {
textcommon += "\n"
}
textcommon += fmt.Sprintf("func %s(%s)%s {\n", funct, strings.Join(in, ", "), outps)
// Prepare arguments tocall.
var argscommon []string // Arguments in the common part
var argscall []string // Arguments for call prototype
var argsgc []string // Arguments for gc call (with syscall6)
var argsgccgo []string // Arguments for gccgo call (with C.name_of_syscall)
n := 0
argN := 0
for _, param := range in {
p := parseParam(param)
if regexp.MustCompile(`^\*`).FindStringSubmatch(p.Type) != nil {
argscommon = append(argscommon, fmt.Sprintf("uintptr(unsafe.Pointer(%s))", p.Name))
argscall = append(argscall, fmt.Sprintf("%s uintptr", p.Name))
argsgc = append(argsgc, p.Name)
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uintptr_t(%s)", p.Name))
} else if p.Type == "string" && errvar != "" {
textcommon += fmt.Sprintf("\tvar _p%d %s\n", n, strconvtype)
textcommon += fmt.Sprintf("\t_p%d, %s = %s(%s)\n", n, errvar, strconvfunc, p.Name)
textcommon += fmt.Sprintf("\tif %s != nil {\n\t\treturn\n\t}\n", errvar)
argscommon = append(argscommon, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n))
argscall = append(argscall, fmt.Sprintf("_p%d uintptr ", n))
argsgc = append(argsgc, fmt.Sprintf("_p%d", n))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uintptr_t(_p%d)", n))
n++
} else if p.Type == "string" {
fmt.Fprintf(os.Stderr, path+":"+funct+" uses string arguments, but has no error return\n")
textcommon += fmt.Sprintf("\tvar _p%d %s\n", n, strconvtype)
textcommon += fmt.Sprintf("\t_p%d, %s = %s(%s)\n", n, errvar, strconvfunc, p.Name)
textcommon += fmt.Sprintf("\tif %s != nil {\n\t\treturn\n\t}\n", errvar)
argscommon = append(argscommon, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n))
argscall = append(argscall, fmt.Sprintf("_p%d uintptr", n))
argsgc = append(argsgc, fmt.Sprintf("_p%d", n))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uintptr_t(_p%d)", n))
n++
} else if m := regexp.MustCompile(`^\[\](.*)`).FindStringSubmatch(p.Type); m != nil {
// Convert slice into pointer, length.
// Have to be careful not to take address of &a[0] if len == 0:
// pass nil in that case.
textcommon += fmt.Sprintf("\tvar _p%d *%s\n", n, m[1])
textcommon += fmt.Sprintf("\tif len(%s) > 0 {\n\t\t_p%d = &%s[0]\n\t}\n", p.Name, n, p.Name)
argscommon = append(argscommon, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n), fmt.Sprintf("len(%s)", p.Name))
argscall = append(argscall, fmt.Sprintf("_p%d uintptr", n), fmt.Sprintf("_lenp%d int", n))
argsgc = append(argsgc, fmt.Sprintf("_p%d", n), fmt.Sprintf("uintptr(_lenp%d)", n))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uintptr_t(_p%d)", n), fmt.Sprintf("C.size_t(_lenp%d)", n))
n++
} else if p.Type == "int64" && endianness != "" {
fmt.Fprintf(os.Stderr, path+":"+funct+" uses int64 with 32 bits mode. Case not yet implemented\n")
} else if p.Type == "bool" {
fmt.Fprintf(os.Stderr, path+":"+funct+" uses bool. Case not yet implemented\n")
} else if regexp.MustCompile(`^_`).FindStringSubmatch(p.Type) != nil || p.Type == "unsafe.Pointer" {
argscommon = append(argscommon, fmt.Sprintf("uintptr(%s)", p.Name))
argscall = append(argscall, fmt.Sprintf("%s uintptr", p.Name))
argsgc = append(argsgc, p.Name)
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uintptr_t(%s)", p.Name))
} else if p.Type == "int" {
if (argN == 0 || argN == 2) && ((funct == "fcntl") || (funct == "FcntlInt") || (funct == "FcntlFlock")) {
// These fcntl arguments need to be uintptr to be able to call FcntlInt and FcntlFlock
argscommon = append(argscommon, fmt.Sprintf("uintptr(%s)", p.Name))
argscall = append(argscall, fmt.Sprintf("%s uintptr", p.Name))
argsgc = append(argsgc, p.Name)
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uintptr_t(%s)", p.Name))
} else {
argscommon = append(argscommon, p.Name)
argscall = append(argscall, fmt.Sprintf("%s int", p.Name))
argsgc = append(argsgc, fmt.Sprintf("uintptr(%s)", p.Name))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.int(%s)", p.Name))
}
} else if p.Type == "int32" {
argscommon = append(argscommon, p.Name)
argscall = append(argscall, fmt.Sprintf("%s int32", p.Name))
argsgc = append(argsgc, fmt.Sprintf("uintptr(%s)", p.Name))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.int(%s)", p.Name))
} else if p.Type == "int64" {
argscommon = append(argscommon, p.Name)
argscall = append(argscall, fmt.Sprintf("%s int64", p.Name))
argsgc = append(argsgc, fmt.Sprintf("uintptr(%s)", p.Name))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.longlong(%s)", p.Name))
} else if p.Type == "uint32" {
argscommon = append(argscommon, p.Name)
argscall = append(argscall, fmt.Sprintf("%s uint32", p.Name))
argsgc = append(argsgc, fmt.Sprintf("uintptr(%s)", p.Name))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uint(%s)", p.Name))
} else if p.Type == "uint64" {
argscommon = append(argscommon, p.Name)
argscall = append(argscall, fmt.Sprintf("%s uint64", p.Name))
argsgc = append(argsgc, fmt.Sprintf("uintptr(%s)", p.Name))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.ulonglong(%s)", p.Name))
} else if p.Type == "uintptr" {
argscommon = append(argscommon, p.Name)
argscall = append(argscall, fmt.Sprintf("%s uintptr", p.Name))
argsgc = append(argsgc, p.Name)
argsgccgo = append(argsgccgo, fmt.Sprintf("C.uintptr_t(%s)", p.Name))
} else {
argscommon = append(argscommon, fmt.Sprintf("int(%s)", p.Name))
argscall = append(argscall, fmt.Sprintf("%s int", p.Name))
argsgc = append(argsgc, fmt.Sprintf("uintptr(%s)", p.Name))
argsgccgo = append(argsgccgo, fmt.Sprintf("C.int(%s)", p.Name))
}
argN++
}
nargs := len(argsgc)
// COMMON function generation
argscommonlist := strings.Join(argscommon, ", ")
callcommon := fmt.Sprintf("call%s(%s)", sysname, argscommonlist)
ret := []string{"_", "_"}
body := ""
doErrno := false
for i := 0; i < len(out); i++ {
p := parseParam(out[i])
reg := ""
if p.Name == "err" {
reg = "e1"
ret[1] = reg
doErrno = true
} else {
reg = "r0"
ret[0] = reg
}
if p.Type == "bool" {
reg = fmt.Sprintf("%s != 0", reg)
}
if reg != "e1" {
body += fmt.Sprintf("\t%s = %s(%s)\n", p.Name, p.Type, reg)
}
}
if ret[0] == "_" && ret[1] == "_" {
textcommon += fmt.Sprintf("\t%s\n", callcommon)
} else {
textcommon += fmt.Sprintf("\t%s, %s := %s\n", ret[0], ret[1], callcommon)
}
textcommon += body
if doErrno {
textcommon += "\tif e1 != 0 {\n"
textcommon += "\t\terr = errnoErr(e1)\n"
textcommon += "\t}\n"
}
textcommon += "\treturn\n"
textcommon += "}\n"
if onlyCommon {
continue
}
// CALL Prototype
callProto := fmt.Sprintf("func call%s(%s) (r1 uintptr, e1 Errno) {\n", sysname, strings.Join(argscall, ", "))
// GC function generation
asm := "syscall6"
if nonblock != nil {
asm = "rawSyscall6"
}
if len(argsgc) <= 6 {
for len(argsgc) < 6 {
argsgc = append(argsgc, "0")
}
} else {
fmt.Fprintf(os.Stderr, "%s: too many arguments to system call", funct)
os.Exit(1)
}
argsgclist := strings.Join(argsgc, ", ")
callgc := fmt.Sprintf("%s(uintptr(unsafe.Pointer(&%s)), %d, %s)", asm, sysvarname, nargs, argsgclist)
textgc += callProto
textgc += fmt.Sprintf("\tr1, _, e1 = %s\n", callgc)
textgc += "\treturn\n}\n"
// GCCGO function generation
argsgccgolist := strings.Join(argsgccgo, ", ")
var callgccgo string
if sysname == "select" {
// select is a keyword of Go. Its name is
// changed to c_select.
callgccgo = fmt.Sprintf("C.c_%s(%s)", sysname, argsgccgolist)
} else {
callgccgo = fmt.Sprintf("C.%s(%s)", sysname, argsgccgolist)
}
textgccgo += callProto
textgccgo += fmt.Sprintf("\tr1 = uintptr(%s)\n", callgccgo)
textgccgo += "\te1 = syscall.GetErrno()\n"
textgccgo += "\treturn\n}\n"
}
if err := s.Err(); err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
file.Close()
}
imp := ""
if pack != "unix" {
imp = "import \"golang.org/x/sys/unix\"\n"
}
// Print zsyscall_aix_ppc64.go
err := ioutil.WriteFile("zsyscall_aix_ppc64.go",
[]byte(fmt.Sprintf(srcTemplate1, cmdLine(), buildTags(), pack, imp, textcommon)),
0644)
if err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
// Print zsyscall_aix_ppc64_gc.go
vardecls := "\t" + strings.Join(vars, ",\n\t")
vardecls += " syscallFunc"
err = ioutil.WriteFile("zsyscall_aix_ppc64_gc.go",
[]byte(fmt.Sprintf(srcTemplate2, cmdLine(), buildTags(), pack, imp, dynimports, linknames, vardecls, textgc)),
0644)
if err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
// Print zsyscall_aix_ppc64_gccgo.go
err = ioutil.WriteFile("zsyscall_aix_ppc64_gccgo.go",
[]byte(fmt.Sprintf(srcTemplate3, cmdLine(), buildTags(), pack, cExtern, imp, textgccgo)),
0644)
if err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
}
const srcTemplate1 = `// %s
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s
package %s
import (
"unsafe"
)
%s
%s
`
const srcTemplate2 = `// %s
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s
// +build !gccgo
package %s
import (
"unsafe"
)
%s
%s
%s
type syscallFunc uintptr
var (
%s
)
// Implemented in runtime/syscall_aix.go.
func rawSyscall6(trap, nargs, a1, a2, a3, a4, a5, a6 uintptr) (r1, r2 uintptr, err Errno)
func syscall6(trap, nargs, a1, a2, a3, a4, a5, a6 uintptr) (r1, r2 uintptr, err Errno)
%s
`
const srcTemplate3 = `// %s
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s
// +build gccgo
package %s
%s
*/
import "C"
import (
"syscall"
)
%s
%s
`

View File

@@ -1,341 +0,0 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
This program reads a file containing function prototypes
(like syscall_solaris.go) and generates system call bodies.
The prototypes are marked by lines beginning with "//sys"
and read like func declarations if //sys is replaced by func, but:
* The parameter lists must give a name for each argument.
This includes return parameters.
* The parameter lists must give a type for each argument:
the (x, y, z int) shorthand is not allowed.
* If the return parameter is an error number, it must be named err.
* If go func name needs to be different than its libc name,
* or the function is not in libc, name could be specified
* at the end, after "=" sign, like
//sys getsockopt(s int, level int, name int, val uintptr, vallen *_Socklen) (err error) = libsocket.getsockopt
*/
package main
import (
"bufio"
"flag"
"fmt"
"os"
"regexp"
"strings"
)
var (
b32 = flag.Bool("b32", false, "32bit big-endian")
l32 = flag.Bool("l32", false, "32bit little-endian")
tags = flag.String("tags", "", "build tags")
illumos = flag.Bool("illumos", false, "illumos specific code generation")
)
// cmdLine returns this programs's commandline arguments
func cmdLine() string {
return "go run mksyscall_solaris.go " + strings.Join(os.Args[1:], " ")
}
// buildTags returns build tags
func buildTags() string {
return *tags
}
// Param is function parameter
type Param struct {
Name string
Type string
}
// usage prints the program usage
func usage() {
fmt.Fprintf(os.Stderr, "usage: go run mksyscall_solaris.go [-b32 | -l32] [-tags x,y] [file ...]\n")
os.Exit(1)
}
// parseParamList parses parameter list and returns a slice of parameters
func parseParamList(list string) []string {
list = strings.TrimSpace(list)
if list == "" {
return []string{}
}
return regexp.MustCompile(`\s*,\s*`).Split(list, -1)
}
// parseParam splits a parameter into name and type
func parseParam(p string) Param {
ps := regexp.MustCompile(`^(\S*) (\S*)$`).FindStringSubmatch(p)
if ps == nil {
fmt.Fprintf(os.Stderr, "malformed parameter: %s\n", p)
os.Exit(1)
}
return Param{ps[1], ps[2]}
}
func main() {
flag.Usage = usage
flag.Parse()
if len(flag.Args()) <= 0 {
fmt.Fprintf(os.Stderr, "no files to parse provided\n")
usage()
}
endianness := ""
if *b32 {
endianness = "big-endian"
} else if *l32 {
endianness = "little-endian"
}
pack := ""
text := ""
dynimports := ""
linknames := ""
var vars []string
for _, path := range flag.Args() {
file, err := os.Open(path)
if err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
s := bufio.NewScanner(file)
for s.Scan() {
t := s.Text()
t = strings.TrimSpace(t)
t = regexp.MustCompile(`\s+`).ReplaceAllString(t, ` `)
if p := regexp.MustCompile(`^package (\S+)$`).FindStringSubmatch(t); p != nil && pack == "" {
pack = p[1]
}
nonblock := regexp.MustCompile(`^\/\/sysnb `).FindStringSubmatch(t)
if regexp.MustCompile(`^\/\/sys `).FindStringSubmatch(t) == nil && nonblock == nil {
continue
}
// Line must be of the form
// func Open(path string, mode int, perm int) (fd int, err error)
// Split into name, in params, out params.
f := regexp.MustCompile(`^\/\/sys(nb)? (\w+)\(([^()]*)\)\s*(?:\(([^()]+)\))?\s*(?:=\s*(?:(\w*)\.)?(\w*))?$`).FindStringSubmatch(t)
if f == nil {
fmt.Fprintf(os.Stderr, "%s:%s\nmalformed //sys declaration\n", path, t)
os.Exit(1)
}
funct, inps, outps, modname, sysname := f[2], f[3], f[4], f[5], f[6]
// Split argument lists on comma.
in := parseParamList(inps)
out := parseParamList(outps)
inps = strings.Join(in, ", ")
outps = strings.Join(out, ", ")
// Try in vain to keep people from editing this file.
// The theory is that they jump into the middle of the file
// without reading the header.
text += "// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT\n\n"
// So file name.
if modname == "" {
modname = "libc"
}
// System call name.
if sysname == "" {
sysname = funct
}
// System call pointer variable name.
sysvarname := fmt.Sprintf("proc%s", sysname)
strconvfunc := "BytePtrFromString"
strconvtype := "*byte"
sysname = strings.ToLower(sysname) // All libc functions are lowercase.
// Runtime import of function to allow cross-platform builds.
dynimports += fmt.Sprintf("//go:cgo_import_dynamic libc_%s %s \"%s.so\"\n", sysname, sysname, modname)
// Link symbol to proc address variable.
linknames += fmt.Sprintf("//go:linkname %s libc_%s\n", sysvarname, sysname)
// Library proc address variable.
vars = append(vars, sysvarname)
// Go function header.
outlist := strings.Join(out, ", ")
if outlist != "" {
outlist = fmt.Sprintf(" (%s)", outlist)
}
if text != "" {
text += "\n"
}
text += fmt.Sprintf("func %s(%s)%s {\n", funct, strings.Join(in, ", "), outlist)
// Check if err return available
errvar := ""
for _, param := range out {
p := parseParam(param)
if p.Type == "error" {
errvar = p.Name
continue
}
}
// Prepare arguments to Syscall.
var args []string
n := 0
for _, param := range in {
p := parseParam(param)
if regexp.MustCompile(`^\*`).FindStringSubmatch(p.Type) != nil {
args = append(args, "uintptr(unsafe.Pointer("+p.Name+"))")
} else if p.Type == "string" && errvar != "" {
text += fmt.Sprintf("\tvar _p%d %s\n", n, strconvtype)
text += fmt.Sprintf("\t_p%d, %s = %s(%s)\n", n, errvar, strconvfunc, p.Name)
text += fmt.Sprintf("\tif %s != nil {\n\t\treturn\n\t}\n", errvar)
args = append(args, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n))
n++
} else if p.Type == "string" {
fmt.Fprintf(os.Stderr, path+":"+funct+" uses string arguments, but has no error return\n")
text += fmt.Sprintf("\tvar _p%d %s\n", n, strconvtype)
text += fmt.Sprintf("\t_p%d, _ = %s(%s)\n", n, strconvfunc, p.Name)
args = append(args, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n))
n++
} else if s := regexp.MustCompile(`^\[\](.*)`).FindStringSubmatch(p.Type); s != nil {
// Convert slice into pointer, length.
// Have to be careful not to take address of &a[0] if len == 0:
// pass nil in that case.
text += fmt.Sprintf("\tvar _p%d *%s\n", n, s[1])
text += fmt.Sprintf("\tif len(%s) > 0 {\n\t\t_p%d = &%s[0]\n\t}\n", p.Name, n, p.Name)
args = append(args, fmt.Sprintf("uintptr(unsafe.Pointer(_p%d))", n), fmt.Sprintf("uintptr(len(%s))", p.Name))
n++
} else if p.Type == "int64" && endianness != "" {
if endianness == "big-endian" {
args = append(args, fmt.Sprintf("uintptr(%s>>32)", p.Name), fmt.Sprintf("uintptr(%s)", p.Name))
} else {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name), fmt.Sprintf("uintptr(%s>>32)", p.Name))
}
} else if p.Type == "bool" {
text += fmt.Sprintf("\tvar _p%d uint32\n", n)
text += fmt.Sprintf("\tif %s {\n\t\t_p%d = 1\n\t} else {\n\t\t_p%d = 0\n\t}\n", p.Name, n, n)
args = append(args, fmt.Sprintf("uintptr(_p%d)", n))
n++
} else {
args = append(args, fmt.Sprintf("uintptr(%s)", p.Name))
}
}
nargs := len(args)
// Determine which form to use; pad args with zeros.
asm := "sysvicall6"
if nonblock != nil {
asm = "rawSysvicall6"
}
if len(args) <= 6 {
for len(args) < 6 {
args = append(args, "0")
}
} else {
fmt.Fprintf(os.Stderr, "%s: too many arguments to system call\n", path)
os.Exit(1)
}
// Actual call.
arglist := strings.Join(args, ", ")
call := fmt.Sprintf("%s(uintptr(unsafe.Pointer(&%s)), %d, %s)", asm, sysvarname, nargs, arglist)
// Assign return values.
body := ""
ret := []string{"_", "_", "_"}
doErrno := false
for i := 0; i < len(out); i++ {
p := parseParam(out[i])
reg := ""
if p.Name == "err" {
reg = "e1"
ret[2] = reg
doErrno = true
} else {
reg = fmt.Sprintf("r%d", i)
ret[i] = reg
}
if p.Type == "bool" {
reg = fmt.Sprintf("%d != 0", reg)
}
if p.Type == "int64" && endianness != "" {
// 64-bit number in r1:r0 or r0:r1.
if i+2 > len(out) {
fmt.Fprintf(os.Stderr, "%s: not enough registers for int64 return\n", path)
os.Exit(1)
}
if endianness == "big-endian" {
reg = fmt.Sprintf("int64(r%d)<<32 | int64(r%d)", i, i+1)
} else {
reg = fmt.Sprintf("int64(r%d)<<32 | int64(r%d)", i+1, i)
}
ret[i] = fmt.Sprintf("r%d", i)
ret[i+1] = fmt.Sprintf("r%d", i+1)
}
if reg != "e1" {
body += fmt.Sprintf("\t%s = %s(%s)\n", p.Name, p.Type, reg)
}
}
if ret[0] == "_" && ret[1] == "_" && ret[2] == "_" {
text += fmt.Sprintf("\t%s\n", call)
} else {
text += fmt.Sprintf("\t%s, %s, %s := %s\n", ret[0], ret[1], ret[2], call)
}
text += body
if doErrno {
text += "\tif e1 != 0 {\n"
text += "\t\terr = e1\n"
text += "\t}\n"
}
text += "\treturn\n"
text += "}\n"
}
if err := s.Err(); err != nil {
fmt.Fprintf(os.Stderr, err.Error())
os.Exit(1)
}
file.Close()
}
imp := ""
if pack != "unix" {
imp = "import \"golang.org/x/sys/unix\"\n"
}
syscallimp := ""
if !*illumos {
syscallimp = "\"syscall\""
}
vardecls := "\t" + strings.Join(vars, ",\n\t")
vardecls += " syscallFunc"
fmt.Printf(srcTemplate, cmdLine(), buildTags(), pack, syscallimp, imp, dynimports, linknames, vardecls, text)
}
const srcTemplate = `// %s
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s
package %s
import (
"unsafe"
%s
)
%s
%s
%s
var (
%s
)
%s
`

View File

@@ -1,355 +0,0 @@
// Copyright 2019 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// Parse the header files for OpenBSD and generate a Go usable sysctl MIB.
//
// Build a MIB with each entry being an array containing the level, type and
// a hash that will contain additional entries if the current entry is a node.
// We then walk this MIB and create a flattened sysctl name to OID hash.
package main
import (
"bufio"
"fmt"
"os"
"path/filepath"
"regexp"
"sort"
"strings"
)
var (
goos, goarch string
)
// cmdLine returns this programs's commandline arguments.
func cmdLine() string {
return "go run mksysctl_openbsd.go " + strings.Join(os.Args[1:], " ")
}
// buildTags returns build tags.
func buildTags() string {
return fmt.Sprintf("%s,%s", goarch, goos)
}
// reMatch performs regular expression match and stores the substring slice to value pointed by m.
func reMatch(re *regexp.Regexp, str string, m *[]string) bool {
*m = re.FindStringSubmatch(str)
if *m != nil {
return true
}
return false
}
type nodeElement struct {
n int
t string
pE *map[string]nodeElement
}
var (
debugEnabled bool
mib map[string]nodeElement
node *map[string]nodeElement
nodeMap map[string]string
sysCtl []string
)
var (
ctlNames1RE = regexp.MustCompile(`^#define\s+(CTL_NAMES)\s+{`)
ctlNames2RE = regexp.MustCompile(`^#define\s+(CTL_(.*)_NAMES)\s+{`)
ctlNames3RE = regexp.MustCompile(`^#define\s+((.*)CTL_NAMES)\s+{`)
netInetRE = regexp.MustCompile(`^netinet/`)
netInet6RE = regexp.MustCompile(`^netinet6/`)
netRE = regexp.MustCompile(`^net/`)
bracesRE = regexp.MustCompile(`{.*}`)
ctlTypeRE = regexp.MustCompile(`{\s+"(\w+)",\s+(CTLTYPE_[A-Z]+)\s+}`)
fsNetKernRE = regexp.MustCompile(`^(fs|net|kern)_`)
)
func debug(s string) {
if debugEnabled {
fmt.Fprintln(os.Stderr, s)
}
}
// Walk the MIB and build a sysctl name to OID mapping.
func buildSysctl(pNode *map[string]nodeElement, name string, oid []int) {
lNode := pNode // local copy of pointer to node
var keys []string
for k := range *lNode {
keys = append(keys, k)
}
sort.Strings(keys)
for _, key := range keys {
nodename := name
if name != "" {
nodename += "."
}
nodename += key
nodeoid := append(oid, (*pNode)[key].n)
if (*pNode)[key].t == `CTLTYPE_NODE` {
if _, ok := nodeMap[nodename]; ok {
lNode = &mib
ctlName := nodeMap[nodename]
for _, part := range strings.Split(ctlName, ".") {
lNode = ((*lNode)[part]).pE
}
} else {
lNode = (*pNode)[key].pE
}
buildSysctl(lNode, nodename, nodeoid)
} else if (*pNode)[key].t != "" {
oidStr := []string{}
for j := range nodeoid {
oidStr = append(oidStr, fmt.Sprintf("%d", nodeoid[j]))
}
text := "\t{ \"" + nodename + "\", []_C_int{ " + strings.Join(oidStr, ", ") + " } }, \n"
sysCtl = append(sysCtl, text)
}
}
}
func main() {
// Get the OS (using GOOS_TARGET if it exist)
goos = os.Getenv("GOOS_TARGET")
if goos == "" {
goos = os.Getenv("GOOS")
}
// Get the architecture (using GOARCH_TARGET if it exists)
goarch = os.Getenv("GOARCH_TARGET")
if goarch == "" {
goarch = os.Getenv("GOARCH")
}
// Check if GOOS and GOARCH environment variables are defined
if goarch == "" || goos == "" {
fmt.Fprintf(os.Stderr, "GOARCH or GOOS not defined in environment\n")
os.Exit(1)
}
mib = make(map[string]nodeElement)
headers := [...]string{
`sys/sysctl.h`,
`sys/socket.h`,
`sys/tty.h`,
`sys/malloc.h`,
`sys/mount.h`,
`sys/namei.h`,
`sys/sem.h`,
`sys/shm.h`,
`sys/vmmeter.h`,
`uvm/uvmexp.h`,
`uvm/uvm_param.h`,
`uvm/uvm_swap_encrypt.h`,
`ddb/db_var.h`,
`net/if.h`,
`net/if_pfsync.h`,
`net/pipex.h`,
`netinet/in.h`,
`netinet/icmp_var.h`,
`netinet/igmp_var.h`,
`netinet/ip_ah.h`,
`netinet/ip_carp.h`,
`netinet/ip_divert.h`,
`netinet/ip_esp.h`,
`netinet/ip_ether.h`,
`netinet/ip_gre.h`,
`netinet/ip_ipcomp.h`,
`netinet/ip_ipip.h`,
`netinet/pim_var.h`,
`netinet/tcp_var.h`,
`netinet/udp_var.h`,
`netinet6/in6.h`,
`netinet6/ip6_divert.h`,
`netinet6/pim6_var.h`,
`netinet/icmp6.h`,
`netmpls/mpls.h`,
}
ctls := [...]string{
`kern`,
`vm`,
`fs`,
`net`,
//debug /* Special handling required */
`hw`,
//machdep /* Arch specific */
`user`,
`ddb`,
//vfs /* Special handling required */
`fs.posix`,
`kern.forkstat`,
`kern.intrcnt`,
`kern.malloc`,
`kern.nchstats`,
`kern.seminfo`,
`kern.shminfo`,
`kern.timecounter`,
`kern.tty`,
`kern.watchdog`,
`net.bpf`,
`net.ifq`,
`net.inet`,
`net.inet.ah`,
`net.inet.carp`,
`net.inet.divert`,
`net.inet.esp`,
`net.inet.etherip`,
`net.inet.gre`,
`net.inet.icmp`,
`net.inet.igmp`,
`net.inet.ip`,
`net.inet.ip.ifq`,
`net.inet.ipcomp`,
`net.inet.ipip`,
`net.inet.mobileip`,
`net.inet.pfsync`,
`net.inet.pim`,
`net.inet.tcp`,
`net.inet.udp`,
`net.inet6`,
`net.inet6.divert`,
`net.inet6.ip6`,
`net.inet6.icmp6`,
`net.inet6.pim6`,
`net.inet6.tcp6`,
`net.inet6.udp6`,
`net.mpls`,
`net.mpls.ifq`,
`net.key`,
`net.pflow`,
`net.pfsync`,
`net.pipex`,
`net.rt`,
`vm.swapencrypt`,
//vfsgenctl /* Special handling required */
}
// Node name "fixups"
ctlMap := map[string]string{
"ipproto": "net.inet",
"net.inet.ipproto": "net.inet",
"net.inet6.ipv6proto": "net.inet6",
"net.inet6.ipv6": "net.inet6.ip6",
"net.inet.icmpv6": "net.inet6.icmp6",
"net.inet6.divert6": "net.inet6.divert",
"net.inet6.tcp6": "net.inet.tcp",
"net.inet6.udp6": "net.inet.udp",
"mpls": "net.mpls",
"swpenc": "vm.swapencrypt",
}
// Node mappings
nodeMap = map[string]string{
"net.inet.ip.ifq": "net.ifq",
"net.inet.pfsync": "net.pfsync",
"net.mpls.ifq": "net.ifq",
}
mCtls := make(map[string]bool)
for _, ctl := range ctls {
mCtls[ctl] = true
}
for _, header := range headers {
debug("Processing " + header)
file, err := os.Open(filepath.Join("/usr/include", header))
if err != nil {
fmt.Fprintf(os.Stderr, "%v\n", err)
os.Exit(1)
}
s := bufio.NewScanner(file)
for s.Scan() {
var sub []string
if reMatch(ctlNames1RE, s.Text(), &sub) ||
reMatch(ctlNames2RE, s.Text(), &sub) ||
reMatch(ctlNames3RE, s.Text(), &sub) {
if sub[1] == `CTL_NAMES` {
// Top level.
node = &mib
} else {
// Node.
nodename := strings.ToLower(sub[2])
ctlName := ""
if reMatch(netInetRE, header, &sub) {
ctlName = "net.inet." + nodename
} else if reMatch(netInet6RE, header, &sub) {
ctlName = "net.inet6." + nodename
} else if reMatch(netRE, header, &sub) {
ctlName = "net." + nodename
} else {
ctlName = nodename
ctlName = fsNetKernRE.ReplaceAllString(ctlName, `$1.`)
}
if val, ok := ctlMap[ctlName]; ok {
ctlName = val
}
if _, ok := mCtls[ctlName]; !ok {
debug("Ignoring " + ctlName + "...")
continue
}
// Walk down from the top of the MIB.
node = &mib
for _, part := range strings.Split(ctlName, ".") {
if _, ok := (*node)[part]; !ok {
debug("Missing node " + part)
(*node)[part] = nodeElement{n: 0, t: "", pE: &map[string]nodeElement{}}
}
node = (*node)[part].pE
}
}
// Populate current node with entries.
i := -1
for !strings.HasPrefix(s.Text(), "}") {
s.Scan()
if reMatch(bracesRE, s.Text(), &sub) {
i++
}
if !reMatch(ctlTypeRE, s.Text(), &sub) {
continue
}
(*node)[sub[1]] = nodeElement{n: i, t: sub[2], pE: &map[string]nodeElement{}}
}
}
}
err = s.Err()
if err != nil {
fmt.Fprintf(os.Stderr, "%v\n", err)
os.Exit(1)
}
file.Close()
}
buildSysctl(&mib, "", []int{})
sort.Strings(sysCtl)
text := strings.Join(sysCtl, "")
fmt.Printf(srcTemplate, cmdLine(), buildTags(), text)
}
const srcTemplate = `// %s
// Code generated by the command above; DO NOT EDIT.
// +build %s
package unix
type mibentry struct {
ctlname string
ctloid []_C_int
}
var sysctlMib = []mibentry {
%s
}
`

View File

@@ -1,190 +0,0 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// Generate system call table for DragonFly, NetBSD,
// FreeBSD, OpenBSD or Darwin from master list
// (for example, /usr/src/sys/kern/syscalls.master or
// sys/syscall.h).
package main
import (
"bufio"
"fmt"
"io"
"io/ioutil"
"net/http"
"os"
"regexp"
"strings"
)
var (
goos, goarch string
)
// cmdLine returns this programs's commandline arguments
func cmdLine() string {
return "go run mksysnum.go " + strings.Join(os.Args[1:], " ")
}
// buildTags returns build tags
func buildTags() string {
return fmt.Sprintf("%s,%s", goarch, goos)
}
func checkErr(err error) {
if err != nil {
fmt.Fprintf(os.Stderr, "%v\n", err)
os.Exit(1)
}
}
// source string and substring slice for regexp
type re struct {
str string // source string
sub []string // matched sub-string
}
// Match performs regular expression match
func (r *re) Match(exp string) bool {
r.sub = regexp.MustCompile(exp).FindStringSubmatch(r.str)
if r.sub != nil {
return true
}
return false
}
// fetchFile fetches a text file from URL
func fetchFile(URL string) io.Reader {
resp, err := http.Get(URL)
checkErr(err)
defer resp.Body.Close()
body, err := ioutil.ReadAll(resp.Body)
checkErr(err)
return strings.NewReader(string(body))
}
// readFile reads a text file from path
func readFile(path string) io.Reader {
file, err := os.Open(os.Args[1])
checkErr(err)
return file
}
func format(name, num, proto string) string {
name = strings.ToUpper(name)
// There are multiple entries for enosys and nosys, so comment them out.
nm := re{str: name}
if nm.Match(`^SYS_E?NOSYS$`) {
name = fmt.Sprintf("// %s", name)
}
if name == `SYS_SYS_EXIT` {
name = `SYS_EXIT`
}
return fmt.Sprintf(" %s = %s; // %s\n", name, num, proto)
}
func main() {
// Get the OS (using GOOS_TARGET if it exist)
goos = os.Getenv("GOOS_TARGET")
if goos == "" {
goos = os.Getenv("GOOS")
}
// Get the architecture (using GOARCH_TARGET if it exists)
goarch = os.Getenv("GOARCH_TARGET")
if goarch == "" {
goarch = os.Getenv("GOARCH")
}
// Check if GOOS and GOARCH environment variables are defined
if goarch == "" || goos == "" {
fmt.Fprintf(os.Stderr, "GOARCH or GOOS not defined in environment\n")
os.Exit(1)
}
file := strings.TrimSpace(os.Args[1])
var syscalls io.Reader
if strings.HasPrefix(file, "https://") || strings.HasPrefix(file, "http://") {
// Download syscalls.master file
syscalls = fetchFile(file)
} else {
syscalls = readFile(file)
}
var text, line string
s := bufio.NewScanner(syscalls)
for s.Scan() {
t := re{str: line}
if t.Match(`^(.*)\\$`) {
// Handle continuation
line = t.sub[1]
line += strings.TrimLeft(s.Text(), " \t")
} else {
// New line
line = s.Text()
}
t = re{str: line}
if t.Match(`\\$`) {
continue
}
t = re{str: line}
switch goos {
case "dragonfly":
if t.Match(`^([0-9]+)\s+STD\s+({ \S+\s+(\w+).*)$`) {
num, proto := t.sub[1], t.sub[2]
name := fmt.Sprintf("SYS_%s", t.sub[3])
text += format(name, num, proto)
}
case "freebsd":
if t.Match(`^([0-9]+)\s+\S+\s+(?:(?:NO)?STD|COMPAT10)\s+({ \S+\s+(\w+).*)$`) {
num, proto := t.sub[1], t.sub[2]
name := fmt.Sprintf("SYS_%s", t.sub[3])
text += format(name, num, proto)
}
case "openbsd":
if t.Match(`^([0-9]+)\s+STD\s+(NOLOCK\s+)?({ \S+\s+\*?(\w+).*)$`) {
num, proto, name := t.sub[1], t.sub[3], t.sub[4]
text += format(name, num, proto)
}
case "netbsd":
if t.Match(`^([0-9]+)\s+((STD)|(NOERR))\s+(RUMP\s+)?({\s+\S+\s*\*?\s*\|(\S+)\|(\S*)\|(\w+).*\s+})(\s+(\S+))?$`) {
num, proto, compat := t.sub[1], t.sub[6], t.sub[8]
name := t.sub[7] + "_" + t.sub[9]
if t.sub[11] != "" {
name = t.sub[7] + "_" + t.sub[11]
}
name = strings.ToUpper(name)
if compat == "" || compat == "13" || compat == "30" || compat == "50" {
text += fmt.Sprintf(" %s = %s; // %s\n", name, num, proto)
}
}
case "darwin":
if t.Match(`^#define\s+SYS_(\w+)\s+([0-9]+)`) {
name, num := t.sub[1], t.sub[2]
name = strings.ToUpper(name)
text += fmt.Sprintf(" SYS_%s = %s;\n", name, num)
}
default:
fmt.Fprintf(os.Stderr, "unrecognized GOOS=%s\n", goos)
os.Exit(1)
}
}
err := s.Err()
checkErr(err)
fmt.Printf(template, cmdLine(), buildTags(), text)
}
const template = `// %s
// Code generated by the command above; see README.md. DO NOT EDIT.
// +build %s
package unix
const(
%s)`

View File

@@ -1,237 +0,0 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// +build aix
/*
Input to cgo -godefs. See also mkerrors.sh and mkall.sh
*/
// +godefs map struct_in_addr [4]byte /* in_addr */
// +godefs map struct_in6_addr [16]byte /* in6_addr */
package unix
/*
#include <sys/types.h>
#include <sys/time.h>
#include <sys/limits.h>
#include <sys/un.h>
#include <utime.h>
#include <sys/utsname.h>
#include <sys/poll.h>
#include <sys/resource.h>
#include <sys/stat.h>
#include <sys/statfs.h>
#include <sys/termio.h>
#include <sys/ioctl.h>
#include <termios.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <netinet/in.h>
#include <netinet/icmp6.h>
#include <dirent.h>
#include <fcntl.h>
enum {
sizeofPtr = sizeof(void*),
};
union sockaddr_all {
struct sockaddr s1; // this one gets used for fields
struct sockaddr_in s2; // these pad it out
struct sockaddr_in6 s3;
struct sockaddr_un s4;
struct sockaddr_dl s5;
};
struct sockaddr_any {
struct sockaddr addr;
char pad[sizeof(union sockaddr_all) - sizeof(struct sockaddr)];
};
*/
import "C"
// Machine characteristics
const (
SizeofPtr = C.sizeofPtr
SizeofShort = C.sizeof_short
SizeofInt = C.sizeof_int
SizeofLong = C.sizeof_long
SizeofLongLong = C.sizeof_longlong
PathMax = C.PATH_MAX
)
// Basic types
type (
_C_short C.short
_C_int C.int
_C_long C.long
_C_long_long C.longlong
)
type off64 C.off64_t
type off C.off_t
type Mode_t C.mode_t
// Time
type Timespec C.struct_timespec
type Timeval C.struct_timeval
type Timeval32 C.struct_timeval32
type Timex C.struct_timex
type Time_t C.time_t
type Tms C.struct_tms
type Utimbuf C.struct_utimbuf
type Timezone C.struct_timezone
// Processes
type Rusage C.struct_rusage
type Rlimit C.struct_rlimit64
type Pid_t C.pid_t
type _Gid_t C.gid_t
type dev_t C.dev_t
// Files
type Stat_t C.struct_stat
type StatxTimestamp C.struct_statx_timestamp
type Statx_t C.struct_statx
type Dirent C.struct_dirent
// Sockets
type RawSockaddrInet4 C.struct_sockaddr_in
type RawSockaddrInet6 C.struct_sockaddr_in6
type RawSockaddrUnix C.struct_sockaddr_un
type RawSockaddrDatalink C.struct_sockaddr_dl
type RawSockaddr C.struct_sockaddr
type RawSockaddrAny C.struct_sockaddr_any
type _Socklen C.socklen_t
type Cmsghdr C.struct_cmsghdr
type ICMPv6Filter C.struct_icmp6_filter
type Iovec C.struct_iovec
type IPMreq C.struct_ip_mreq
type IPv6Mreq C.struct_ipv6_mreq
type IPv6MTUInfo C.struct_ip6_mtuinfo
type Linger C.struct_linger
type Msghdr C.struct_msghdr
const (
SizeofSockaddrInet4 = C.sizeof_struct_sockaddr_in
SizeofSockaddrInet6 = C.sizeof_struct_sockaddr_in6
SizeofSockaddrAny = C.sizeof_struct_sockaddr_any
SizeofSockaddrUnix = C.sizeof_struct_sockaddr_un
SizeofSockaddrDatalink = C.sizeof_struct_sockaddr_dl
SizeofLinger = C.sizeof_struct_linger
SizeofIPMreq = C.sizeof_struct_ip_mreq
SizeofIPv6Mreq = C.sizeof_struct_ipv6_mreq
SizeofIPv6MTUInfo = C.sizeof_struct_ip6_mtuinfo
SizeofMsghdr = C.sizeof_struct_msghdr
SizeofCmsghdr = C.sizeof_struct_cmsghdr
SizeofICMPv6Filter = C.sizeof_struct_icmp6_filter
)
// Routing and interface messages
const (
SizeofIfMsghdr = C.sizeof_struct_if_msghdr
)
type IfMsgHdr C.struct_if_msghdr
// Misc
type FdSet C.fd_set
type Utsname C.struct_utsname
type Ustat_t C.struct_ustat
type Sigset_t C.sigset_t
const (
AT_FDCWD = C.AT_FDCWD
AT_REMOVEDIR = C.AT_REMOVEDIR
AT_SYMLINK_NOFOLLOW = C.AT_SYMLINK_NOFOLLOW
)
// Terminal handling
type Termios C.struct_termios
type Termio C.struct_termio
type Winsize C.struct_winsize
//poll
type PollFd struct {
Fd int32
Events uint16
Revents uint16
}
const (
POLLERR = C.POLLERR
POLLHUP = C.POLLHUP
POLLIN = C.POLLIN
POLLNVAL = C.POLLNVAL
POLLOUT = C.POLLOUT
POLLPRI = C.POLLPRI
POLLRDBAND = C.POLLRDBAND
POLLRDNORM = C.POLLRDNORM
POLLWRBAND = C.POLLWRBAND
POLLWRNORM = C.POLLWRNORM
)
//flock_t
type Flock_t C.struct_flock64
// Statfs
type Fsid_t C.struct_fsid_t
type Fsid64_t C.struct_fsid64_t
type Statfs_t C.struct_statfs
const RNDGETENTCNT = 0x80045200

View File

@@ -1,283 +0,0 @@
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
Input to cgo -godefs. See README.md
*/
// +godefs map struct_in_addr [4]byte /* in_addr */
// +godefs map struct_in6_addr [16]byte /* in6_addr */
package unix
/*
#define __DARWIN_UNIX03 0
#define KERNEL
#define _DARWIN_USE_64_BIT_INODE
#include <dirent.h>
#include <fcntl.h>
#include <poll.h>
#include <signal.h>
#include <termios.h>
#include <unistd.h>
#include <mach/mach.h>
#include <mach/message.h>
#include <sys/event.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/param.h>
#include <sys/ptrace.h>
#include <sys/resource.h>
#include <sys/select.h>
#include <sys/signal.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <sys/types.h>
#include <sys/uio.h>
#include <sys/un.h>
#include <sys/utsname.h>
#include <sys/wait.h>
#include <net/bpf.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/if_var.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/icmp6.h>
#include <netinet/tcp.h>
enum {
sizeofPtr = sizeof(void*),
};
union sockaddr_all {
struct sockaddr s1; // this one gets used for fields
struct sockaddr_in s2; // these pad it out
struct sockaddr_in6 s3;
struct sockaddr_un s4;
struct sockaddr_dl s5;
};
struct sockaddr_any {
struct sockaddr addr;
char pad[sizeof(union sockaddr_all) - sizeof(struct sockaddr)];
};
*/
import "C"
// Machine characteristics
const (
SizeofPtr = C.sizeofPtr
SizeofShort = C.sizeof_short
SizeofInt = C.sizeof_int
SizeofLong = C.sizeof_long
SizeofLongLong = C.sizeof_longlong
)
// Basic types
type (
_C_short C.short
_C_int C.int
_C_long C.long
_C_long_long C.longlong
)
// Time
type Timespec C.struct_timespec
type Timeval C.struct_timeval
type Timeval32 C.struct_timeval32
// Processes
type Rusage C.struct_rusage
type Rlimit C.struct_rlimit
type _Gid_t C.gid_t
// Files
type Stat_t C.struct_stat64
type Statfs_t C.struct_statfs64
type Flock_t C.struct_flock
type Fstore_t C.struct_fstore
type Radvisory_t C.struct_radvisory
type Fbootstraptransfer_t C.struct_fbootstraptransfer
type Log2phys_t C.struct_log2phys
type Fsid C.struct_fsid
type Dirent C.struct_dirent
// Sockets
type RawSockaddrInet4 C.struct_sockaddr_in
type RawSockaddrInet6 C.struct_sockaddr_in6
type RawSockaddrUnix C.struct_sockaddr_un
type RawSockaddrDatalink C.struct_sockaddr_dl
type RawSockaddr C.struct_sockaddr
type RawSockaddrAny C.struct_sockaddr_any
type _Socklen C.socklen_t
type Linger C.struct_linger
type Iovec C.struct_iovec
type IPMreq C.struct_ip_mreq
type IPv6Mreq C.struct_ipv6_mreq
type Msghdr C.struct_msghdr
type Cmsghdr C.struct_cmsghdr
type Inet4Pktinfo C.struct_in_pktinfo
type Inet6Pktinfo C.struct_in6_pktinfo
type IPv6MTUInfo C.struct_ip6_mtuinfo
type ICMPv6Filter C.struct_icmp6_filter
const (
SizeofSockaddrInet4 = C.sizeof_struct_sockaddr_in
SizeofSockaddrInet6 = C.sizeof_struct_sockaddr_in6
SizeofSockaddrAny = C.sizeof_struct_sockaddr_any
SizeofSockaddrUnix = C.sizeof_struct_sockaddr_un
SizeofSockaddrDatalink = C.sizeof_struct_sockaddr_dl
SizeofLinger = C.sizeof_struct_linger
SizeofIPMreq = C.sizeof_struct_ip_mreq
SizeofIPv6Mreq = C.sizeof_struct_ipv6_mreq
SizeofMsghdr = C.sizeof_struct_msghdr
SizeofCmsghdr = C.sizeof_struct_cmsghdr
SizeofInet4Pktinfo = C.sizeof_struct_in_pktinfo
SizeofInet6Pktinfo = C.sizeof_struct_in6_pktinfo
SizeofIPv6MTUInfo = C.sizeof_struct_ip6_mtuinfo
SizeofICMPv6Filter = C.sizeof_struct_icmp6_filter
)
// Ptrace requests
const (
PTRACE_TRACEME = C.PT_TRACE_ME
PTRACE_CONT = C.PT_CONTINUE
PTRACE_KILL = C.PT_KILL
)
// Events (kqueue, kevent)
type Kevent_t C.struct_kevent
// Select
type FdSet C.fd_set
// Routing and interface messages
const (
SizeofIfMsghdr = C.sizeof_struct_if_msghdr
SizeofIfData = C.sizeof_struct_if_data
SizeofIfaMsghdr = C.sizeof_struct_ifa_msghdr
SizeofIfmaMsghdr = C.sizeof_struct_ifma_msghdr
SizeofIfmaMsghdr2 = C.sizeof_struct_ifma_msghdr2
SizeofRtMsghdr = C.sizeof_struct_rt_msghdr
SizeofRtMetrics = C.sizeof_struct_rt_metrics
)
type IfMsghdr C.struct_if_msghdr
type IfData C.struct_if_data
type IfaMsghdr C.struct_ifa_msghdr
type IfmaMsghdr C.struct_ifma_msghdr
type IfmaMsghdr2 C.struct_ifma_msghdr2
type RtMsghdr C.struct_rt_msghdr
type RtMetrics C.struct_rt_metrics
// Berkeley packet filter
const (
SizeofBpfVersion = C.sizeof_struct_bpf_version
SizeofBpfStat = C.sizeof_struct_bpf_stat
SizeofBpfProgram = C.sizeof_struct_bpf_program
SizeofBpfInsn = C.sizeof_struct_bpf_insn
SizeofBpfHdr = C.sizeof_struct_bpf_hdr
)
type BpfVersion C.struct_bpf_version
type BpfStat C.struct_bpf_stat
type BpfProgram C.struct_bpf_program
type BpfInsn C.struct_bpf_insn
type BpfHdr C.struct_bpf_hdr
// Terminal handling
type Termios C.struct_termios
type Winsize C.struct_winsize
// fchmodat-like syscalls.
const (
AT_FDCWD = C.AT_FDCWD
AT_REMOVEDIR = C.AT_REMOVEDIR
AT_SYMLINK_FOLLOW = C.AT_SYMLINK_FOLLOW
AT_SYMLINK_NOFOLLOW = C.AT_SYMLINK_NOFOLLOW
)
// poll
type PollFd C.struct_pollfd
const (
POLLERR = C.POLLERR
POLLHUP = C.POLLHUP
POLLIN = C.POLLIN
POLLNVAL = C.POLLNVAL
POLLOUT = C.POLLOUT
POLLPRI = C.POLLPRI
POLLRDBAND = C.POLLRDBAND
POLLRDNORM = C.POLLRDNORM
POLLWRBAND = C.POLLWRBAND
POLLWRNORM = C.POLLWRNORM
)
// uname
type Utsname C.struct_utsname
// Clockinfo
const SizeofClockinfo = C.sizeof_struct_clockinfo
type Clockinfo C.struct_clockinfo

View File

@@ -1,269 +0,0 @@
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
Input to cgo -godefs. See README.md
*/
// +godefs map struct_in_addr [4]byte /* in_addr */
// +godefs map struct_in6_addr [16]byte /* in6_addr */
package unix
/*
#define KERNEL
#include <dirent.h>
#include <fcntl.h>
#include <poll.h>
#include <signal.h>
#include <termios.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/event.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/param.h>
#include <sys/ptrace.h>
#include <sys/resource.h>
#include <sys/select.h>
#include <sys/signal.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <sys/types.h>
#include <sys/un.h>
#include <sys/utsname.h>
#include <sys/wait.h>
#include <net/bpf.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/icmp6.h>
#include <netinet/tcp.h>
enum {
sizeofPtr = sizeof(void*),
};
union sockaddr_all {
struct sockaddr s1; // this one gets used for fields
struct sockaddr_in s2; // these pad it out
struct sockaddr_in6 s3;
struct sockaddr_un s4;
struct sockaddr_dl s5;
};
struct sockaddr_any {
struct sockaddr addr;
char pad[sizeof(union sockaddr_all) - sizeof(struct sockaddr)];
};
*/
import "C"
// Machine characteristics
const (
SizeofPtr = C.sizeofPtr
SizeofShort = C.sizeof_short
SizeofInt = C.sizeof_int
SizeofLong = C.sizeof_long
SizeofLongLong = C.sizeof_longlong
)
// Basic types
type (
_C_short C.short
_C_int C.int
_C_long C.long
_C_long_long C.longlong
)
// Time
type Timespec C.struct_timespec
type Timeval C.struct_timeval
// Processes
type Rusage C.struct_rusage
type Rlimit C.struct_rlimit
type _Gid_t C.gid_t
// Files
type Stat_t C.struct_stat
type Statfs_t C.struct_statfs
type Flock_t C.struct_flock
type Dirent C.struct_dirent
type Fsid C.struct_fsid
// File system limits
const (
PathMax = C.PATH_MAX
)
// Sockets
type RawSockaddrInet4 C.struct_sockaddr_in
type RawSockaddrInet6 C.struct_sockaddr_in6
type RawSockaddrUnix C.struct_sockaddr_un
type RawSockaddrDatalink C.struct_sockaddr_dl
type RawSockaddr C.struct_sockaddr
type RawSockaddrAny C.struct_sockaddr_any
type _Socklen C.socklen_t
type Linger C.struct_linger
type Iovec C.struct_iovec
type IPMreq C.struct_ip_mreq
type IPv6Mreq C.struct_ipv6_mreq
type Msghdr C.struct_msghdr
type Cmsghdr C.struct_cmsghdr
type Inet6Pktinfo C.struct_in6_pktinfo
type IPv6MTUInfo C.struct_ip6_mtuinfo
type ICMPv6Filter C.struct_icmp6_filter
const (
SizeofSockaddrInet4 = C.sizeof_struct_sockaddr_in
SizeofSockaddrInet6 = C.sizeof_struct_sockaddr_in6
SizeofSockaddrAny = C.sizeof_struct_sockaddr_any
SizeofSockaddrUnix = C.sizeof_struct_sockaddr_un
SizeofSockaddrDatalink = C.sizeof_struct_sockaddr_dl
SizeofLinger = C.sizeof_struct_linger
SizeofIPMreq = C.sizeof_struct_ip_mreq
SizeofIPv6Mreq = C.sizeof_struct_ipv6_mreq
SizeofMsghdr = C.sizeof_struct_msghdr
SizeofCmsghdr = C.sizeof_struct_cmsghdr
SizeofInet6Pktinfo = C.sizeof_struct_in6_pktinfo
SizeofIPv6MTUInfo = C.sizeof_struct_ip6_mtuinfo
SizeofICMPv6Filter = C.sizeof_struct_icmp6_filter
)
// Ptrace requests
const (
PTRACE_TRACEME = C.PT_TRACE_ME
PTRACE_CONT = C.PT_CONTINUE
PTRACE_KILL = C.PT_KILL
)
// Events (kqueue, kevent)
type Kevent_t C.struct_kevent
// Select
type FdSet C.fd_set
// Routing and interface messages
const (
SizeofIfMsghdr = C.sizeof_struct_if_msghdr
SizeofIfData = C.sizeof_struct_if_data
SizeofIfaMsghdr = C.sizeof_struct_ifa_msghdr
SizeofIfmaMsghdr = C.sizeof_struct_ifma_msghdr
SizeofIfAnnounceMsghdr = C.sizeof_struct_if_announcemsghdr
SizeofRtMsghdr = C.sizeof_struct_rt_msghdr
SizeofRtMetrics = C.sizeof_struct_rt_metrics
)
type IfMsghdr C.struct_if_msghdr
type IfData C.struct_if_data
type IfaMsghdr C.struct_ifa_msghdr
type IfmaMsghdr C.struct_ifma_msghdr
type IfAnnounceMsghdr C.struct_if_announcemsghdr
type RtMsghdr C.struct_rt_msghdr
type RtMetrics C.struct_rt_metrics
// Berkeley packet filter
const (
SizeofBpfVersion = C.sizeof_struct_bpf_version
SizeofBpfStat = C.sizeof_struct_bpf_stat
SizeofBpfProgram = C.sizeof_struct_bpf_program
SizeofBpfInsn = C.sizeof_struct_bpf_insn
SizeofBpfHdr = C.sizeof_struct_bpf_hdr
)
type BpfVersion C.struct_bpf_version
type BpfStat C.struct_bpf_stat
type BpfProgram C.struct_bpf_program
type BpfInsn C.struct_bpf_insn
type BpfHdr C.struct_bpf_hdr
// Terminal handling
type Termios C.struct_termios
type Winsize C.struct_winsize
// fchmodat-like syscalls.
const (
AT_FDCWD = C.AT_FDCWD
AT_SYMLINK_NOFOLLOW = C.AT_SYMLINK_NOFOLLOW
)
// poll
type PollFd C.struct_pollfd
const (
POLLERR = C.POLLERR
POLLHUP = C.POLLHUP
POLLIN = C.POLLIN
POLLNVAL = C.POLLNVAL
POLLOUT = C.POLLOUT
POLLPRI = C.POLLPRI
POLLRDBAND = C.POLLRDBAND
POLLRDNORM = C.POLLRDNORM
POLLWRBAND = C.POLLWRBAND
POLLWRNORM = C.POLLWRNORM
)
// Uname
type Utsname C.struct_utsname
// Clockinfo
const SizeofClockinfo = C.sizeof_struct_clockinfo
type Clockinfo C.struct_clockinfo

View File

@@ -1,406 +0,0 @@
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
Input to cgo -godefs. See README.md
*/
// +godefs map struct_in_addr [4]byte /* in_addr */
// +godefs map struct_in6_addr [16]byte /* in6_addr */
package unix
/*
#define _WANT_FREEBSD11_STAT 1
#define _WANT_FREEBSD11_STATFS 1
#define _WANT_FREEBSD11_DIRENT 1
#define _WANT_FREEBSD11_KEVENT 1
#include <dirent.h>
#include <fcntl.h>
#include <poll.h>
#include <signal.h>
#include <termios.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/capsicum.h>
#include <sys/event.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/param.h>
#include <sys/ptrace.h>
#include <sys/resource.h>
#include <sys/select.h>
#include <sys/signal.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <sys/types.h>
#include <sys/un.h>
#include <sys/utsname.h>
#include <sys/wait.h>
#include <net/bpf.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/icmp6.h>
#include <netinet/tcp.h>
enum {
sizeofPtr = sizeof(void*),
};
union sockaddr_all {
struct sockaddr s1; // this one gets used for fields
struct sockaddr_in s2; // these pad it out
struct sockaddr_in6 s3;
struct sockaddr_un s4;
struct sockaddr_dl s5;
};
struct sockaddr_any {
struct sockaddr addr;
char pad[sizeof(union sockaddr_all) - sizeof(struct sockaddr)];
};
// This structure is a duplicate of if_data on FreeBSD 8-STABLE.
// See /usr/include/net/if.h.
struct if_data8 {
u_char ifi_type;
u_char ifi_physical;
u_char ifi_addrlen;
u_char ifi_hdrlen;
u_char ifi_link_state;
u_char ifi_spare_char1;
u_char ifi_spare_char2;
u_char ifi_datalen;
u_long ifi_mtu;
u_long ifi_metric;
u_long ifi_baudrate;
u_long ifi_ipackets;
u_long ifi_ierrors;
u_long ifi_opackets;
u_long ifi_oerrors;
u_long ifi_collisions;
u_long ifi_ibytes;
u_long ifi_obytes;
u_long ifi_imcasts;
u_long ifi_omcasts;
u_long ifi_iqdrops;
u_long ifi_noproto;
u_long ifi_hwassist;
// FIXME: these are now unions, so maybe need to change definitions?
#undef ifi_epoch
time_t ifi_epoch;
#undef ifi_lastchange
struct timeval ifi_lastchange;
};
// This structure is a duplicate of if_msghdr on FreeBSD 8-STABLE.
// See /usr/include/net/if.h.
struct if_msghdr8 {
u_short ifm_msglen;
u_char ifm_version;
u_char ifm_type;
int ifm_addrs;
int ifm_flags;
u_short ifm_index;
struct if_data8 ifm_data;
};
*/
import "C"
// Machine characteristics
const (
SizeofPtr = C.sizeofPtr
SizeofShort = C.sizeof_short
SizeofInt = C.sizeof_int
SizeofLong = C.sizeof_long
SizeofLongLong = C.sizeof_longlong
)
// Basic types
type (
_C_short C.short
_C_int C.int
_C_long C.long
_C_long_long C.longlong
)
// Time
type Timespec C.struct_timespec
type Timeval C.struct_timeval
// Processes
type Rusage C.struct_rusage
type Rlimit C.struct_rlimit
type _Gid_t C.gid_t
// Files
const (
_statfsVersion = C.STATFS_VERSION
_dirblksiz = C.DIRBLKSIZ
)
type Stat_t C.struct_stat
type stat_freebsd11_t C.struct_freebsd11_stat
type Statfs_t C.struct_statfs
type statfs_freebsd11_t C.struct_freebsd11_statfs
type Flock_t C.struct_flock
type Dirent C.struct_dirent
type dirent_freebsd11 C.struct_freebsd11_dirent
type Fsid C.struct_fsid
// File system limits
const (
PathMax = C.PATH_MAX
)
// Advice to Fadvise
const (
FADV_NORMAL = C.POSIX_FADV_NORMAL
FADV_RANDOM = C.POSIX_FADV_RANDOM
FADV_SEQUENTIAL = C.POSIX_FADV_SEQUENTIAL
FADV_WILLNEED = C.POSIX_FADV_WILLNEED
FADV_DONTNEED = C.POSIX_FADV_DONTNEED
FADV_NOREUSE = C.POSIX_FADV_NOREUSE
)
// Sockets
type RawSockaddrInet4 C.struct_sockaddr_in
type RawSockaddrInet6 C.struct_sockaddr_in6
type RawSockaddrUnix C.struct_sockaddr_un
type RawSockaddrDatalink C.struct_sockaddr_dl
type RawSockaddr C.struct_sockaddr
type RawSockaddrAny C.struct_sockaddr_any
type _Socklen C.socklen_t
type Linger C.struct_linger
type Iovec C.struct_iovec
type IPMreq C.struct_ip_mreq
type IPMreqn C.struct_ip_mreqn
type IPv6Mreq C.struct_ipv6_mreq
type Msghdr C.struct_msghdr
type Cmsghdr C.struct_cmsghdr
type Inet6Pktinfo C.struct_in6_pktinfo
type IPv6MTUInfo C.struct_ip6_mtuinfo
type ICMPv6Filter C.struct_icmp6_filter
const (
SizeofSockaddrInet4 = C.sizeof_struct_sockaddr_in
SizeofSockaddrInet6 = C.sizeof_struct_sockaddr_in6
SizeofSockaddrAny = C.sizeof_struct_sockaddr_any
SizeofSockaddrUnix = C.sizeof_struct_sockaddr_un
SizeofSockaddrDatalink = C.sizeof_struct_sockaddr_dl
SizeofLinger = C.sizeof_struct_linger
SizeofIPMreq = C.sizeof_struct_ip_mreq
SizeofIPMreqn = C.sizeof_struct_ip_mreqn
SizeofIPv6Mreq = C.sizeof_struct_ipv6_mreq
SizeofMsghdr = C.sizeof_struct_msghdr
SizeofCmsghdr = C.sizeof_struct_cmsghdr
SizeofInet6Pktinfo = C.sizeof_struct_in6_pktinfo
SizeofIPv6MTUInfo = C.sizeof_struct_ip6_mtuinfo
SizeofICMPv6Filter = C.sizeof_struct_icmp6_filter
)
// Ptrace requests
const (
PTRACE_ATTACH = C.PT_ATTACH
PTRACE_CONT = C.PT_CONTINUE
PTRACE_DETACH = C.PT_DETACH
PTRACE_GETFPREGS = C.PT_GETFPREGS
PTRACE_GETFSBASE = C.PT_GETFSBASE
PTRACE_GETLWPLIST = C.PT_GETLWPLIST
PTRACE_GETNUMLWPS = C.PT_GETNUMLWPS
PTRACE_GETREGS = C.PT_GETREGS
PTRACE_GETXSTATE = C.PT_GETXSTATE
PTRACE_IO = C.PT_IO
PTRACE_KILL = C.PT_KILL
PTRACE_LWPEVENTS = C.PT_LWP_EVENTS
PTRACE_LWPINFO = C.PT_LWPINFO
PTRACE_SETFPREGS = C.PT_SETFPREGS
PTRACE_SETREGS = C.PT_SETREGS
PTRACE_SINGLESTEP = C.PT_STEP
PTRACE_TRACEME = C.PT_TRACE_ME
)
const (
PIOD_READ_D = C.PIOD_READ_D
PIOD_WRITE_D = C.PIOD_WRITE_D
PIOD_READ_I = C.PIOD_READ_I
PIOD_WRITE_I = C.PIOD_WRITE_I
)
const (
PL_FLAG_BORN = C.PL_FLAG_BORN
PL_FLAG_EXITED = C.PL_FLAG_EXITED
PL_FLAG_SI = C.PL_FLAG_SI
)
const (
TRAP_BRKPT = C.TRAP_BRKPT
TRAP_TRACE = C.TRAP_TRACE
)
type PtraceLwpInfoStruct C.struct_ptrace_lwpinfo
type __Siginfo C.struct___siginfo
type Sigset_t C.sigset_t
type Reg C.struct_reg
type FpReg C.struct_fpreg
type PtraceIoDesc C.struct_ptrace_io_desc
// Events (kqueue, kevent)
type Kevent_t C.struct_kevent_freebsd11
// Select
type FdSet C.fd_set
// Routing and interface messages
const (
sizeofIfMsghdr = C.sizeof_struct_if_msghdr
SizeofIfMsghdr = C.sizeof_struct_if_msghdr8
sizeofIfData = C.sizeof_struct_if_data
SizeofIfData = C.sizeof_struct_if_data8
SizeofIfaMsghdr = C.sizeof_struct_ifa_msghdr
SizeofIfmaMsghdr = C.sizeof_struct_ifma_msghdr
SizeofIfAnnounceMsghdr = C.sizeof_struct_if_announcemsghdr
SizeofRtMsghdr = C.sizeof_struct_rt_msghdr
SizeofRtMetrics = C.sizeof_struct_rt_metrics
)
type ifMsghdr C.struct_if_msghdr
type IfMsghdr C.struct_if_msghdr8
type ifData C.struct_if_data
type IfData C.struct_if_data8
type IfaMsghdr C.struct_ifa_msghdr
type IfmaMsghdr C.struct_ifma_msghdr
type IfAnnounceMsghdr C.struct_if_announcemsghdr
type RtMsghdr C.struct_rt_msghdr
type RtMetrics C.struct_rt_metrics
// Berkeley packet filter
const (
SizeofBpfVersion = C.sizeof_struct_bpf_version
SizeofBpfStat = C.sizeof_struct_bpf_stat
SizeofBpfZbuf = C.sizeof_struct_bpf_zbuf
SizeofBpfProgram = C.sizeof_struct_bpf_program
SizeofBpfInsn = C.sizeof_struct_bpf_insn
SizeofBpfHdr = C.sizeof_struct_bpf_hdr
SizeofBpfZbufHeader = C.sizeof_struct_bpf_zbuf_header
)
type BpfVersion C.struct_bpf_version
type BpfStat C.struct_bpf_stat
type BpfZbuf C.struct_bpf_zbuf
type BpfProgram C.struct_bpf_program
type BpfInsn C.struct_bpf_insn
type BpfHdr C.struct_bpf_hdr
type BpfZbufHeader C.struct_bpf_zbuf_header
// Terminal handling
type Termios C.struct_termios
type Winsize C.struct_winsize
// fchmodat-like syscalls.
const (
AT_FDCWD = C.AT_FDCWD
AT_REMOVEDIR = C.AT_REMOVEDIR
AT_SYMLINK_FOLLOW = C.AT_SYMLINK_FOLLOW
AT_SYMLINK_NOFOLLOW = C.AT_SYMLINK_NOFOLLOW
)
// poll
type PollFd C.struct_pollfd
const (
POLLERR = C.POLLERR
POLLHUP = C.POLLHUP
POLLIN = C.POLLIN
POLLINIGNEOF = C.POLLINIGNEOF
POLLNVAL = C.POLLNVAL
POLLOUT = C.POLLOUT
POLLPRI = C.POLLPRI
POLLRDBAND = C.POLLRDBAND
POLLRDNORM = C.POLLRDNORM
POLLWRBAND = C.POLLWRBAND
POLLWRNORM = C.POLLWRNORM
)
// Capabilities
type CapRights C.struct_cap_rights
// Uname
type Utsname C.struct_utsname
// Clockinfo
const SizeofClockinfo = C.sizeof_struct_clockinfo
type Clockinfo C.struct_clockinfo

View File

@@ -1,300 +0,0 @@
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
Input to cgo -godefs. See README.md
*/
// +godefs map struct_in_addr [4]byte /* in_addr */
// +godefs map struct_in6_addr [16]byte /* in6_addr */
package unix
/*
#define KERNEL
#include <dirent.h>
#include <fcntl.h>
#include <poll.h>
#include <signal.h>
#include <termios.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/param.h>
#include <sys/types.h>
#include <sys/event.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/ptrace.h>
#include <sys/resource.h>
#include <sys/select.h>
#include <sys/signal.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/statvfs.h>
#include <sys/sysctl.h>
#include <sys/time.h>
#include <sys/uio.h>
#include <sys/un.h>
#include <sys/utsname.h>
#include <sys/wait.h>
#include <net/bpf.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/icmp6.h>
#include <netinet/tcp.h>
enum {
sizeofPtr = sizeof(void*),
};
union sockaddr_all {
struct sockaddr s1; // this one gets used for fields
struct sockaddr_in s2; // these pad it out
struct sockaddr_in6 s3;
struct sockaddr_un s4;
struct sockaddr_dl s5;
};
struct sockaddr_any {
struct sockaddr addr;
char pad[sizeof(union sockaddr_all) - sizeof(struct sockaddr)];
};
*/
import "C"
// Machine characteristics
const (
SizeofPtr = C.sizeofPtr
SizeofShort = C.sizeof_short
SizeofInt = C.sizeof_int
SizeofLong = C.sizeof_long
SizeofLongLong = C.sizeof_longlong
)
// Basic types
type (
_C_short C.short
_C_int C.int
_C_long C.long
_C_long_long C.longlong
)
// Time
type Timespec C.struct_timespec
type Timeval C.struct_timeval
// Processes
type Rusage C.struct_rusage
type Rlimit C.struct_rlimit
type _Gid_t C.gid_t
// Files
type Stat_t C.struct_stat
type Statfs_t C.struct_statfs
type Statvfs_t C.struct_statvfs
type Flock_t C.struct_flock
type Dirent C.struct_dirent
type Fsid C.fsid_t
// File system limits
const (
PathMax = C.PATH_MAX
)
// Fstatvfs/Statvfs flags
const (
ST_WAIT = C.ST_WAIT
ST_NOWAIT = C.ST_NOWAIT
)
// Advice to Fadvise
const (
FADV_NORMAL = C.POSIX_FADV_NORMAL
FADV_RANDOM = C.POSIX_FADV_RANDOM
FADV_SEQUENTIAL = C.POSIX_FADV_SEQUENTIAL
FADV_WILLNEED = C.POSIX_FADV_WILLNEED
FADV_DONTNEED = C.POSIX_FADV_DONTNEED
FADV_NOREUSE = C.POSIX_FADV_NOREUSE
)
// Sockets
type RawSockaddrInet4 C.struct_sockaddr_in
type RawSockaddrInet6 C.struct_sockaddr_in6
type RawSockaddrUnix C.struct_sockaddr_un
type RawSockaddrDatalink C.struct_sockaddr_dl
type RawSockaddr C.struct_sockaddr
type RawSockaddrAny C.struct_sockaddr_any
type _Socklen C.socklen_t
type Linger C.struct_linger
type Iovec C.struct_iovec
type IPMreq C.struct_ip_mreq
type IPv6Mreq C.struct_ipv6_mreq
type Msghdr C.struct_msghdr
type Cmsghdr C.struct_cmsghdr
type Inet6Pktinfo C.struct_in6_pktinfo
type IPv6MTUInfo C.struct_ip6_mtuinfo
type ICMPv6Filter C.struct_icmp6_filter
const (
SizeofSockaddrInet4 = C.sizeof_struct_sockaddr_in
SizeofSockaddrInet6 = C.sizeof_struct_sockaddr_in6
SizeofSockaddrAny = C.sizeof_struct_sockaddr_any
SizeofSockaddrUnix = C.sizeof_struct_sockaddr_un
SizeofSockaddrDatalink = C.sizeof_struct_sockaddr_dl
SizeofLinger = C.sizeof_struct_linger
SizeofIPMreq = C.sizeof_struct_ip_mreq
SizeofIPv6Mreq = C.sizeof_struct_ipv6_mreq
SizeofMsghdr = C.sizeof_struct_msghdr
SizeofCmsghdr = C.sizeof_struct_cmsghdr
SizeofInet6Pktinfo = C.sizeof_struct_in6_pktinfo
SizeofIPv6MTUInfo = C.sizeof_struct_ip6_mtuinfo
SizeofICMPv6Filter = C.sizeof_struct_icmp6_filter
)
// Ptrace requests
const (
PTRACE_TRACEME = C.PT_TRACE_ME
PTRACE_CONT = C.PT_CONTINUE
PTRACE_KILL = C.PT_KILL
)
// Events (kqueue, kevent)
type Kevent_t C.struct_kevent
// Select
type FdSet C.fd_set
// Routing and interface messages
const (
SizeofIfMsghdr = C.sizeof_struct_if_msghdr
SizeofIfData = C.sizeof_struct_if_data
SizeofIfaMsghdr = C.sizeof_struct_ifa_msghdr
SizeofIfAnnounceMsghdr = C.sizeof_struct_if_announcemsghdr
SizeofRtMsghdr = C.sizeof_struct_rt_msghdr
SizeofRtMetrics = C.sizeof_struct_rt_metrics
)
type IfMsghdr C.struct_if_msghdr
type IfData C.struct_if_data
type IfaMsghdr C.struct_ifa_msghdr
type IfAnnounceMsghdr C.struct_if_announcemsghdr
type RtMsghdr C.struct_rt_msghdr
type RtMetrics C.struct_rt_metrics
type Mclpool C.struct_mclpool
// Berkeley packet filter
const (
SizeofBpfVersion = C.sizeof_struct_bpf_version
SizeofBpfStat = C.sizeof_struct_bpf_stat
SizeofBpfProgram = C.sizeof_struct_bpf_program
SizeofBpfInsn = C.sizeof_struct_bpf_insn
SizeofBpfHdr = C.sizeof_struct_bpf_hdr
)
type BpfVersion C.struct_bpf_version
type BpfStat C.struct_bpf_stat
type BpfProgram C.struct_bpf_program
type BpfInsn C.struct_bpf_insn
type BpfHdr C.struct_bpf_hdr
type BpfTimeval C.struct_bpf_timeval
// Terminal handling
type Termios C.struct_termios
type Winsize C.struct_winsize
type Ptmget C.struct_ptmget
// fchmodat-like syscalls.
const (
AT_FDCWD = C.AT_FDCWD
AT_SYMLINK_FOLLOW = C.AT_SYMLINK_FOLLOW
AT_SYMLINK_NOFOLLOW = C.AT_SYMLINK_NOFOLLOW
)
// poll
type PollFd C.struct_pollfd
const (
POLLERR = C.POLLERR
POLLHUP = C.POLLHUP
POLLIN = C.POLLIN
POLLNVAL = C.POLLNVAL
POLLOUT = C.POLLOUT
POLLPRI = C.POLLPRI
POLLRDBAND = C.POLLRDBAND
POLLRDNORM = C.POLLRDNORM
POLLWRBAND = C.POLLWRBAND
POLLWRNORM = C.POLLWRNORM
)
// Sysctl
type Sysctlnode C.struct_sysctlnode
// Uname
type Utsname C.struct_utsname
// Clockinfo
const SizeofClockinfo = C.sizeof_struct_clockinfo
type Clockinfo C.struct_clockinfo

View File

@@ -1,283 +0,0 @@
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
Input to cgo -godefs. See README.md
*/
// +godefs map struct_in_addr [4]byte /* in_addr */
// +godefs map struct_in6_addr [16]byte /* in6_addr */
package unix
/*
#define KERNEL
#include <dirent.h>
#include <fcntl.h>
#include <poll.h>
#include <signal.h>
#include <termios.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/param.h>
#include <sys/types.h>
#include <sys/event.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/ptrace.h>
#include <sys/resource.h>
#include <sys/select.h>
#include <sys/signal.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <sys/uio.h>
#include <sys/un.h>
#include <sys/utsname.h>
#include <sys/wait.h>
#include <uvm/uvmexp.h>
#include <net/bpf.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/icmp6.h>
#include <netinet/tcp.h>
enum {
sizeofPtr = sizeof(void*),
};
union sockaddr_all {
struct sockaddr s1; // this one gets used for fields
struct sockaddr_in s2; // these pad it out
struct sockaddr_in6 s3;
struct sockaddr_un s4;
struct sockaddr_dl s5;
};
struct sockaddr_any {
struct sockaddr addr;
char pad[sizeof(union sockaddr_all) - sizeof(struct sockaddr)];
};
*/
import "C"
// Machine characteristics
const (
SizeofPtr = C.sizeofPtr
SizeofShort = C.sizeof_short
SizeofInt = C.sizeof_int
SizeofLong = C.sizeof_long
SizeofLongLong = C.sizeof_longlong
)
// Basic types
type (
_C_short C.short
_C_int C.int
_C_long C.long
_C_long_long C.longlong
)
// Time
type Timespec C.struct_timespec
type Timeval C.struct_timeval
// Processes
type Rusage C.struct_rusage
type Rlimit C.struct_rlimit
type _Gid_t C.gid_t
// Files
type Stat_t C.struct_stat
type Statfs_t C.struct_statfs
type Flock_t C.struct_flock
type Dirent C.struct_dirent
type Fsid C.fsid_t
// File system limits
const (
PathMax = C.PATH_MAX
)
// Sockets
type RawSockaddrInet4 C.struct_sockaddr_in
type RawSockaddrInet6 C.struct_sockaddr_in6
type RawSockaddrUnix C.struct_sockaddr_un
type RawSockaddrDatalink C.struct_sockaddr_dl
type RawSockaddr C.struct_sockaddr
type RawSockaddrAny C.struct_sockaddr_any
type _Socklen C.socklen_t
type Linger C.struct_linger
type Iovec C.struct_iovec
type IPMreq C.struct_ip_mreq
type IPv6Mreq C.struct_ipv6_mreq
type Msghdr C.struct_msghdr
type Cmsghdr C.struct_cmsghdr
type Inet6Pktinfo C.struct_in6_pktinfo
type IPv6MTUInfo C.struct_ip6_mtuinfo
type ICMPv6Filter C.struct_icmp6_filter
const (
SizeofSockaddrInet4 = C.sizeof_struct_sockaddr_in
SizeofSockaddrInet6 = C.sizeof_struct_sockaddr_in6
SizeofSockaddrAny = C.sizeof_struct_sockaddr_any
SizeofSockaddrUnix = C.sizeof_struct_sockaddr_un
SizeofSockaddrDatalink = C.sizeof_struct_sockaddr_dl
SizeofLinger = C.sizeof_struct_linger
SizeofIPMreq = C.sizeof_struct_ip_mreq
SizeofIPv6Mreq = C.sizeof_struct_ipv6_mreq
SizeofMsghdr = C.sizeof_struct_msghdr
SizeofCmsghdr = C.sizeof_struct_cmsghdr
SizeofInet6Pktinfo = C.sizeof_struct_in6_pktinfo
SizeofIPv6MTUInfo = C.sizeof_struct_ip6_mtuinfo
SizeofICMPv6Filter = C.sizeof_struct_icmp6_filter
)
// Ptrace requests
const (
PTRACE_TRACEME = C.PT_TRACE_ME
PTRACE_CONT = C.PT_CONTINUE
PTRACE_KILL = C.PT_KILL
)
// Events (kqueue, kevent)
type Kevent_t C.struct_kevent
// Select
type FdSet C.fd_set
// Routing and interface messages
const (
SizeofIfMsghdr = C.sizeof_struct_if_msghdr
SizeofIfData = C.sizeof_struct_if_data
SizeofIfaMsghdr = C.sizeof_struct_ifa_msghdr
SizeofIfAnnounceMsghdr = C.sizeof_struct_if_announcemsghdr
SizeofRtMsghdr = C.sizeof_struct_rt_msghdr
SizeofRtMetrics = C.sizeof_struct_rt_metrics
)
type IfMsghdr C.struct_if_msghdr
type IfData C.struct_if_data
type IfaMsghdr C.struct_ifa_msghdr
type IfAnnounceMsghdr C.struct_if_announcemsghdr
type RtMsghdr C.struct_rt_msghdr
type RtMetrics C.struct_rt_metrics
type Mclpool C.struct_mclpool
// Berkeley packet filter
const (
SizeofBpfVersion = C.sizeof_struct_bpf_version
SizeofBpfStat = C.sizeof_struct_bpf_stat
SizeofBpfProgram = C.sizeof_struct_bpf_program
SizeofBpfInsn = C.sizeof_struct_bpf_insn
SizeofBpfHdr = C.sizeof_struct_bpf_hdr
)
type BpfVersion C.struct_bpf_version
type BpfStat C.struct_bpf_stat
type BpfProgram C.struct_bpf_program
type BpfInsn C.struct_bpf_insn
type BpfHdr C.struct_bpf_hdr
type BpfTimeval C.struct_bpf_timeval
// Terminal handling
type Termios C.struct_termios
type Winsize C.struct_winsize
// fchmodat-like syscalls.
const (
AT_FDCWD = C.AT_FDCWD
AT_SYMLINK_FOLLOW = C.AT_SYMLINK_FOLLOW
AT_SYMLINK_NOFOLLOW = C.AT_SYMLINK_NOFOLLOW
)
// poll
type PollFd C.struct_pollfd
const (
POLLERR = C.POLLERR
POLLHUP = C.POLLHUP
POLLIN = C.POLLIN
POLLNVAL = C.POLLNVAL
POLLOUT = C.POLLOUT
POLLPRI = C.POLLPRI
POLLRDBAND = C.POLLRDBAND
POLLRDNORM = C.POLLRDNORM
POLLWRBAND = C.POLLWRBAND
POLLWRNORM = C.POLLWRNORM
)
// Signal Sets
type Sigset_t C.sigset_t
// Uname
type Utsname C.struct_utsname
// Uvmexp
const SizeofUvmexp = C.sizeof_struct_uvmexp
type Uvmexp C.struct_uvmexp
// Clockinfo
const SizeofClockinfo = C.sizeof_struct_clockinfo
type Clockinfo C.struct_clockinfo

View File

@@ -1,269 +0,0 @@
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
/*
Input to cgo -godefs. See README.md
*/
// +godefs map struct_in_addr [4]byte /* in_addr */
// +godefs map struct_in6_addr [16]byte /* in6_addr */
package unix
/*
#define KERNEL
// These defines ensure that builds done on newer versions of Solaris are
// backwards-compatible with older versions of Solaris and
// OpenSolaris-based derivatives.
#define __USE_SUNOS_SOCKETS__ // msghdr
#define __USE_LEGACY_PROTOTYPES__ // iovec
#include <dirent.h>
#include <fcntl.h>
#include <netdb.h>
#include <limits.h>
#include <poll.h>
#include <signal.h>
#include <termios.h>
#include <termio.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/param.h>
#include <sys/resource.h>
#include <sys/select.h>
#include <sys/signal.h>
#include <sys/socket.h>
#include <sys/stat.h>
#include <sys/statvfs.h>
#include <sys/time.h>
#include <sys/times.h>
#include <sys/types.h>
#include <sys/utsname.h>
#include <sys/un.h>
#include <sys/wait.h>
#include <net/bpf.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/icmp6.h>
#include <netinet/tcp.h>
#include <ustat.h>
#include <utime.h>
enum {
sizeofPtr = sizeof(void*),
};
union sockaddr_all {
struct sockaddr s1; // this one gets used for fields
struct sockaddr_in s2; // these pad it out
struct sockaddr_in6 s3;
struct sockaddr_un s4;
struct sockaddr_dl s5;
};
struct sockaddr_any {
struct sockaddr addr;
char pad[sizeof(union sockaddr_all) - sizeof(struct sockaddr)];
};
*/
import "C"
// Machine characteristics
const (
SizeofPtr = C.sizeofPtr
SizeofShort = C.sizeof_short
SizeofInt = C.sizeof_int
SizeofLong = C.sizeof_long
SizeofLongLong = C.sizeof_longlong
PathMax = C.PATH_MAX
MaxHostNameLen = C.MAXHOSTNAMELEN
)
// Basic types
type (
_C_short C.short
_C_int C.int
_C_long C.long
_C_long_long C.longlong
)
// Time
type Timespec C.struct_timespec
type Timeval C.struct_timeval
type Timeval32 C.struct_timeval32
type Tms C.struct_tms
type Utimbuf C.struct_utimbuf
// Processes
type Rusage C.struct_rusage
type Rlimit C.struct_rlimit
type _Gid_t C.gid_t
// Files
type Stat_t C.struct_stat
type Flock_t C.struct_flock
type Dirent C.struct_dirent
// Filesystems
type _Fsblkcnt_t C.fsblkcnt_t
type Statvfs_t C.struct_statvfs
// Sockets
type RawSockaddrInet4 C.struct_sockaddr_in
type RawSockaddrInet6 C.struct_sockaddr_in6
type RawSockaddrUnix C.struct_sockaddr_un
type RawSockaddrDatalink C.struct_sockaddr_dl
type RawSockaddr C.struct_sockaddr
type RawSockaddrAny C.struct_sockaddr_any
type _Socklen C.socklen_t
type Linger C.struct_linger
type Iovec C.struct_iovec
type IPMreq C.struct_ip_mreq
type IPv6Mreq C.struct_ipv6_mreq
type Msghdr C.struct_msghdr
type Cmsghdr C.struct_cmsghdr
type Inet4Pktinfo C.struct_in_pktinfo
type Inet6Pktinfo C.struct_in6_pktinfo
type IPv6MTUInfo C.struct_ip6_mtuinfo
type ICMPv6Filter C.struct_icmp6_filter
const (
SizeofSockaddrInet4 = C.sizeof_struct_sockaddr_in
SizeofSockaddrInet6 = C.sizeof_struct_sockaddr_in6
SizeofSockaddrAny = C.sizeof_struct_sockaddr_any
SizeofSockaddrUnix = C.sizeof_struct_sockaddr_un
SizeofSockaddrDatalink = C.sizeof_struct_sockaddr_dl
SizeofLinger = C.sizeof_struct_linger
SizeofIPMreq = C.sizeof_struct_ip_mreq
SizeofIPv6Mreq = C.sizeof_struct_ipv6_mreq
SizeofMsghdr = C.sizeof_struct_msghdr
SizeofCmsghdr = C.sizeof_struct_cmsghdr
SizeofInet4Pktinfo = C.sizeof_struct_in_pktinfo
SizeofInet6Pktinfo = C.sizeof_struct_in6_pktinfo
SizeofIPv6MTUInfo = C.sizeof_struct_ip6_mtuinfo
SizeofICMPv6Filter = C.sizeof_struct_icmp6_filter
)
// Select
type FdSet C.fd_set
// Misc
type Utsname C.struct_utsname
type Ustat_t C.struct_ustat
const (
AT_FDCWD = C.AT_FDCWD
AT_SYMLINK_NOFOLLOW = C.AT_SYMLINK_NOFOLLOW
AT_SYMLINK_FOLLOW = C.AT_SYMLINK_FOLLOW
AT_REMOVEDIR = C.AT_REMOVEDIR
AT_EACCESS = C.AT_EACCESS
)
// Routing and interface messages
const (
SizeofIfMsghdr = C.sizeof_struct_if_msghdr
SizeofIfData = C.sizeof_struct_if_data
SizeofIfaMsghdr = C.sizeof_struct_ifa_msghdr
SizeofRtMsghdr = C.sizeof_struct_rt_msghdr
SizeofRtMetrics = C.sizeof_struct_rt_metrics
)
type IfMsghdr C.struct_if_msghdr
type IfData C.struct_if_data
type IfaMsghdr C.struct_ifa_msghdr
type RtMsghdr C.struct_rt_msghdr
type RtMetrics C.struct_rt_metrics
// Berkeley packet filter
const (
SizeofBpfVersion = C.sizeof_struct_bpf_version
SizeofBpfStat = C.sizeof_struct_bpf_stat
SizeofBpfProgram = C.sizeof_struct_bpf_program
SizeofBpfInsn = C.sizeof_struct_bpf_insn
SizeofBpfHdr = C.sizeof_struct_bpf_hdr
)
type BpfVersion C.struct_bpf_version
type BpfStat C.struct_bpf_stat
type BpfProgram C.struct_bpf_program
type BpfInsn C.struct_bpf_insn
type BpfTimeval C.struct_bpf_timeval
type BpfHdr C.struct_bpf_hdr
// Terminal handling
type Termios C.struct_termios
type Termio C.struct_termio
type Winsize C.struct_winsize
// poll
type PollFd C.struct_pollfd
const (
POLLERR = C.POLLERR
POLLHUP = C.POLLHUP
POLLIN = C.POLLIN
POLLNVAL = C.POLLNVAL
POLLOUT = C.POLLOUT
POLLPRI = C.POLLPRI
POLLRDBAND = C.POLLRDBAND
POLLRDNORM = C.POLLRDNORM
POLLWRBAND = C.POLLWRBAND
POLLWRNORM = C.POLLWRNORM
)

View File

@@ -1,556 +0,0 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
import (
"bufio"
"fmt"
"log"
"net/http"
"sort"
"strings"
"unicode/utf8"
"golang.org/x/text/encoding"
"golang.org/x/text/internal/gen"
)
const ascii = "\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" +
"\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" +
` !"#$%&'()*+,-./0123456789:;<=>?` +
`@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_` +
"`abcdefghijklmnopqrstuvwxyz{|}~\u007f"
var encodings = []struct {
name string
mib string
comment string
varName string
replacement byte
mapping string
}{
{
"IBM Code Page 037",
"IBM037",
"",
"CodePage037",
0x3f,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM037-2.1.2.ucm",
},
{
"IBM Code Page 437",
"PC8CodePage437",
"",
"CodePage437",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM437-2.1.2.ucm",
},
{
"IBM Code Page 850",
"PC850Multilingual",
"",
"CodePage850",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM850-2.1.2.ucm",
},
{
"IBM Code Page 852",
"PCp852",
"",
"CodePage852",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM852-2.1.2.ucm",
},
{
"IBM Code Page 855",
"IBM855",
"",
"CodePage855",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM855-2.1.2.ucm",
},
{
"Windows Code Page 858", // PC latin1 with Euro
"IBM00858",
"",
"CodePage858",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/windows-858-2000.ucm",
},
{
"IBM Code Page 860",
"IBM860",
"",
"CodePage860",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM860-2.1.2.ucm",
},
{
"IBM Code Page 862",
"PC862LatinHebrew",
"",
"CodePage862",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM862-2.1.2.ucm",
},
{
"IBM Code Page 863",
"IBM863",
"",
"CodePage863",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM863-2.1.2.ucm",
},
{
"IBM Code Page 865",
"IBM865",
"",
"CodePage865",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM865-2.1.2.ucm",
},
{
"IBM Code Page 866",
"IBM866",
"",
"CodePage866",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-ibm866.txt",
},
{
"IBM Code Page 1047",
"IBM1047",
"",
"CodePage1047",
0x3f,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/glibc-IBM1047-2.1.2.ucm",
},
{
"IBM Code Page 1140",
"IBM01140",
"",
"CodePage1140",
0x3f,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/ibm-1140_P100-1997.ucm",
},
{
"ISO 8859-1",
"ISOLatin1",
"",
"ISO8859_1",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/iso-8859_1-1998.ucm",
},
{
"ISO 8859-2",
"ISOLatin2",
"",
"ISO8859_2",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-2.txt",
},
{
"ISO 8859-3",
"ISOLatin3",
"",
"ISO8859_3",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-3.txt",
},
{
"ISO 8859-4",
"ISOLatin4",
"",
"ISO8859_4",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-4.txt",
},
{
"ISO 8859-5",
"ISOLatinCyrillic",
"",
"ISO8859_5",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-5.txt",
},
{
"ISO 8859-6",
"ISOLatinArabic",
"",
"ISO8859_6,ISO8859_6E,ISO8859_6I",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-6.txt",
},
{
"ISO 8859-7",
"ISOLatinGreek",
"",
"ISO8859_7",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-7.txt",
},
{
"ISO 8859-8",
"ISOLatinHebrew",
"",
"ISO8859_8,ISO8859_8E,ISO8859_8I",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-8.txt",
},
{
"ISO 8859-9",
"ISOLatin5",
"",
"ISO8859_9",
encoding.ASCIISub,
"http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/iso-8859_9-1999.ucm",
},
{
"ISO 8859-10",
"ISOLatin6",
"",
"ISO8859_10",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-10.txt",
},
{
"ISO 8859-13",
"ISO885913",
"",
"ISO8859_13",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-13.txt",
},
{
"ISO 8859-14",
"ISO885914",
"",
"ISO8859_14",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-14.txt",
},
{
"ISO 8859-15",
"ISO885915",
"",
"ISO8859_15",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-15.txt",
},
{
"ISO 8859-16",
"ISO885916",
"",
"ISO8859_16",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-iso-8859-16.txt",
},
{
"KOI8-R",
"KOI8R",
"",
"KOI8R",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-koi8-r.txt",
},
{
"KOI8-U",
"KOI8U",
"",
"KOI8U",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-koi8-u.txt",
},
{
"Macintosh",
"Macintosh",
"",
"Macintosh",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-macintosh.txt",
},
{
"Macintosh Cyrillic",
"MacintoshCyrillic",
"",
"MacintoshCyrillic",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-x-mac-cyrillic.txt",
},
{
"Windows 874",
"Windows874",
"",
"Windows874",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-874.txt",
},
{
"Windows 1250",
"Windows1250",
"",
"Windows1250",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1250.txt",
},
{
"Windows 1251",
"Windows1251",
"",
"Windows1251",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1251.txt",
},
{
"Windows 1252",
"Windows1252",
"",
"Windows1252",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1252.txt",
},
{
"Windows 1253",
"Windows1253",
"",
"Windows1253",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1253.txt",
},
{
"Windows 1254",
"Windows1254",
"",
"Windows1254",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1254.txt",
},
{
"Windows 1255",
"Windows1255",
"",
"Windows1255",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1255.txt",
},
{
"Windows 1256",
"Windows1256",
"",
"Windows1256",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1256.txt",
},
{
"Windows 1257",
"Windows1257",
"",
"Windows1257",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1257.txt",
},
{
"Windows 1258",
"Windows1258",
"",
"Windows1258",
encoding.ASCIISub,
"http://encoding.spec.whatwg.org/index-windows-1258.txt",
},
{
"X-User-Defined",
"XUserDefined",
"It is defined at http://encoding.spec.whatwg.org/#x-user-defined",
"XUserDefined",
encoding.ASCIISub,
ascii +
"\uf780\uf781\uf782\uf783\uf784\uf785\uf786\uf787" +
"\uf788\uf789\uf78a\uf78b\uf78c\uf78d\uf78e\uf78f" +
"\uf790\uf791\uf792\uf793\uf794\uf795\uf796\uf797" +
"\uf798\uf799\uf79a\uf79b\uf79c\uf79d\uf79e\uf79f" +
"\uf7a0\uf7a1\uf7a2\uf7a3\uf7a4\uf7a5\uf7a6\uf7a7" +
"\uf7a8\uf7a9\uf7aa\uf7ab\uf7ac\uf7ad\uf7ae\uf7af" +
"\uf7b0\uf7b1\uf7b2\uf7b3\uf7b4\uf7b5\uf7b6\uf7b7" +
"\uf7b8\uf7b9\uf7ba\uf7bb\uf7bc\uf7bd\uf7be\uf7bf" +
"\uf7c0\uf7c1\uf7c2\uf7c3\uf7c4\uf7c5\uf7c6\uf7c7" +
"\uf7c8\uf7c9\uf7ca\uf7cb\uf7cc\uf7cd\uf7ce\uf7cf" +
"\uf7d0\uf7d1\uf7d2\uf7d3\uf7d4\uf7d5\uf7d6\uf7d7" +
"\uf7d8\uf7d9\uf7da\uf7db\uf7dc\uf7dd\uf7de\uf7df" +
"\uf7e0\uf7e1\uf7e2\uf7e3\uf7e4\uf7e5\uf7e6\uf7e7" +
"\uf7e8\uf7e9\uf7ea\uf7eb\uf7ec\uf7ed\uf7ee\uf7ef" +
"\uf7f0\uf7f1\uf7f2\uf7f3\uf7f4\uf7f5\uf7f6\uf7f7" +
"\uf7f8\uf7f9\uf7fa\uf7fb\uf7fc\uf7fd\uf7fe\uf7ff",
},
}
func getWHATWG(url string) string {
res, err := http.Get(url)
if err != nil {
log.Fatalf("%q: Get: %v", url, err)
}
defer res.Body.Close()
mapping := make([]rune, 128)
for i := range mapping {
mapping[i] = '\ufffd'
}
scanner := bufio.NewScanner(res.Body)
for scanner.Scan() {
s := strings.TrimSpace(scanner.Text())
if s == "" || s[0] == '#' {
continue
}
x, y := 0, 0
if _, err := fmt.Sscanf(s, "%d\t0x%x", &x, &y); err != nil {
log.Fatalf("could not parse %q", s)
}
if x < 0 || 128 <= x {
log.Fatalf("code %d is out of range", x)
}
if 0x80 <= y && y < 0xa0 {
// We diverge from the WHATWG spec by mapping control characters
// in the range [0x80, 0xa0) to U+FFFD.
continue
}
mapping[x] = rune(y)
}
return ascii + string(mapping)
}
func getUCM(url string) string {
res, err := http.Get(url)
if err != nil {
log.Fatalf("%q: Get: %v", url, err)
}
defer res.Body.Close()
mapping := make([]rune, 256)
for i := range mapping {
mapping[i] = '\ufffd'
}
charsFound := 0
scanner := bufio.NewScanner(res.Body)
for scanner.Scan() {
s := strings.TrimSpace(scanner.Text())
if s == "" || s[0] == '#' {
continue
}
var c byte
var r rune
if _, err := fmt.Sscanf(s, `<U%x> \x%x |0`, &r, &c); err != nil {
continue
}
mapping[c] = r
charsFound++
}
if charsFound < 200 {
log.Fatalf("%q: only %d characters found (wrong page format?)", url, charsFound)
}
return string(mapping)
}
func main() {
mibs := map[string]bool{}
all := []string{}
w := gen.NewCodeWriter()
defer w.WriteGoFile("tables.go", "charmap")
printf := func(s string, a ...interface{}) { fmt.Fprintf(w, s, a...) }
printf("import (\n")
printf("\t\"golang.org/x/text/encoding\"\n")
printf("\t\"golang.org/x/text/encoding/internal/identifier\"\n")
printf(")\n\n")
for _, e := range encodings {
varNames := strings.Split(e.varName, ",")
all = append(all, varNames...)
varName := varNames[0]
switch {
case strings.HasPrefix(e.mapping, "http://encoding.spec.whatwg.org/"):
e.mapping = getWHATWG(e.mapping)
case strings.HasPrefix(e.mapping, "http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/"):
e.mapping = getUCM(e.mapping)
}
asciiSuperset, low := strings.HasPrefix(e.mapping, ascii), 0x00
if asciiSuperset {
low = 0x80
}
lvn := 1
if strings.HasPrefix(varName, "ISO") || strings.HasPrefix(varName, "KOI") {
lvn = 3
}
lowerVarName := strings.ToLower(varName[:lvn]) + varName[lvn:]
printf("// %s is the %s encoding.\n", varName, e.name)
if e.comment != "" {
printf("//\n// %s\n", e.comment)
}
printf("var %s *Charmap = &%s\n\nvar %s = Charmap{\nname: %q,\n",
varName, lowerVarName, lowerVarName, e.name)
if mibs[e.mib] {
log.Fatalf("MIB type %q declared multiple times.", e.mib)
}
printf("mib: identifier.%s,\n", e.mib)
printf("asciiSuperset: %t,\n", asciiSuperset)
printf("low: 0x%02x,\n", low)
printf("replacement: 0x%02x,\n", e.replacement)
printf("decode: [256]utf8Enc{\n")
i, backMapping := 0, map[rune]byte{}
for _, c := range e.mapping {
if _, ok := backMapping[c]; !ok && c != utf8.RuneError {
backMapping[c] = byte(i)
}
var buf [8]byte
n := utf8.EncodeRune(buf[:], c)
if n > 3 {
panic(fmt.Sprintf("rune %q (%U) is too long", c, c))
}
printf("{%d,[3]byte{0x%02x,0x%02x,0x%02x}},", n, buf[0], buf[1], buf[2])
if i%2 == 1 {
printf("\n")
}
i++
}
printf("},\n")
printf("encode: [256]uint32{\n")
encode := make([]uint32, 0, 256)
for c, i := range backMapping {
encode = append(encode, uint32(i)<<24|uint32(c))
}
sort.Sort(byRune(encode))
for len(encode) < cap(encode) {
encode = append(encode, encode[len(encode)-1])
}
for i, enc := range encode {
printf("0x%08x,", enc)
if i%8 == 7 {
printf("\n")
}
}
printf("},\n}\n")
// Add an estimate of the size of a single Charmap{} struct value, which
// includes two 256 elem arrays of 4 bytes and some extra fields, which
// align to 3 uint64s on 64-bit architectures.
w.Size += 2*4*256 + 3*8
}
// TODO: add proper line breaking.
printf("var listAll = []encoding.Encoding{\n%s,\n}\n\n", strings.Join(all, ",\n"))
}
type byRune []uint32
func (b byRune) Len() int { return len(b) }
func (b byRune) Less(i, j int) bool { return b[i]&0xffffff < b[j]&0xffffff }
func (b byRune) Swap(i, j int) { b[i], b[j] = b[j], b[i] }

View File

@@ -1,173 +0,0 @@
// Copyright 2015 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
import (
"bytes"
"encoding/json"
"fmt"
"log"
"strings"
"golang.org/x/text/internal/gen"
)
type group struct {
Encodings []struct {
Labels []string
Name string
}
}
func main() {
gen.Init()
r := gen.Open("https://encoding.spec.whatwg.org", "whatwg", "encodings.json")
var groups []group
if err := json.NewDecoder(r).Decode(&groups); err != nil {
log.Fatalf("Error reading encodings.json: %v", err)
}
w := &bytes.Buffer{}
fmt.Fprintln(w, "type htmlEncoding byte")
fmt.Fprintln(w, "const (")
for i, g := range groups {
for _, e := range g.Encodings {
key := strings.ToLower(e.Name)
name := consts[key]
if name == "" {
log.Fatalf("No const defined for %s.", key)
}
if i == 0 {
fmt.Fprintf(w, "%s htmlEncoding = iota\n", name)
} else {
fmt.Fprintf(w, "%s\n", name)
}
}
}
fmt.Fprintln(w, "numEncodings")
fmt.Fprint(w, ")\n\n")
fmt.Fprintln(w, "var canonical = [numEncodings]string{")
for _, g := range groups {
for _, e := range g.Encodings {
fmt.Fprintf(w, "%q,\n", strings.ToLower(e.Name))
}
}
fmt.Fprint(w, "}\n\n")
fmt.Fprintln(w, "var nameMap = map[string]htmlEncoding{")
for _, g := range groups {
for _, e := range g.Encodings {
for _, l := range e.Labels {
key := strings.ToLower(e.Name)
name := consts[key]
fmt.Fprintf(w, "%q: %s,\n", l, name)
}
}
}
fmt.Fprint(w, "}\n\n")
var tags []string
fmt.Fprintln(w, "var localeMap = []htmlEncoding{")
for _, loc := range locales {
tags = append(tags, loc.tag)
fmt.Fprintf(w, "%s, // %s \n", consts[loc.name], loc.tag)
}
fmt.Fprint(w, "}\n\n")
fmt.Fprintf(w, "const locales = %q\n", strings.Join(tags, " "))
gen.WriteGoFile("tables.go", "htmlindex", w.Bytes())
}
// consts maps canonical encoding name to internal constant.
var consts = map[string]string{
"utf-8": "utf8",
"ibm866": "ibm866",
"iso-8859-2": "iso8859_2",
"iso-8859-3": "iso8859_3",
"iso-8859-4": "iso8859_4",
"iso-8859-5": "iso8859_5",
"iso-8859-6": "iso8859_6",
"iso-8859-7": "iso8859_7",
"iso-8859-8": "iso8859_8",
"iso-8859-8-i": "iso8859_8I",
"iso-8859-10": "iso8859_10",
"iso-8859-13": "iso8859_13",
"iso-8859-14": "iso8859_14",
"iso-8859-15": "iso8859_15",
"iso-8859-16": "iso8859_16",
"koi8-r": "koi8r",
"koi8-u": "koi8u",
"macintosh": "macintosh",
"windows-874": "windows874",
"windows-1250": "windows1250",
"windows-1251": "windows1251",
"windows-1252": "windows1252",
"windows-1253": "windows1253",
"windows-1254": "windows1254",
"windows-1255": "windows1255",
"windows-1256": "windows1256",
"windows-1257": "windows1257",
"windows-1258": "windows1258",
"x-mac-cyrillic": "macintoshCyrillic",
"gbk": "gbk",
"gb18030": "gb18030",
// "hz-gb-2312": "hzgb2312", // Was removed from WhatWG
"big5": "big5",
"euc-jp": "eucjp",
"iso-2022-jp": "iso2022jp",
"shift_jis": "shiftJIS",
"euc-kr": "euckr",
"replacement": "replacement",
"utf-16be": "utf16be",
"utf-16le": "utf16le",
"x-user-defined": "xUserDefined",
}
// locales is taken from
// https://html.spec.whatwg.org/multipage/syntax.html#encoding-sniffing-algorithm.
var locales = []struct{ tag, name string }{
// The default value. Explicitly state latin to benefit from the exact
// script option, while still making 1252 the default encoding for languages
// written in Latin script.
{"und_Latn", "windows-1252"},
{"ar", "windows-1256"},
{"ba", "windows-1251"},
{"be", "windows-1251"},
{"bg", "windows-1251"},
{"cs", "windows-1250"},
{"el", "iso-8859-7"},
{"et", "windows-1257"},
{"fa", "windows-1256"},
{"he", "windows-1255"},
{"hr", "windows-1250"},
{"hu", "iso-8859-2"},
{"ja", "shift_jis"},
{"kk", "windows-1251"},
{"ko", "euc-kr"},
{"ku", "windows-1254"},
{"ky", "windows-1251"},
{"lt", "windows-1257"},
{"lv", "windows-1257"},
{"mk", "windows-1251"},
{"pl", "iso-8859-2"},
{"ru", "windows-1251"},
{"sah", "windows-1251"},
{"sk", "windows-1250"},
{"sl", "iso-8859-2"},
{"sr", "windows-1251"},
{"tg", "windows-1251"},
{"th", "windows-874"},
{"tr", "windows-1254"},
{"tt", "windows-1251"},
{"uk", "windows-1251"},
{"vi", "windows-1258"},
{"zh-hans", "gb18030"},
{"zh-hant", "big5"},
}

View File

@@ -1,142 +0,0 @@
// Copyright 2015 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
import (
"bytes"
"encoding/xml"
"fmt"
"io"
"log"
"strings"
"golang.org/x/text/internal/gen"
)
type registry struct {
XMLName xml.Name `xml:"registry"`
Updated string `xml:"updated"`
Registry []struct {
ID string `xml:"id,attr"`
Record []struct {
Name string `xml:"name"`
Xref []struct {
Type string `xml:"type,attr"`
Data string `xml:"data,attr"`
} `xml:"xref"`
Desc struct {
Data string `xml:",innerxml"`
// Any []struct {
// Data string `xml:",chardata"`
// } `xml:",any"`
// Data string `xml:",chardata"`
} `xml:"description,"`
MIB string `xml:"value"`
Alias []string `xml:"alias"`
MIME string `xml:"preferred_alias"`
} `xml:"record"`
} `xml:"registry"`
}
func main() {
r := gen.OpenIANAFile("assignments/character-sets/character-sets.xml")
reg := &registry{}
if err := xml.NewDecoder(r).Decode(&reg); err != nil && err != io.EOF {
log.Fatalf("Error decoding charset registry: %v", err)
}
if len(reg.Registry) == 0 || reg.Registry[0].ID != "character-sets-1" {
log.Fatalf("Unexpected ID %s", reg.Registry[0].ID)
}
w := &bytes.Buffer{}
fmt.Fprintf(w, "const (\n")
for _, rec := range reg.Registry[0].Record {
constName := ""
for _, a := range rec.Alias {
if strings.HasPrefix(a, "cs") && strings.IndexByte(a, '-') == -1 {
// Some of the constant definitions have comments in them. Strip those.
constName = strings.Title(strings.SplitN(a[2:], "\n", 2)[0])
}
}
if constName == "" {
switch rec.MIB {
case "2085":
constName = "HZGB2312" // Not listed as alias for some reason.
default:
log.Fatalf("No cs alias defined for %s.", rec.MIB)
}
}
if rec.MIME != "" {
rec.MIME = fmt.Sprintf(" (MIME: %s)", rec.MIME)
}
fmt.Fprintf(w, "// %s is the MIB identifier with IANA name %s%s.\n//\n", constName, rec.Name, rec.MIME)
if len(rec.Desc.Data) > 0 {
fmt.Fprint(w, "// ")
d := xml.NewDecoder(strings.NewReader(rec.Desc.Data))
inElem := true
attr := ""
for {
t, err := d.Token()
if err != nil {
if err != io.EOF {
log.Fatal(err)
}
break
}
switch x := t.(type) {
case xml.CharData:
attr = "" // Don't need attribute info.
a := bytes.Split([]byte(x), []byte("\n"))
for i, b := range a {
if b = bytes.TrimSpace(b); len(b) != 0 {
if !inElem && i > 0 {
fmt.Fprint(w, "\n// ")
}
inElem = false
fmt.Fprintf(w, "%s ", string(b))
}
}
case xml.StartElement:
if x.Name.Local == "xref" {
inElem = true
use := false
for _, a := range x.Attr {
if a.Name.Local == "type" {
use = use || a.Value != "person"
}
if a.Name.Local == "data" && use {
// Patch up URLs to use https. From some links, the
// https version is different from the http one.
s := a.Value
s = strings.Replace(s, "http://", "https://", -1)
s = strings.Replace(s, "/unicode/", "/", -1)
attr = s + " "
}
}
}
case xml.EndElement:
inElem = false
fmt.Fprint(w, attr)
}
}
fmt.Fprint(w, "\n")
}
for _, x := range rec.Xref {
switch x.Type {
case "rfc":
fmt.Fprintf(w, "// Reference: %s\n", strings.ToUpper(x.Data))
case "uri":
fmt.Fprintf(w, "// Reference: %s\n", x.Data)
}
}
fmt.Fprintf(w, "%s MIB = %s\n", constName, rec.MIB)
fmt.Fprintln(w)
}
fmt.Fprintln(w, ")")
gen.WriteGoFile("mib.go", "identifier", w.Bytes())
}

View File

@@ -1,161 +0,0 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
// This program generates tables.go:
// go run maketables.go | gofmt > tables.go
// TODO: Emoji extensions?
// https://www.unicode.org/faq/emoji_dingbats.html
// https://www.unicode.org/Public/UNIDATA/EmojiSources.txt
import (
"bufio"
"fmt"
"log"
"net/http"
"sort"
"strings"
)
type entry struct {
jisCode, table int
}
func main() {
fmt.Printf("// generated by go run maketables.go; DO NOT EDIT\n\n")
fmt.Printf("// Package japanese provides Japanese encodings such as EUC-JP and Shift JIS.\n")
fmt.Printf(`package japanese // import "golang.org/x/text/encoding/japanese"` + "\n\n")
reverse := [65536]entry{}
for i := range reverse {
reverse[i].table = -1
}
tables := []struct {
url string
name string
}{
{"http://encoding.spec.whatwg.org/index-jis0208.txt", "0208"},
{"http://encoding.spec.whatwg.org/index-jis0212.txt", "0212"},
}
for i, table := range tables {
res, err := http.Get(table.url)
if err != nil {
log.Fatalf("%q: Get: %v", table.url, err)
}
defer res.Body.Close()
mapping := [65536]uint16{}
scanner := bufio.NewScanner(res.Body)
for scanner.Scan() {
s := strings.TrimSpace(scanner.Text())
if s == "" || s[0] == '#' {
continue
}
x, y := 0, uint16(0)
if _, err := fmt.Sscanf(s, "%d 0x%x", &x, &y); err != nil {
log.Fatalf("%q: could not parse %q", table.url, s)
}
if x < 0 || 120*94 <= x {
log.Fatalf("%q: JIS code %d is out of range", table.url, x)
}
mapping[x] = y
if reverse[y].table == -1 {
reverse[y] = entry{jisCode: x, table: i}
}
}
if err := scanner.Err(); err != nil {
log.Fatalf("%q: scanner error: %v", table.url, err)
}
fmt.Printf("// jis%sDecode is the decoding table from JIS %s code to Unicode.\n// It is defined at %s\n",
table.name, table.name, table.url)
fmt.Printf("var jis%sDecode = [...]uint16{\n", table.name)
for i, m := range mapping {
if m != 0 {
fmt.Printf("\t%d: 0x%04X,\n", i, m)
}
}
fmt.Printf("}\n\n")
}
// Any run of at least separation continuous zero entries in the reverse map will
// be a separate encode table.
const separation = 1024
intervals := []interval(nil)
low, high := -1, -1
for i, v := range reverse {
if v.table == -1 {
continue
}
if low < 0 {
low = i
} else if i-high >= separation {
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
low = i
}
high = i + 1
}
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
sort.Sort(byDecreasingLength(intervals))
fmt.Printf("const (\n")
fmt.Printf("\tjis0208 = 1\n")
fmt.Printf("\tjis0212 = 2\n")
fmt.Printf("\tcodeMask = 0x7f\n")
fmt.Printf("\tcodeShift = 7\n")
fmt.Printf("\ttableShift = 14\n")
fmt.Printf(")\n\n")
fmt.Printf("const numEncodeTables = %d\n\n", len(intervals))
fmt.Printf("// encodeX are the encoding tables from Unicode to JIS code,\n")
fmt.Printf("// sorted by decreasing length.\n")
for i, v := range intervals {
fmt.Printf("// encode%d: %5d entries for runes in [%5d, %5d).\n", i, v.len(), v.low, v.high)
}
fmt.Printf("//\n")
fmt.Printf("// The high two bits of the value record whether the JIS code comes from the\n")
fmt.Printf("// JIS0208 table (high bits == 1) or the JIS0212 table (high bits == 2).\n")
fmt.Printf("// The low 14 bits are two 7-bit unsigned integers j1 and j2 that form the\n")
fmt.Printf("// JIS code (94*j1 + j2) within that table.\n")
fmt.Printf("\n")
for i, v := range intervals {
fmt.Printf("const encode%dLow, encode%dHigh = %d, %d\n\n", i, i, v.low, v.high)
fmt.Printf("var encode%d = [...]uint16{\n", i)
for j := v.low; j < v.high; j++ {
x := reverse[j]
if x.table == -1 {
continue
}
fmt.Printf("\t%d - %d: jis%s<<14 | 0x%02X<<7 | 0x%02X,\n",
j, v.low, tables[x.table].name, x.jisCode/94, x.jisCode%94)
}
fmt.Printf("}\n\n")
}
}
// interval is a half-open interval [low, high).
type interval struct {
low, high int
}
func (i interval) len() int { return i.high - i.low }
// byDecreasingLength sorts intervals by decreasing length.
type byDecreasingLength []interval
func (b byDecreasingLength) Len() int { return len(b) }
func (b byDecreasingLength) Less(i, j int) bool { return b[i].len() > b[j].len() }
func (b byDecreasingLength) Swap(i, j int) { b[i], b[j] = b[j], b[i] }

View File

@@ -1,143 +0,0 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
// This program generates tables.go:
// go run maketables.go | gofmt > tables.go
import (
"bufio"
"fmt"
"log"
"net/http"
"sort"
"strings"
)
func main() {
fmt.Printf("// generated by go run maketables.go; DO NOT EDIT\n\n")
fmt.Printf("// Package korean provides Korean encodings such as EUC-KR.\n")
fmt.Printf(`package korean // import "golang.org/x/text/encoding/korean"` + "\n\n")
res, err := http.Get("http://encoding.spec.whatwg.org/index-euc-kr.txt")
if err != nil {
log.Fatalf("Get: %v", err)
}
defer res.Body.Close()
mapping := [65536]uint16{}
reverse := [65536]uint16{}
scanner := bufio.NewScanner(res.Body)
for scanner.Scan() {
s := strings.TrimSpace(scanner.Text())
if s == "" || s[0] == '#' {
continue
}
x, y := uint16(0), uint16(0)
if _, err := fmt.Sscanf(s, "%d 0x%x", &x, &y); err != nil {
log.Fatalf("could not parse %q", s)
}
if x < 0 || 178*(0xc7-0x81)+(0xfe-0xc7)*94+(0xff-0xa1) <= x {
log.Fatalf("EUC-KR code %d is out of range", x)
}
mapping[x] = y
if reverse[y] == 0 {
c0, c1 := uint16(0), uint16(0)
if x < 178*(0xc7-0x81) {
c0 = uint16(x/178) + 0x81
c1 = uint16(x % 178)
switch {
case c1 < 1*26:
c1 += 0x41
case c1 < 2*26:
c1 += 0x47
default:
c1 += 0x4d
}
} else {
x -= 178 * (0xc7 - 0x81)
c0 = uint16(x/94) + 0xc7
c1 = uint16(x%94) + 0xa1
}
reverse[y] = c0<<8 | c1
}
}
if err := scanner.Err(); err != nil {
log.Fatalf("scanner error: %v", err)
}
fmt.Printf("// decode is the decoding table from EUC-KR code to Unicode.\n")
fmt.Printf("// It is defined at http://encoding.spec.whatwg.org/index-euc-kr.txt\n")
fmt.Printf("var decode = [...]uint16{\n")
for i, v := range mapping {
if v != 0 {
fmt.Printf("\t%d: 0x%04X,\n", i, v)
}
}
fmt.Printf("}\n\n")
// Any run of at least separation continuous zero entries in the reverse map will
// be a separate encode table.
const separation = 1024
intervals := []interval(nil)
low, high := -1, -1
for i, v := range reverse {
if v == 0 {
continue
}
if low < 0 {
low = i
} else if i-high >= separation {
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
low = i
}
high = i + 1
}
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
sort.Sort(byDecreasingLength(intervals))
fmt.Printf("const numEncodeTables = %d\n\n", len(intervals))
fmt.Printf("// encodeX are the encoding tables from Unicode to EUC-KR code,\n")
fmt.Printf("// sorted by decreasing length.\n")
for i, v := range intervals {
fmt.Printf("// encode%d: %5d entries for runes in [%5d, %5d).\n", i, v.len(), v.low, v.high)
}
fmt.Printf("\n")
for i, v := range intervals {
fmt.Printf("const encode%dLow, encode%dHigh = %d, %d\n\n", i, i, v.low, v.high)
fmt.Printf("var encode%d = [...]uint16{\n", i)
for j := v.low; j < v.high; j++ {
x := reverse[j]
if x == 0 {
continue
}
fmt.Printf("\t%d-%d: 0x%04X,\n", j, v.low, x)
}
fmt.Printf("}\n\n")
}
}
// interval is a half-open interval [low, high).
type interval struct {
low, high int
}
func (i interval) len() int { return i.high - i.low }
// byDecreasingLength sorts intervals by decreasing length.
type byDecreasingLength []interval
func (b byDecreasingLength) Len() int { return len(b) }
func (b byDecreasingLength) Less(i, j int) bool { return b[i].len() > b[j].len() }
func (b byDecreasingLength) Swap(i, j int) { b[i], b[j] = b[j], b[i] }

View File

@@ -1,161 +0,0 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
// This program generates tables.go:
// go run maketables.go | gofmt > tables.go
import (
"bufio"
"fmt"
"log"
"net/http"
"sort"
"strings"
)
func main() {
fmt.Printf("// generated by go run maketables.go; DO NOT EDIT\n\n")
fmt.Printf("// Package simplifiedchinese provides Simplified Chinese encodings such as GBK.\n")
fmt.Printf(`package simplifiedchinese // import "golang.org/x/text/encoding/simplifiedchinese"` + "\n\n")
printGB18030()
printGBK()
}
func printGB18030() {
res, err := http.Get("http://encoding.spec.whatwg.org/index-gb18030.txt")
if err != nil {
log.Fatalf("Get: %v", err)
}
defer res.Body.Close()
fmt.Printf("// gb18030 is the table from http://encoding.spec.whatwg.org/index-gb18030.txt\n")
fmt.Printf("var gb18030 = [...][2]uint16{\n")
scanner := bufio.NewScanner(res.Body)
for scanner.Scan() {
s := strings.TrimSpace(scanner.Text())
if s == "" || s[0] == '#' {
continue
}
x, y := uint32(0), uint32(0)
if _, err := fmt.Sscanf(s, "%d 0x%x", &x, &y); err != nil {
log.Fatalf("could not parse %q", s)
}
if x < 0x10000 && y < 0x10000 {
fmt.Printf("\t{0x%04x, 0x%04x},\n", x, y)
}
}
fmt.Printf("}\n\n")
}
func printGBK() {
res, err := http.Get("http://encoding.spec.whatwg.org/index-gbk.txt")
if err != nil {
log.Fatalf("Get: %v", err)
}
defer res.Body.Close()
mapping := [65536]uint16{}
reverse := [65536]uint16{}
scanner := bufio.NewScanner(res.Body)
for scanner.Scan() {
s := strings.TrimSpace(scanner.Text())
if s == "" || s[0] == '#' {
continue
}
x, y := uint16(0), uint16(0)
if _, err := fmt.Sscanf(s, "%d 0x%x", &x, &y); err != nil {
log.Fatalf("could not parse %q", s)
}
if x < 0 || 126*190 <= x {
log.Fatalf("GBK code %d is out of range", x)
}
mapping[x] = y
if reverse[y] == 0 {
c0, c1 := x/190, x%190
if c1 >= 0x3f {
c1++
}
reverse[y] = (0x81+c0)<<8 | (0x40 + c1)
}
}
if err := scanner.Err(); err != nil {
log.Fatalf("scanner error: %v", err)
}
fmt.Printf("// decode is the decoding table from GBK code to Unicode.\n")
fmt.Printf("// It is defined at http://encoding.spec.whatwg.org/index-gbk.txt\n")
fmt.Printf("var decode = [...]uint16{\n")
for i, v := range mapping {
if v != 0 {
fmt.Printf("\t%d: 0x%04X,\n", i, v)
}
}
fmt.Printf("}\n\n")
// Any run of at least separation continuous zero entries in the reverse map will
// be a separate encode table.
const separation = 1024
intervals := []interval(nil)
low, high := -1, -1
for i, v := range reverse {
if v == 0 {
continue
}
if low < 0 {
low = i
} else if i-high >= separation {
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
low = i
}
high = i + 1
}
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
sort.Sort(byDecreasingLength(intervals))
fmt.Printf("const numEncodeTables = %d\n\n", len(intervals))
fmt.Printf("// encodeX are the encoding tables from Unicode to GBK code,\n")
fmt.Printf("// sorted by decreasing length.\n")
for i, v := range intervals {
fmt.Printf("// encode%d: %5d entries for runes in [%5d, %5d).\n", i, v.len(), v.low, v.high)
}
fmt.Printf("\n")
for i, v := range intervals {
fmt.Printf("const encode%dLow, encode%dHigh = %d, %d\n\n", i, i, v.low, v.high)
fmt.Printf("var encode%d = [...]uint16{\n", i)
for j := v.low; j < v.high; j++ {
x := reverse[j]
if x == 0 {
continue
}
fmt.Printf("\t%d-%d: 0x%04X,\n", j, v.low, x)
}
fmt.Printf("}\n\n")
}
}
// interval is a half-open interval [low, high).
type interval struct {
low, high int
}
func (i interval) len() int { return i.high - i.low }
// byDecreasingLength sorts intervals by decreasing length.
type byDecreasingLength []interval
func (b byDecreasingLength) Len() int { return len(b) }
func (b byDecreasingLength) Less(i, j int) bool { return b[i].len() > b[j].len() }
func (b byDecreasingLength) Swap(i, j int) { b[i], b[j] = b[j], b[i] }

View File

@@ -1,140 +0,0 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
// This program generates tables.go:
// go run maketables.go | gofmt > tables.go
import (
"bufio"
"fmt"
"log"
"net/http"
"sort"
"strings"
)
func main() {
fmt.Printf("// generated by go run maketables.go; DO NOT EDIT\n\n")
fmt.Printf("// Package traditionalchinese provides Traditional Chinese encodings such as Big5.\n")
fmt.Printf(`package traditionalchinese // import "golang.org/x/text/encoding/traditionalchinese"` + "\n\n")
res, err := http.Get("http://encoding.spec.whatwg.org/index-big5.txt")
if err != nil {
log.Fatalf("Get: %v", err)
}
defer res.Body.Close()
mapping := [65536]uint32{}
reverse := [65536 * 4]uint16{}
scanner := bufio.NewScanner(res.Body)
for scanner.Scan() {
s := strings.TrimSpace(scanner.Text())
if s == "" || s[0] == '#' {
continue
}
x, y := uint16(0), uint32(0)
if _, err := fmt.Sscanf(s, "%d 0x%x", &x, &y); err != nil {
log.Fatalf("could not parse %q", s)
}
if x < 0 || 126*157 <= x {
log.Fatalf("Big5 code %d is out of range", x)
}
mapping[x] = y
// The WHATWG spec http://encoding.spec.whatwg.org/#indexes says that
// "The index pointer for code point in index is the first pointer
// corresponding to code point in index", which would normally mean
// that the code below should be guarded by "if reverse[y] == 0", but
// last instead of first seems to match the behavior of
// "iconv -f UTF-8 -t BIG5". For example, U+8005 者 occurs twice in
// http://encoding.spec.whatwg.org/index-big5.txt, as index 2148
// (encoded as "\x8e\xcd") and index 6543 (encoded as "\xaa\xcc")
// and "echo 者 | iconv -f UTF-8 -t BIG5 | xxd" gives "\xaa\xcc".
c0, c1 := x/157, x%157
if c1 < 0x3f {
c1 += 0x40
} else {
c1 += 0x62
}
reverse[y] = (0x81+c0)<<8 | c1
}
if err := scanner.Err(); err != nil {
log.Fatalf("scanner error: %v", err)
}
fmt.Printf("// decode is the decoding table from Big5 code to Unicode.\n")
fmt.Printf("// It is defined at http://encoding.spec.whatwg.org/index-big5.txt\n")
fmt.Printf("var decode = [...]uint32{\n")
for i, v := range mapping {
if v != 0 {
fmt.Printf("\t%d: 0x%08X,\n", i, v)
}
}
fmt.Printf("}\n\n")
// Any run of at least separation continuous zero entries in the reverse map will
// be a separate encode table.
const separation = 1024
intervals := []interval(nil)
low, high := -1, -1
for i, v := range reverse {
if v == 0 {
continue
}
if low < 0 {
low = i
} else if i-high >= separation {
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
low = i
}
high = i + 1
}
if high >= 0 {
intervals = append(intervals, interval{low, high})
}
sort.Sort(byDecreasingLength(intervals))
fmt.Printf("const numEncodeTables = %d\n\n", len(intervals))
fmt.Printf("// encodeX are the encoding tables from Unicode to Big5 code,\n")
fmt.Printf("// sorted by decreasing length.\n")
for i, v := range intervals {
fmt.Printf("// encode%d: %5d entries for runes in [%6d, %6d).\n", i, v.len(), v.low, v.high)
}
fmt.Printf("\n")
for i, v := range intervals {
fmt.Printf("const encode%dLow, encode%dHigh = %d, %d\n\n", i, i, v.low, v.high)
fmt.Printf("var encode%d = [...]uint16{\n", i)
for j := v.low; j < v.high; j++ {
x := reverse[j]
if x == 0 {
continue
}
fmt.Printf("\t%d-%d: 0x%04X,\n", j, v.low, x)
}
fmt.Printf("}\n\n")
}
}
// interval is a half-open interval [low, high).
type interval struct {
low, high int
}
func (i interval) len() int { return i.high - i.low }
// byDecreasingLength sorts intervals by decreasing length.
type byDecreasingLength []interval
func (b byDecreasingLength) Len() int { return len(b) }
func (b byDecreasingLength) Less(i, j int) bool { return b[i].len() > b[j].len() }
func (b byDecreasingLength) Swap(i, j int) { b[i], b[j] = b[j], b[i] }

View File

@@ -1,64 +0,0 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// Language tag table generator.
// Data read from the web.
package main
import (
"flag"
"fmt"
"log"
"golang.org/x/text/internal/gen"
"golang.org/x/text/unicode/cldr"
)
var (
test = flag.Bool("test",
false,
"test existing tables; can be used to compare web data with package data.")
outputFile = flag.String("output",
"tables.go",
"output file for generated tables")
)
func main() {
gen.Init()
w := gen.NewCodeWriter()
defer w.WriteGoFile("tables.go", "compact")
fmt.Fprintln(w, `import "golang.org/x/text/internal/language"`)
b := newBuilder(w)
gen.WriteCLDRVersion(w)
b.writeCompactIndex()
}
type builder struct {
w *gen.CodeWriter
data *cldr.CLDR
supp *cldr.SupplementalData
}
func newBuilder(w *gen.CodeWriter) *builder {
r := gen.OpenCLDRCoreZip()
defer r.Close()
d := &cldr.Decoder{}
data, err := d.DecodeZip(r)
if err != nil {
log.Fatal(err)
}
b := builder{
w: w,
data: data,
supp: data.Supplemental(),
}
return &b
}

View File

@@ -1,113 +0,0 @@
// Copyright 2015 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
// This file generates derivative tables based on the language package itself.
import (
"fmt"
"log"
"sort"
"strings"
"golang.org/x/text/internal/language"
)
// Compact indices:
// Note -va-X variants only apply to localization variants.
// BCP variants only ever apply to language.
// The only ambiguity between tags is with regions.
func (b *builder) writeCompactIndex() {
// Collect all language tags for which we have any data in CLDR.
m := map[language.Tag]bool{}
for _, lang := range b.data.Locales() {
// We include all locales unconditionally to be consistent with en_US.
// We want en_US, even though it has no data associated with it.
// TODO: put any of the languages for which no data exists at the end
// of the index. This allows all components based on ICU to use that
// as the cutoff point.
// if x := data.RawLDML(lang); false ||
// x.LocaleDisplayNames != nil ||
// x.Characters != nil ||
// x.Delimiters != nil ||
// x.Measurement != nil ||
// x.Dates != nil ||
// x.Numbers != nil ||
// x.Units != nil ||
// x.ListPatterns != nil ||
// x.Collations != nil ||
// x.Segmentations != nil ||
// x.Rbnf != nil ||
// x.Annotations != nil ||
// x.Metadata != nil {
// TODO: support POSIX natively, albeit non-standard.
tag := language.Make(strings.Replace(lang, "_POSIX", "-u-va-posix", 1))
m[tag] = true
// }
}
// TODO: plural rules are also defined for the deprecated tags:
// iw mo sh tl
// Consider removing these as compact tags.
// Include locales for plural rules, which uses a different structure.
for _, plurals := range b.supp.Plurals {
for _, rules := range plurals.PluralRules {
for _, lang := range strings.Split(rules.Locales, " ") {
m[language.Make(lang)] = true
}
}
}
var coreTags []language.CompactCoreInfo
var special []string
for t := range m {
if x := t.Extensions(); len(x) != 0 && fmt.Sprint(x) != "[u-va-posix]" {
log.Fatalf("Unexpected extension %v in %v", x, t)
}
if len(t.Variants()) == 0 && len(t.Extensions()) == 0 {
cci, ok := language.GetCompactCore(t)
if !ok {
log.Fatalf("Locale for non-basic language %q", t)
}
coreTags = append(coreTags, cci)
} else {
special = append(special, t.String())
}
}
w := b.w
sort.Slice(coreTags, func(i, j int) bool { return coreTags[i] < coreTags[j] })
sort.Strings(special)
w.WriteComment(`
NumCompactTags is the number of common tags. The maximum tag is
NumCompactTags-1.`)
w.WriteConst("NumCompactTags", len(m))
fmt.Fprintln(w, "const (")
for i, t := range coreTags {
fmt.Fprintf(w, "%s ID = %d\n", ident(t.Tag().String()), i)
}
for i, t := range special {
fmt.Fprintf(w, "%s ID = %d\n", ident(t), i+len(coreTags))
}
fmt.Fprintln(w, ")")
w.WriteVar("coreTags", coreTags)
w.WriteConst("specialTagsStr", strings.Join(special, " "))
}
func ident(s string) string {
return strings.Replace(s, "-", "", -1) + "Index"
}

View File

@@ -1,54 +0,0 @@
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
import (
"log"
"golang.org/x/text/internal/gen"
"golang.org/x/text/internal/language"
"golang.org/x/text/internal/language/compact"
"golang.org/x/text/unicode/cldr"
)
func main() {
r := gen.OpenCLDRCoreZip()
defer r.Close()
d := &cldr.Decoder{}
data, err := d.DecodeZip(r)
if err != nil {
log.Fatalf("DecodeZip: %v", err)
}
w := gen.NewCodeWriter()
defer w.WriteGoFile("parents.go", "compact")
// Create parents table.
type ID uint16
parents := make([]ID, compact.NumCompactTags)
for _, loc := range data.Locales() {
tag := language.MustParse(loc)
index, ok := compact.FromTag(tag)
if !ok {
continue
}
parentIndex := compact.ID(0) // und
for p := tag.Parent(); p != language.Und; p = p.Parent() {
if x, ok := compact.FromTag(p); ok {
parentIndex = x
break
}
}
parents[index] = ID(parentIndex)
}
w.WriteComment(`
parents maps a compact index of a tag to the compact index of the parent of
this tag.`)
w.WriteVar("parents", parents)
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,20 +0,0 @@
// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
package main
// This file contains code common to the maketables.go and the package code.
// AliasType is the type of an alias in AliasMap.
type AliasType int8
const (
Deprecated AliasType = iota
Macro
Legacy
AliasTypeUnknown AliasType = -1
)

View File

@@ -1,305 +0,0 @@
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// Language tag table generator.
// Data read from the web.
package main
import (
"flag"
"fmt"
"io"
"log"
"sort"
"strconv"
"strings"
"golang.org/x/text/internal/gen"
"golang.org/x/text/internal/language"
"golang.org/x/text/unicode/cldr"
)
var (
test = flag.Bool("test",
false,
"test existing tables; can be used to compare web data with package data.")
outputFile = flag.String("output",
"tables.go",
"output file for generated tables")
)
func main() {
gen.Init()
w := gen.NewCodeWriter()
defer w.WriteGoFile("tables.go", "language")
b := newBuilder(w)
gen.WriteCLDRVersion(w)
b.writeConstants()
b.writeMatchData()
}
type builder struct {
w *gen.CodeWriter
hw io.Writer // MultiWriter for w and w.Hash
data *cldr.CLDR
supp *cldr.SupplementalData
}
func (b *builder) langIndex(s string) uint16 {
return uint16(language.MustParseBase(s))
}
func (b *builder) regionIndex(s string) int {
return int(language.MustParseRegion(s))
}
func (b *builder) scriptIndex(s string) int {
return int(language.MustParseScript(s))
}
func newBuilder(w *gen.CodeWriter) *builder {
r := gen.OpenCLDRCoreZip()
defer r.Close()
d := &cldr.Decoder{}
data, err := d.DecodeZip(r)
if err != nil {
log.Fatal(err)
}
b := builder{
w: w,
hw: io.MultiWriter(w, w.Hash),
data: data,
supp: data.Supplemental(),
}
return &b
}
// writeConsts computes f(v) for all v in values and writes the results
// as constants named _v to a single constant block.
func (b *builder) writeConsts(f func(string) int, values ...string) {
fmt.Fprintln(b.w, "const (")
for _, v := range values {
fmt.Fprintf(b.w, "\t_%s = %v\n", v, f(v))
}
fmt.Fprintln(b.w, ")")
}
// TODO: region inclusion data will probably not be use used in future matchers.
var langConsts = []string{
"de", "en", "fr", "it", "mo", "no", "nb", "pt", "sh", "mul", "und",
}
var scriptConsts = []string{
"Latn", "Hani", "Hans", "Hant", "Qaaa", "Qaai", "Qabx", "Zinh", "Zyyy",
"Zzzz",
}
var regionConsts = []string{
"001", "419", "BR", "CA", "ES", "GB", "MD", "PT", "UK", "US",
"ZZ", "XA", "XC", "XK", // Unofficial tag for Kosovo.
}
func (b *builder) writeConstants() {
b.writeConsts(func(s string) int { return int(b.langIndex(s)) }, langConsts...)
b.writeConsts(b.regionIndex, regionConsts...)
b.writeConsts(b.scriptIndex, scriptConsts...)
}
type mutualIntelligibility struct {
want, have uint16
distance uint8
oneway bool
}
type scriptIntelligibility struct {
wantLang, haveLang uint16
wantScript, haveScript uint8
distance uint8
// Always oneway
}
type regionIntelligibility struct {
lang uint16 // compact language id
script uint8 // 0 means any
group uint8 // 0 means any; if bit 7 is set it means inverse
distance uint8
// Always twoway.
}
// writeMatchData writes tables with languages and scripts for which there is
// mutual intelligibility. The data is based on CLDR's languageMatching data.
// Note that we use a different algorithm than the one defined by CLDR and that
// we slightly modify the data. For example, we convert scores to confidence levels.
// We also drop all region-related data as we use a different algorithm to
// determine region equivalence.
func (b *builder) writeMatchData() {
lm := b.supp.LanguageMatching.LanguageMatches
cldr.MakeSlice(&lm).SelectAnyOf("type", "written_new")
regionHierarchy := map[string][]string{}
for _, g := range b.supp.TerritoryContainment.Group {
regions := strings.Split(g.Contains, " ")
regionHierarchy[g.Type] = append(regionHierarchy[g.Type], regions...)
}
regionToGroups := make([]uint8, language.NumRegions)
idToIndex := map[string]uint8{}
for i, mv := range lm[0].MatchVariable {
if i > 6 {
log.Fatalf("Too many groups: %d", i)
}
idToIndex[mv.Id] = uint8(i + 1)
// TODO: also handle '-'
for _, r := range strings.Split(mv.Value, "+") {
todo := []string{r}
for k := 0; k < len(todo); k++ {
r := todo[k]
regionToGroups[b.regionIndex(r)] |= 1 << uint8(i)
todo = append(todo, regionHierarchy[r]...)
}
}
}
b.w.WriteVar("regionToGroups", regionToGroups)
// maps language id to in- and out-of-group region.
paradigmLocales := [][3]uint16{}
locales := strings.Split(lm[0].ParadigmLocales[0].Locales, " ")
for i := 0; i < len(locales); i += 2 {
x := [3]uint16{}
for j := 0; j < 2; j++ {
pc := strings.SplitN(locales[i+j], "-", 2)
x[0] = b.langIndex(pc[0])
if len(pc) == 2 {
x[1+j] = uint16(b.regionIndex(pc[1]))
}
}
paradigmLocales = append(paradigmLocales, x)
}
b.w.WriteVar("paradigmLocales", paradigmLocales)
b.w.WriteType(mutualIntelligibility{})
b.w.WriteType(scriptIntelligibility{})
b.w.WriteType(regionIntelligibility{})
matchLang := []mutualIntelligibility{}
matchScript := []scriptIntelligibility{}
matchRegion := []regionIntelligibility{}
// Convert the languageMatch entries in lists keyed by desired language.
for _, m := range lm[0].LanguageMatch {
// Different versions of CLDR use different separators.
desired := strings.Replace(m.Desired, "-", "_", -1)
supported := strings.Replace(m.Supported, "-", "_", -1)
d := strings.Split(desired, "_")
s := strings.Split(supported, "_")
if len(d) != len(s) {
log.Fatalf("not supported: desired=%q; supported=%q", desired, supported)
continue
}
distance, _ := strconv.ParseInt(m.Distance, 10, 8)
switch len(d) {
case 2:
if desired == supported && desired == "*_*" {
continue
}
// language-script pair.
matchScript = append(matchScript, scriptIntelligibility{
wantLang: uint16(b.langIndex(d[0])),
haveLang: uint16(b.langIndex(s[0])),
wantScript: uint8(b.scriptIndex(d[1])),
haveScript: uint8(b.scriptIndex(s[1])),
distance: uint8(distance),
})
if m.Oneway != "true" {
matchScript = append(matchScript, scriptIntelligibility{
wantLang: uint16(b.langIndex(s[0])),
haveLang: uint16(b.langIndex(d[0])),
wantScript: uint8(b.scriptIndex(s[1])),
haveScript: uint8(b.scriptIndex(d[1])),
distance: uint8(distance),
})
}
case 1:
if desired == supported && desired == "*" {
continue
}
if distance == 1 {
// nb == no is already handled by macro mapping. Check there
// really is only this case.
if d[0] != "no" || s[0] != "nb" {
log.Fatalf("unhandled equivalence %s == %s", s[0], d[0])
}
continue
}
// TODO: consider dropping oneway field and just doubling the entry.
matchLang = append(matchLang, mutualIntelligibility{
want: uint16(b.langIndex(d[0])),
have: uint16(b.langIndex(s[0])),
distance: uint8(distance),
oneway: m.Oneway == "true",
})
case 3:
if desired == supported && desired == "*_*_*" {
continue
}
if desired != supported {
// This is now supported by CLDR, but only one case, which
// should already be covered by paradigm locales. For instance,
// test case "und, en, en-GU, en-IN, en-GB ; en-ZA ; en-GB" in
// testdata/CLDRLocaleMatcherTest.txt tests this.
if supported != "en_*_GB" {
log.Fatalf("not supported: desired=%q; supported=%q", desired, supported)
}
continue
}
ri := regionIntelligibility{
lang: b.langIndex(d[0]),
distance: uint8(distance),
}
if d[1] != "*" {
ri.script = uint8(b.scriptIndex(d[1]))
}
switch {
case d[2] == "*":
ri.group = 0x80 // not contained in anything
case strings.HasPrefix(d[2], "$!"):
ri.group = 0x80
d[2] = "$" + d[2][len("$!"):]
fallthrough
case strings.HasPrefix(d[2], "$"):
ri.group |= idToIndex[d[2]]
}
matchRegion = append(matchRegion, ri)
default:
log.Fatalf("not supported: desired=%q; supported=%q", desired, supported)
}
}
sort.SliceStable(matchLang, func(i, j int) bool {
return matchLang[i].distance < matchLang[j].distance
})
b.w.WriteComment(`
matchLang holds pairs of langIDs of base languages that are typically
mutually intelligible. Each pair is associated with a confidence and
whether the intelligibility goes one or both ways.`)
b.w.WriteVar("matchLang", matchLang)
b.w.WriteComment(`
matchScript holds pairs of scriptIDs where readers of one script
can typically also read the other. Each is associated with a confidence.`)
sort.SliceStable(matchScript, func(i, j int) bool {
return matchScript[i].distance < matchScript[j].distance
})
b.w.WriteVar("matchScript", matchScript)
sort.SliceStable(matchRegion, func(i, j int) bool {
return matchRegion[i].distance < matchRegion[j].distance
})
b.w.WriteVar("matchRegion", matchRegion)
}

View File

@@ -1,986 +0,0 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// Normalization table generator.
// Data read from the web.
// See forminfo.go for a description of the trie values associated with each rune.
package main
import (
"bytes"
"encoding/binary"
"flag"
"fmt"
"io"
"log"
"sort"
"strconv"
"strings"
"golang.org/x/text/internal/gen"
"golang.org/x/text/internal/triegen"
"golang.org/x/text/internal/ucd"
)
func main() {
gen.Init()
loadUnicodeData()
compactCCC()
loadCompositionExclusions()
completeCharFields(FCanonical)
completeCharFields(FCompatibility)
computeNonStarterCounts()
verifyComputed()
printChars()
testDerived()
printTestdata()
makeTables()
}
var (
tablelist = flag.String("tables",
"all",
"comma-separated list of which tables to generate; "+
"can be 'decomp', 'recomp', 'info' and 'all'")
test = flag.Bool("test",
false,
"test existing tables against DerivedNormalizationProps and generate test data for regression testing")
verbose = flag.Bool("verbose",
false,
"write data to stdout as it is parsed")
)
const MaxChar = 0x10FFFF // anything above this shouldn't exist
// Quick Check properties of runes allow us to quickly
// determine whether a rune may occur in a normal form.
// For a given normal form, a rune may be guaranteed to occur
// verbatim (QC=Yes), may or may not combine with another
// rune (QC=Maybe), or may not occur (QC=No).
type QCResult int
const (
QCUnknown QCResult = iota
QCYes
QCNo
QCMaybe
)
func (r QCResult) String() string {
switch r {
case QCYes:
return "Yes"
case QCNo:
return "No"
case QCMaybe:
return "Maybe"
}
return "***UNKNOWN***"
}
const (
FCanonical = iota // NFC or NFD
FCompatibility // NFKC or NFKD
FNumberOfFormTypes
)
const (
MComposed = iota // NFC or NFKC
MDecomposed // NFD or NFKD
MNumberOfModes
)
// This contains only the properties we're interested in.
type Char struct {
name string
codePoint rune // if zero, this index is not a valid code point.
ccc uint8 // canonical combining class
origCCC uint8
excludeInComp bool // from CompositionExclusions.txt
compatDecomp bool // it has a compatibility expansion
nTrailingNonStarters uint8
nLeadingNonStarters uint8 // must be equal to trailing if non-zero
forms [FNumberOfFormTypes]FormInfo // For FCanonical and FCompatibility
state State
}
var chars = make([]Char, MaxChar+1)
var cccMap = make(map[uint8]uint8)
func (c Char) String() string {
buf := new(bytes.Buffer)
fmt.Fprintf(buf, "%U [%s]:\n", c.codePoint, c.name)
fmt.Fprintf(buf, " ccc: %v\n", c.ccc)
fmt.Fprintf(buf, " excludeInComp: %v\n", c.excludeInComp)
fmt.Fprintf(buf, " compatDecomp: %v\n", c.compatDecomp)
fmt.Fprintf(buf, " state: %v\n", c.state)
fmt.Fprintf(buf, " NFC:\n")
fmt.Fprint(buf, c.forms[FCanonical])
fmt.Fprintf(buf, " NFKC:\n")
fmt.Fprint(buf, c.forms[FCompatibility])
return buf.String()
}
// In UnicodeData.txt, some ranges are marked like this:
// 3400;<CJK Ideograph Extension A, First>;Lo;0;L;;;;;N;;;;;
// 4DB5;<CJK Ideograph Extension A, Last>;Lo;0;L;;;;;N;;;;;
// parseCharacter keeps a state variable indicating the weirdness.
type State int
const (
SNormal State = iota // known to be zero for the type
SFirst
SLast
SMissing
)
var lastChar = rune('\u0000')
func (c Char) isValid() bool {
return c.codePoint != 0 && c.state != SMissing
}
type FormInfo struct {
quickCheck [MNumberOfModes]QCResult // index: MComposed or MDecomposed
verified [MNumberOfModes]bool // index: MComposed or MDecomposed
combinesForward bool // May combine with rune on the right
combinesBackward bool // May combine with rune on the left
isOneWay bool // Never appears in result
inDecomp bool // Some decompositions result in this char.
decomp Decomposition
expandedDecomp Decomposition
}
func (f FormInfo) String() string {
buf := bytes.NewBuffer(make([]byte, 0))
fmt.Fprintf(buf, " quickCheck[C]: %v\n", f.quickCheck[MComposed])
fmt.Fprintf(buf, " quickCheck[D]: %v\n", f.quickCheck[MDecomposed])
fmt.Fprintf(buf, " cmbForward: %v\n", f.combinesForward)
fmt.Fprintf(buf, " cmbBackward: %v\n", f.combinesBackward)
fmt.Fprintf(buf, " isOneWay: %v\n", f.isOneWay)
fmt.Fprintf(buf, " inDecomp: %v\n", f.inDecomp)
fmt.Fprintf(buf, " decomposition: %X\n", f.decomp)
fmt.Fprintf(buf, " expandedDecomp: %X\n", f.expandedDecomp)
return buf.String()
}
type Decomposition []rune
func parseDecomposition(s string, skipfirst bool) (a []rune, err error) {
decomp := strings.Split(s, " ")
if len(decomp) > 0 && skipfirst {
decomp = decomp[1:]
}
for _, d := range decomp {
point, err := strconv.ParseUint(d, 16, 64)
if err != nil {
return a, err
}
a = append(a, rune(point))
}
return a, nil
}
func loadUnicodeData() {
f := gen.OpenUCDFile("UnicodeData.txt")
defer f.Close()
p := ucd.New(f)
for p.Next() {
r := p.Rune(ucd.CodePoint)
char := &chars[r]
char.ccc = uint8(p.Uint(ucd.CanonicalCombiningClass))
decmap := p.String(ucd.DecompMapping)
exp, err := parseDecomposition(decmap, false)
isCompat := false
if err != nil {
if len(decmap) > 0 {
exp, err = parseDecomposition(decmap, true)
if err != nil {
log.Fatalf(`%U: bad decomp |%v|: "%s"`, r, decmap, err)
}
isCompat = true
}
}
char.name = p.String(ucd.Name)
char.codePoint = r
char.forms[FCompatibility].decomp = exp
if !isCompat {
char.forms[FCanonical].decomp = exp
} else {
char.compatDecomp = true
}
if len(decmap) > 0 {
char.forms[FCompatibility].decomp = exp
}
}
if err := p.Err(); err != nil {
log.Fatal(err)
}
}
// compactCCC converts the sparse set of CCC values to a continguous one,
// reducing the number of bits needed from 8 to 6.
func compactCCC() {
m := make(map[uint8]uint8)
for i := range chars {
c := &chars[i]
m[c.ccc] = 0
}
cccs := []int{}
for v, _ := range m {
cccs = append(cccs, int(v))
}
sort.Ints(cccs)
for i, c := range cccs {
cccMap[uint8(i)] = uint8(c)
m[uint8(c)] = uint8(i)
}
for i := range chars {
c := &chars[i]
c.origCCC = c.ccc
c.ccc = m[c.ccc]
}
if len(m) >= 1<<6 {
log.Fatalf("too many difference CCC values: %d >= 64", len(m))
}
}
// CompositionExclusions.txt has form:
// 0958 # ...
// See https://unicode.org/reports/tr44/ for full explanation
func loadCompositionExclusions() {
f := gen.OpenUCDFile("CompositionExclusions.txt")
defer f.Close()
p := ucd.New(f)
for p.Next() {
c := &chars[p.Rune(0)]
if c.excludeInComp {
log.Fatalf("%U: Duplicate entry in exclusions.", c.codePoint)
}
c.excludeInComp = true
}
if e := p.Err(); e != nil {
log.Fatal(e)
}
}
// hasCompatDecomp returns true if any of the recursive
// decompositions contains a compatibility expansion.
// In this case, the character may not occur in NFK*.
func hasCompatDecomp(r rune) bool {
c := &chars[r]
if c.compatDecomp {
return true
}
for _, d := range c.forms[FCompatibility].decomp {
if hasCompatDecomp(d) {
return true
}
}
return false
}
// Hangul related constants.
const (
HangulBase = 0xAC00
HangulEnd = 0xD7A4 // hangulBase + Jamo combinations (19 * 21 * 28)
JamoLBase = 0x1100
JamoLEnd = 0x1113
JamoVBase = 0x1161
JamoVEnd = 0x1176
JamoTBase = 0x11A8
JamoTEnd = 0x11C3
JamoLVTCount = 19 * 21 * 28
JamoTCount = 28
)
func isHangul(r rune) bool {
return HangulBase <= r && r < HangulEnd
}
func isHangulWithoutJamoT(r rune) bool {
if !isHangul(r) {
return false
}
r -= HangulBase
return r < JamoLVTCount && r%JamoTCount == 0
}
func ccc(r rune) uint8 {
return chars[r].ccc
}
// Insert a rune in a buffer, ordered by Canonical Combining Class.
func insertOrdered(b Decomposition, r rune) Decomposition {
n := len(b)
b = append(b, 0)
cc := ccc(r)
if cc > 0 {
// Use bubble sort.
for ; n > 0; n-- {
if ccc(b[n-1]) <= cc {
break
}
b[n] = b[n-1]
}
}
b[n] = r
return b
}
// Recursively decompose.
func decomposeRecursive(form int, r rune, d Decomposition) Decomposition {
dcomp := chars[r].forms[form].decomp
if len(dcomp) == 0 {
return insertOrdered(d, r)
}
for _, c := range dcomp {
d = decomposeRecursive(form, c, d)
}
return d
}
func completeCharFields(form int) {
// Phase 0: pre-expand decomposition.
for i := range chars {
f := &chars[i].forms[form]
if len(f.decomp) == 0 {
continue
}
exp := make(Decomposition, 0)
for _, c := range f.decomp {
exp = decomposeRecursive(form, c, exp)
}
f.expandedDecomp = exp
}
// Phase 1: composition exclusion, mark decomposition.
for i := range chars {
c := &chars[i]
f := &c.forms[form]
// Marks script-specific exclusions and version restricted.
f.isOneWay = c.excludeInComp
// Singletons
f.isOneWay = f.isOneWay || len(f.decomp) == 1
// Non-starter decompositions
if len(f.decomp) > 1 {
chk := c.ccc != 0 || chars[f.decomp[0]].ccc != 0
f.isOneWay = f.isOneWay || chk
}
// Runes that decompose into more than two runes.
f.isOneWay = f.isOneWay || len(f.decomp) > 2
if form == FCompatibility {
f.isOneWay = f.isOneWay || hasCompatDecomp(c.codePoint)
}
for _, r := range f.decomp {
chars[r].forms[form].inDecomp = true
}
}
// Phase 2: forward and backward combining.
for i := range chars {
c := &chars[i]
f := &c.forms[form]
if !f.isOneWay && len(f.decomp) == 2 {
f0 := &chars[f.decomp[0]].forms[form]
f1 := &chars[f.decomp[1]].forms[form]
if !f0.isOneWay {
f0.combinesForward = true
}
if !f1.isOneWay {
f1.combinesBackward = true
}
}
if isHangulWithoutJamoT(rune(i)) {
f.combinesForward = true
}
}
// Phase 3: quick check values.
for i := range chars {
c := &chars[i]
f := &c.forms[form]
switch {
case len(f.decomp) > 0:
f.quickCheck[MDecomposed] = QCNo
case isHangul(rune(i)):
f.quickCheck[MDecomposed] = QCNo
default:
f.quickCheck[MDecomposed] = QCYes
}
switch {
case f.isOneWay:
f.quickCheck[MComposed] = QCNo
case (i & 0xffff00) == JamoLBase:
f.quickCheck[MComposed] = QCYes
if JamoLBase <= i && i < JamoLEnd {
f.combinesForward = true
}
if JamoVBase <= i && i < JamoVEnd {
f.quickCheck[MComposed] = QCMaybe
f.combinesBackward = true
f.combinesForward = true
}
if JamoTBase <= i && i < JamoTEnd {
f.quickCheck[MComposed] = QCMaybe
f.combinesBackward = true
}
case !f.combinesBackward:
f.quickCheck[MComposed] = QCYes
default:
f.quickCheck[MComposed] = QCMaybe
}
}
}
func computeNonStarterCounts() {
// Phase 4: leading and trailing non-starter count
for i := range chars {
c := &chars[i]
runes := []rune{rune(i)}
// We always use FCompatibility so that the CGJ insertion points do not
// change for repeated normalizations with different forms.
if exp := c.forms[FCompatibility].expandedDecomp; len(exp) > 0 {
runes = exp
}
// We consider runes that combine backwards to be non-starters for the
// purpose of Stream-Safe Text Processing.
for _, r := range runes {
if cr := &chars[r]; cr.ccc == 0 && !cr.forms[FCompatibility].combinesBackward {
break
}
c.nLeadingNonStarters++
}
for i := len(runes) - 1; i >= 0; i-- {
if cr := &chars[runes[i]]; cr.ccc == 0 && !cr.forms[FCompatibility].combinesBackward {
break
}
c.nTrailingNonStarters++
}
if c.nTrailingNonStarters > 3 {
log.Fatalf("%U: Decomposition with more than 3 (%d) trailing modifiers (%U)", i, c.nTrailingNonStarters, runes)
}
if isHangul(rune(i)) {
c.nTrailingNonStarters = 2
if isHangulWithoutJamoT(rune(i)) {
c.nTrailingNonStarters = 1
}
}
if l, t := c.nLeadingNonStarters, c.nTrailingNonStarters; l > 0 && l != t {
log.Fatalf("%U: number of leading and trailing non-starters should be equal (%d vs %d)", i, l, t)
}
if t := c.nTrailingNonStarters; t > 3 {
log.Fatalf("%U: number of trailing non-starters is %d > 3", t)
}
}
}
func printBytes(w io.Writer, b []byte, name string) {
fmt.Fprintf(w, "// %s: %d bytes\n", name, len(b))
fmt.Fprintf(w, "var %s = [...]byte {", name)
for i, c := range b {
switch {
case i%64 == 0:
fmt.Fprintf(w, "\n// Bytes %x - %x\n", i, i+63)
case i%8 == 0:
fmt.Fprintf(w, "\n")
}
fmt.Fprintf(w, "0x%.2X, ", c)
}
fmt.Fprint(w, "\n}\n\n")
}
// See forminfo.go for format.
func makeEntry(f *FormInfo, c *Char) uint16 {
e := uint16(0)
if r := c.codePoint; HangulBase <= r && r < HangulEnd {
e |= 0x40
}
if f.combinesForward {
e |= 0x20
}
if f.quickCheck[MDecomposed] == QCNo {
e |= 0x4
}
switch f.quickCheck[MComposed] {
case QCYes:
case QCNo:
e |= 0x10
case QCMaybe:
e |= 0x18
default:
log.Fatalf("Illegal quickcheck value %v.", f.quickCheck[MComposed])
}
e |= uint16(c.nTrailingNonStarters)
return e
}
// decompSet keeps track of unique decompositions, grouped by whether
// the decomposition is followed by a trailing and/or leading CCC.
type decompSet [7]map[string]bool
const (
normalDecomp = iota
firstMulti
firstCCC
endMulti
firstLeadingCCC
firstCCCZeroExcept
firstStarterWithNLead
lastDecomp
)
var cname = []string{"firstMulti", "firstCCC", "endMulti", "firstLeadingCCC", "firstCCCZeroExcept", "firstStarterWithNLead", "lastDecomp"}
func makeDecompSet() decompSet {
m := decompSet{}
for i := range m {
m[i] = make(map[string]bool)
}
return m
}
func (m *decompSet) insert(key int, s string) {
m[key][s] = true
}
func printCharInfoTables(w io.Writer) int {
mkstr := func(r rune, f *FormInfo) (int, string) {
d := f.expandedDecomp
s := string([]rune(d))
if max := 1 << 6; len(s) >= max {
const msg = "%U: too many bytes in decomposition: %d >= %d"
log.Fatalf(msg, r, len(s), max)
}
head := uint8(len(s))
if f.quickCheck[MComposed] != QCYes {
head |= 0x40
}
if f.combinesForward {
head |= 0x80
}
s = string([]byte{head}) + s
lccc := ccc(d[0])
tccc := ccc(d[len(d)-1])
cc := ccc(r)
if cc != 0 && lccc == 0 && tccc == 0 {
log.Fatalf("%U: trailing and leading ccc are 0 for non-zero ccc %d", r, cc)
}
if tccc < lccc && lccc != 0 {
const msg = "%U: lccc (%d) must be <= tcc (%d)"
log.Fatalf(msg, r, lccc, tccc)
}
index := normalDecomp
nTrail := chars[r].nTrailingNonStarters
nLead := chars[r].nLeadingNonStarters
if tccc > 0 || lccc > 0 || nTrail > 0 {
tccc <<= 2
tccc |= nTrail
s += string([]byte{tccc})
index = endMulti
for _, r := range d[1:] {
if ccc(r) == 0 {
index = firstCCC
}
}
if lccc > 0 || nLead > 0 {
s += string([]byte{lccc})
if index == firstCCC {
log.Fatalf("%U: multi-segment decomposition not supported for decompositions with leading CCC != 0", r)
}
index = firstLeadingCCC
}
if cc != lccc {
if cc != 0 {
log.Fatalf("%U: for lccc != ccc, expected ccc to be 0; was %d", r, cc)
}
index = firstCCCZeroExcept
}
} else if len(d) > 1 {
index = firstMulti
}
return index, s
}
decompSet := makeDecompSet()
const nLeadStr = "\x00\x01" // 0-byte length and tccc with nTrail.
decompSet.insert(firstStarterWithNLead, nLeadStr)
// Store the uniqued decompositions in a byte buffer,
// preceded by their byte length.
for _, c := range chars {
for _, f := range c.forms {
if len(f.expandedDecomp) == 0 {
continue
}
if f.combinesBackward {
log.Fatalf("%U: combinesBackward and decompose", c.codePoint)
}
index, s := mkstr(c.codePoint, &f)
decompSet.insert(index, s)
}
}
decompositions := bytes.NewBuffer(make([]byte, 0, 10000))
size := 0
positionMap := make(map[string]uint16)
decompositions.WriteString("\000")
fmt.Fprintln(w, "const (")
for i, m := range decompSet {
sa := []string{}
for s := range m {
sa = append(sa, s)
}
sort.Strings(sa)
for _, s := range sa {
p := decompositions.Len()
decompositions.WriteString(s)
positionMap[s] = uint16(p)
}
if cname[i] != "" {
fmt.Fprintf(w, "%s = 0x%X\n", cname[i], decompositions.Len())
}
}
fmt.Fprintln(w, "maxDecomp = 0x8000")
fmt.Fprintln(w, ")")
b := decompositions.Bytes()
printBytes(w, b, "decomps")
size += len(b)
varnames := []string{"nfc", "nfkc"}
for i := 0; i < FNumberOfFormTypes; i++ {
trie := triegen.NewTrie(varnames[i])
for r, c := range chars {
f := c.forms[i]
d := f.expandedDecomp
if len(d) != 0 {
_, key := mkstr(c.codePoint, &f)
trie.Insert(rune(r), uint64(positionMap[key]))
if c.ccc != ccc(d[0]) {
// We assume the lead ccc of a decomposition !=0 in this case.
if ccc(d[0]) == 0 {
log.Fatalf("Expected leading CCC to be non-zero; ccc is %d", c.ccc)
}
}
} else if c.nLeadingNonStarters > 0 && len(f.expandedDecomp) == 0 && c.ccc == 0 && !f.combinesBackward {
// Handle cases where it can't be detected that the nLead should be equal
// to nTrail.
trie.Insert(c.codePoint, uint64(positionMap[nLeadStr]))
} else if v := makeEntry(&f, &c)<<8 | uint16(c.ccc); v != 0 {
trie.Insert(c.codePoint, uint64(0x8000|v))
}
}
sz, err := trie.Gen(w, triegen.Compact(&normCompacter{name: varnames[i]}))
if err != nil {
log.Fatal(err)
}
size += sz
}
return size
}
func contains(sa []string, s string) bool {
for _, a := range sa {
if a == s {
return true
}
}
return false
}
func makeTables() {
w := &bytes.Buffer{}
size := 0
if *tablelist == "" {
return
}
list := strings.Split(*tablelist, ",")
if *tablelist == "all" {
list = []string{"recomp", "info"}
}
// Compute maximum decomposition size.
max := 0
for _, c := range chars {
if n := len(string(c.forms[FCompatibility].expandedDecomp)); n > max {
max = n
}
}
fmt.Fprintln(w, `import "sync"`)
fmt.Fprintln(w)
fmt.Fprintln(w, "const (")
fmt.Fprintln(w, "\t// Version is the Unicode edition from which the tables are derived.")
fmt.Fprintf(w, "\tVersion = %q\n", gen.UnicodeVersion())
fmt.Fprintln(w)
fmt.Fprintln(w, "\t// MaxTransformChunkSize indicates the maximum number of bytes that Transform")
fmt.Fprintln(w, "\t// may need to write atomically for any Form. Making a destination buffer at")
fmt.Fprintln(w, "\t// least this size ensures that Transform can always make progress and that")
fmt.Fprintln(w, "\t// the user does not need to grow the buffer on an ErrShortDst.")
fmt.Fprintf(w, "\tMaxTransformChunkSize = %d+maxNonStarters*4\n", len(string(0x034F))+max)
fmt.Fprintln(w, ")\n")
// Print the CCC remap table.
size += len(cccMap)
fmt.Fprintf(w, "var ccc = [%d]uint8{", len(cccMap))
for i := 0; i < len(cccMap); i++ {
if i%8 == 0 {
fmt.Fprintln(w)
}
fmt.Fprintf(w, "%3d, ", cccMap[uint8(i)])
}
fmt.Fprintln(w, "\n}\n")
if contains(list, "info") {
size += printCharInfoTables(w)
}
if contains(list, "recomp") {
// Note that we use 32 bit keys, instead of 64 bit.
// This clips the bits of three entries, but we know
// this won't cause a collision. The compiler will catch
// any changes made to UnicodeData.txt that introduces
// a collision.
// Note that the recomposition map for NFC and NFKC
// are identical.
// Recomposition map
nrentries := 0
for _, c := range chars {
f := c.forms[FCanonical]
if !f.isOneWay && len(f.decomp) > 0 {
nrentries++
}
}
sz := nrentries * 8
size += sz
fmt.Fprintf(w, "// recompMap: %d bytes (entries only)\n", sz)
fmt.Fprintln(w, "var recompMap map[uint32]rune")
fmt.Fprintln(w, "var recompMapOnce sync.Once\n")
fmt.Fprintln(w, `const recompMapPacked = "" +`)
var buf [8]byte
for i, c := range chars {
f := c.forms[FCanonical]
d := f.decomp
if !f.isOneWay && len(d) > 0 {
key := uint32(uint16(d[0]))<<16 + uint32(uint16(d[1]))
binary.BigEndian.PutUint32(buf[:4], key)
binary.BigEndian.PutUint32(buf[4:], uint32(i))
fmt.Fprintf(w, "\t\t%q + // 0x%.8X: 0x%.8X\n", string(buf[:]), key, uint32(i))
}
}
// hack so we don't have to special case the trailing plus sign
fmt.Fprintf(w, ` ""`)
fmt.Fprintln(w)
}
fmt.Fprintf(w, "// Total size of tables: %dKB (%d bytes)\n", (size+512)/1024, size)
gen.WriteVersionedGoFile("tables.go", "norm", w.Bytes())
}
func printChars() {
if *verbose {
for _, c := range chars {
if !c.isValid() || c.state == SMissing {
continue
}
fmt.Println(c)
}
}
}
// verifyComputed does various consistency tests.
func verifyComputed() {
for i, c := range chars {
for _, f := range c.forms {
isNo := (f.quickCheck[MDecomposed] == QCNo)
if (len(f.decomp) > 0) != isNo && !isHangul(rune(i)) {
log.Fatalf("%U: NF*D QC must be No if rune decomposes", i)
}
isMaybe := f.quickCheck[MComposed] == QCMaybe
if f.combinesBackward != isMaybe {
log.Fatalf("%U: NF*C QC must be Maybe if combinesBackward", i)
}
if len(f.decomp) > 0 && f.combinesForward && isMaybe {
log.Fatalf("%U: NF*C QC must be Yes or No if combinesForward and decomposes", i)
}
if len(f.expandedDecomp) != 0 {
continue
}
if a, b := c.nLeadingNonStarters > 0, (c.ccc > 0 || f.combinesBackward); a != b {
// We accept these runes to be treated differently (it only affects
// segment breaking in iteration, most likely on improper use), but
// reconsider if more characters are added.
// U+FF9E HALFWIDTH KATAKANA VOICED SOUND MARK;Lm;0;L;<narrow> 3099;;;;N;;;;;
// U+FF9F HALFWIDTH KATAKANA SEMI-VOICED SOUND MARK;Lm;0;L;<narrow> 309A;;;;N;;;;;
// U+3133 HANGUL LETTER KIYEOK-SIOS;Lo;0;L;<compat> 11AA;;;;N;HANGUL LETTER GIYEOG SIOS;;;;
// U+318E HANGUL LETTER ARAEAE;Lo;0;L;<compat> 11A1;;;;N;HANGUL LETTER ALAE AE;;;;
// U+FFA3 HALFWIDTH HANGUL LETTER KIYEOK-SIOS;Lo;0;L;<narrow> 3133;;;;N;HALFWIDTH HANGUL LETTER GIYEOG SIOS;;;;
// U+FFDC HALFWIDTH HANGUL LETTER I;Lo;0;L;<narrow> 3163;;;;N;;;;;
if i != 0xFF9E && i != 0xFF9F && !(0x3133 <= i && i <= 0x318E) && !(0xFFA3 <= i && i <= 0xFFDC) {
log.Fatalf("%U: nLead was %v; want %v", i, a, b)
}
}
}
nfc := c.forms[FCanonical]
nfkc := c.forms[FCompatibility]
if nfc.combinesBackward != nfkc.combinesBackward {
log.Fatalf("%U: Cannot combine combinesBackward\n", c.codePoint)
}
}
}
// Use values in DerivedNormalizationProps.txt to compare against the
// values we computed.
// DerivedNormalizationProps.txt has form:
// 00C0..00C5 ; NFD_QC; N # ...
// 0374 ; NFD_QC; N # ...
// See https://unicode.org/reports/tr44/ for full explanation
func testDerived() {
f := gen.OpenUCDFile("DerivedNormalizationProps.txt")
defer f.Close()
p := ucd.New(f)
for p.Next() {
r := p.Rune(0)
c := &chars[r]
var ftype, mode int
qt := p.String(1)
switch qt {
case "NFC_QC":
ftype, mode = FCanonical, MComposed
case "NFD_QC":
ftype, mode = FCanonical, MDecomposed
case "NFKC_QC":
ftype, mode = FCompatibility, MComposed
case "NFKD_QC":
ftype, mode = FCompatibility, MDecomposed
default:
continue
}
var qr QCResult
switch p.String(2) {
case "Y":
qr = QCYes
case "N":
qr = QCNo
case "M":
qr = QCMaybe
default:
log.Fatalf(`Unexpected quick check value "%s"`, p.String(2))
}
if got := c.forms[ftype].quickCheck[mode]; got != qr {
log.Printf("%U: FAILED %s (was %v need %v)\n", r, qt, got, qr)
}
c.forms[ftype].verified[mode] = true
}
if err := p.Err(); err != nil {
log.Fatal(err)
}
// Any unspecified value must be QCYes. Verify this.
for i, c := range chars {
for j, fd := range c.forms {
for k, qr := range fd.quickCheck {
if !fd.verified[k] && qr != QCYes {
m := "%U: FAIL F:%d M:%d (was %v need Yes) %s\n"
log.Printf(m, i, j, k, qr, c.name)
}
}
}
}
}
var testHeader = `const (
Yes = iota
No
Maybe
)
type formData struct {
qc uint8
combinesForward bool
decomposition string
}
type runeData struct {
r rune
ccc uint8
nLead uint8
nTrail uint8
f [2]formData // 0: canonical; 1: compatibility
}
func f(qc uint8, cf bool, dec string) [2]formData {
return [2]formData{{qc, cf, dec}, {qc, cf, dec}}
}
func g(qc, qck uint8, cf, cfk bool, d, dk string) [2]formData {
return [2]formData{{qc, cf, d}, {qck, cfk, dk}}
}
var testData = []runeData{
`
func printTestdata() {
type lastInfo struct {
ccc uint8
nLead uint8
nTrail uint8
f string
}
last := lastInfo{}
w := &bytes.Buffer{}
fmt.Fprintf(w, testHeader)
for r, c := range chars {
f := c.forms[FCanonical]
qc, cf, d := f.quickCheck[MComposed], f.combinesForward, string(f.expandedDecomp)
f = c.forms[FCompatibility]
qck, cfk, dk := f.quickCheck[MComposed], f.combinesForward, string(f.expandedDecomp)
s := ""
if d == dk && qc == qck && cf == cfk {
s = fmt.Sprintf("f(%s, %v, %q)", qc, cf, d)
} else {
s = fmt.Sprintf("g(%s, %s, %v, %v, %q, %q)", qc, qck, cf, cfk, d, dk)
}
current := lastInfo{c.ccc, c.nLeadingNonStarters, c.nTrailingNonStarters, s}
if last != current {
fmt.Fprintf(w, "\t{0x%x, %d, %d, %d, %s},\n", r, c.origCCC, c.nLeadingNonStarters, c.nTrailingNonStarters, s)
last = current
}
}
fmt.Fprintln(w, "}")
gen.WriteVersionedGoFile("data_test.go", "norm", w.Bytes())
}

View File

@@ -1,117 +0,0 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// Trie table generator.
// Used by make*tables tools to generate a go file with trie data structures
// for mapping UTF-8 to a 16-bit value. All but the last byte in a UTF-8 byte
// sequence are used to lookup offsets in the index table to be used for the
// next byte. The last byte is used to index into a table with 16-bit values.
package main
import (
"fmt"
"io"
)
const maxSparseEntries = 16
type normCompacter struct {
sparseBlocks [][]uint64
sparseOffset []uint16
sparseCount int
name string
}
func mostFrequentStride(a []uint64) int {
counts := make(map[int]int)
var v int
for _, x := range a {
if stride := int(x) - v; v != 0 && stride >= 0 {
counts[stride]++
}
v = int(x)
}
var maxs, maxc int
for stride, cnt := range counts {
if cnt > maxc || (cnt == maxc && stride < maxs) {
maxs, maxc = stride, cnt
}
}
return maxs
}
func countSparseEntries(a []uint64) int {
stride := mostFrequentStride(a)
var v, count int
for _, tv := range a {
if int(tv)-v != stride {
if tv != 0 {
count++
}
}
v = int(tv)
}
return count
}
func (c *normCompacter) Size(v []uint64) (sz int, ok bool) {
if n := countSparseEntries(v); n <= maxSparseEntries {
return (n+1)*4 + 2, true
}
return 0, false
}
func (c *normCompacter) Store(v []uint64) uint32 {
h := uint32(len(c.sparseOffset))
c.sparseBlocks = append(c.sparseBlocks, v)
c.sparseOffset = append(c.sparseOffset, uint16(c.sparseCount))
c.sparseCount += countSparseEntries(v) + 1
return h
}
func (c *normCompacter) Handler() string {
return c.name + "Sparse.lookup"
}
func (c *normCompacter) Print(w io.Writer) (retErr error) {
p := func(f string, x ...interface{}) {
if _, err := fmt.Fprintf(w, f, x...); retErr == nil && err != nil {
retErr = err
}
}
ls := len(c.sparseBlocks)
p("// %sSparseOffset: %d entries, %d bytes\n", c.name, ls, ls*2)
p("var %sSparseOffset = %#v\n\n", c.name, c.sparseOffset)
ns := c.sparseCount
p("// %sSparseValues: %d entries, %d bytes\n", c.name, ns, ns*4)
p("var %sSparseValues = [%d]valueRange {", c.name, ns)
for i, b := range c.sparseBlocks {
p("\n// Block %#x, offset %#x", i, c.sparseOffset[i])
var v int
stride := mostFrequentStride(b)
n := countSparseEntries(b)
p("\n{value:%#04x,lo:%#02x},", stride, uint8(n))
for i, nv := range b {
if int(nv)-v != stride {
if v != 0 {
p(",hi:%#02x},", 0x80+i-1)
}
if nv != 0 {
p("\n{value:%#04x,lo:%#02x", nv, 0x80+i)
}
}
v = int(nv)
}
if v != 0 {
p(",hi:%#02x},", 0x80+len(b)-1)
}
}
p("\n}\n\n")
return
}

View File

@@ -1,99 +0,0 @@
// Copyright 2017 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ignore
// The gcexportdata command is a diagnostic tool that displays the
// contents of gc export data files.
package main
import (
"flag"
"fmt"
"go/token"
"go/types"
"log"
"os"
"golang.org/x/tools/go/gcexportdata"
"golang.org/x/tools/go/types/typeutil"
)
var packageFlag = flag.String("package", "", "alternative package to print")
func main() {
log.SetPrefix("gcexportdata: ")
log.SetFlags(0)
flag.Usage = func() {
fmt.Fprintln(os.Stderr, "usage: gcexportdata [-package path] file.a")
}
flag.Parse()
if flag.NArg() != 1 {
flag.Usage()
os.Exit(2)
}
filename := flag.Args()[0]
f, err := os.Open(filename)
if err != nil {
log.Fatal(err)
}
r, err := gcexportdata.NewReader(f)
if err != nil {
log.Fatalf("%s: %s", filename, err)
}
// Decode the package.
const primary = "<primary>"
imports := make(map[string]*types.Package)
fset := token.NewFileSet()
pkg, err := gcexportdata.Read(r, fset, imports, primary)
if err != nil {
log.Fatalf("%s: %s", filename, err)
}
// Optionally select an indirectly mentioned package.
if *packageFlag != "" {
pkg = imports[*packageFlag]
if pkg == nil {
fmt.Fprintf(os.Stderr, "export data file %s does not mention %s; has:\n",
filename, *packageFlag)
for p := range imports {
if p != primary {
fmt.Fprintf(os.Stderr, "\t%s\n", p)
}
}
os.Exit(1)
}
}
// Print all package-level declarations, including non-exported ones.
fmt.Printf("package %s\n", pkg.Name())
for _, imp := range pkg.Imports() {
fmt.Printf("import %q\n", imp.Path())
}
qual := func(p *types.Package) string {
if pkg == p {
return ""
}
return p.Name()
}
scope := pkg.Scope()
for _, name := range scope.Names() {
obj := scope.Lookup(name)
fmt.Printf("%s: %s\n",
fset.Position(obj.Pos()),
types.ObjectString(obj, qual))
// For types, print each method.
if _, ok := obj.(*types.TypeName); ok {
for _, method := range typeutil.IntuitiveMethodSet(obj.Type(), nil) {
fmt.Printf("%s: %s\n",
fset.Position(method.Obj().Pos()),
types.SelectionString(method, qual))
}
}
}
}

View File

@@ -1,173 +0,0 @@
// +build ignore
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Command mkindex creates the file "pkgindex.go" containing an index of the Go
// standard library. The file is intended to be built as part of the imports
// package, so that the package may be used in environments where a GOROOT is
// not available (such as App Engine).
package main
import (
"bytes"
"fmt"
"go/ast"
"go/build"
"go/format"
"go/parser"
"go/token"
"io/ioutil"
"log"
"os"
"path"
"path/filepath"
"strings"
)
var (
pkgIndex = make(map[string][]pkg)
exports = make(map[string]map[string]bool)
)
func main() {
// Don't use GOPATH.
ctx := build.Default
ctx.GOPATH = ""
// Populate pkgIndex global from GOROOT.
for _, path := range ctx.SrcDirs() {
f, err := os.Open(path)
if err != nil {
log.Print(err)
continue
}
children, err := f.Readdir(-1)
f.Close()
if err != nil {
log.Print(err)
continue
}
for _, child := range children {
if child.IsDir() {
loadPkg(path, child.Name())
}
}
}
// Populate exports global.
for _, ps := range pkgIndex {
for _, p := range ps {
e := loadExports(p.dir)
if e != nil {
exports[p.dir] = e
}
}
}
// Construct source file.
var buf bytes.Buffer
fmt.Fprint(&buf, pkgIndexHead)
fmt.Fprintf(&buf, "var pkgIndexMaster = %#v\n", pkgIndex)
fmt.Fprintf(&buf, "var exportsMaster = %#v\n", exports)
src := buf.Bytes()
// Replace main.pkg type name with pkg.
src = bytes.Replace(src, []byte("main.pkg"), []byte("pkg"), -1)
// Replace actual GOROOT with "/go".
src = bytes.Replace(src, []byte(ctx.GOROOT), []byte("/go"), -1)
// Add some line wrapping.
src = bytes.Replace(src, []byte("}, "), []byte("},\n"), -1)
src = bytes.Replace(src, []byte("true, "), []byte("true,\n"), -1)
var err error
src, err = format.Source(src)
if err != nil {
log.Fatal(err)
}
// Write out source file.
err = ioutil.WriteFile("pkgindex.go", src, 0644)
if err != nil {
log.Fatal(err)
}
}
const pkgIndexHead = `package imports
func init() {
pkgIndexOnce.Do(func() {
pkgIndex.m = pkgIndexMaster
})
loadExports = func(dir string) map[string]bool {
return exportsMaster[dir]
}
}
`
type pkg struct {
importpath string // full pkg import path, e.g. "net/http"
dir string // absolute file path to pkg directory e.g. "/usr/lib/go/src/fmt"
}
var fset = token.NewFileSet()
func loadPkg(root, importpath string) {
shortName := path.Base(importpath)
if shortName == "testdata" {
return
}
dir := filepath.Join(root, importpath)
pkgIndex[shortName] = append(pkgIndex[shortName], pkg{
importpath: importpath,
dir: dir,
})
pkgDir, err := os.Open(dir)
if err != nil {
return
}
children, err := pkgDir.Readdir(-1)
pkgDir.Close()
if err != nil {
return
}
for _, child := range children {
name := child.Name()
if name == "" {
continue
}
if c := name[0]; c == '.' || ('0' <= c && c <= '9') {
continue
}
if child.IsDir() {
loadPkg(root, filepath.Join(importpath, name))
}
}
}
func loadExports(dir string) map[string]bool {
exports := make(map[string]bool)
buildPkg, err := build.ImportDir(dir, 0)
if err != nil {
if strings.Contains(err.Error(), "no buildable Go source files in") {
return nil
}
log.Printf("could not import %q: %v", dir, err)
return nil
}
for _, file := range buildPkg.GoFiles {
f, err := parser.ParseFile(fset, filepath.Join(dir, file), nil, 0)
if err != nil {
log.Printf("could not parse %q: %v", file, err)
continue
}
for name := range f.Scope.Objects {
if ast.IsExported(name) {
exports[name] = true
}
}
}
return exports
}

View File

@@ -1,132 +0,0 @@
// +build ignore
// mkstdlib generates the zstdlib.go file, containing the Go standard
// library API symbols. It's baked into the binary to avoid scanning
// GOPATH in the common case.
package main
import (
"bufio"
"bytes"
"fmt"
"go/format"
"io"
"io/ioutil"
"log"
"os"
"os/exec"
"path/filepath"
"regexp"
"runtime"
"sort"
"strings"
)
func mustOpen(name string) io.Reader {
f, err := os.Open(name)
if err != nil {
log.Fatal(err)
}
return f
}
func api(base string) string {
return filepath.Join(runtime.GOROOT(), "api", base)
}
var sym = regexp.MustCompile(`^pkg (\S+).*?, (?:var|func|type|const) ([A-Z]\w*)`)
var unsafeSyms = map[string]bool{"Alignof": true, "ArbitraryType": true, "Offsetof": true, "Pointer": true, "Sizeof": true}
func main() {
var buf bytes.Buffer
outf := func(format string, args ...interface{}) {
fmt.Fprintf(&buf, format, args...)
}
outf("// Code generated by mkstdlib.go. DO NOT EDIT.\n\n")
outf("package imports\n")
outf("var stdlib = map[string]map[string]bool{\n")
f := io.MultiReader(
mustOpen(api("go1.txt")),
mustOpen(api("go1.1.txt")),
mustOpen(api("go1.2.txt")),
mustOpen(api("go1.3.txt")),
mustOpen(api("go1.4.txt")),
mustOpen(api("go1.5.txt")),
mustOpen(api("go1.6.txt")),
mustOpen(api("go1.7.txt")),
mustOpen(api("go1.8.txt")),
mustOpen(api("go1.9.txt")),
mustOpen(api("go1.10.txt")),
mustOpen(api("go1.11.txt")),
mustOpen(api("go1.12.txt")),
// The API of the syscall/js package needs to be computed explicitly,
// because it's not included in the GOROOT/api/go1.*.txt files at this time.
syscallJSAPI(),
)
sc := bufio.NewScanner(f)
pkgs := map[string]map[string]bool{
"unsafe": unsafeSyms,
}
paths := []string{"unsafe"}
for sc.Scan() {
l := sc.Text()
has := func(v string) bool { return strings.Contains(l, v) }
if has("struct, ") || has("interface, ") || has(", method (") {
continue
}
if m := sym.FindStringSubmatch(l); m != nil {
path, sym := m[1], m[2]
if _, ok := pkgs[path]; !ok {
pkgs[path] = map[string]bool{}
paths = append(paths, path)
}
pkgs[path][sym] = true
}
}
if err := sc.Err(); err != nil {
log.Fatal(err)
}
sort.Strings(paths)
for _, path := range paths {
outf("\t%q: map[string]bool{\n", path)
pkg := pkgs[path]
var syms []string
for sym := range pkg {
syms = append(syms, sym)
}
sort.Strings(syms)
for _, sym := range syms {
outf("\t\t%q: true,\n", sym)
}
outf("},\n")
}
outf("}\n")
fmtbuf, err := format.Source(buf.Bytes())
if err != nil {
log.Fatal(err)
}
err = ioutil.WriteFile("zstdlib.go", fmtbuf, 0666)
if err != nil {
log.Fatal(err)
}
}
// syscallJSAPI returns the API of the syscall/js package.
// It's computed from the contents of $(go env GOROOT)/src/syscall/js.
func syscallJSAPI() io.Reader {
var exeSuffix string
if runtime.GOOS == "windows" {
exeSuffix = ".exe"
}
cmd := exec.Command("go"+exeSuffix, "run", "cmd/api", "-contexts", "js-wasm", "syscall/js")
out, err := cmd.Output()
if err != nil {
log.Fatalln(err)
}
return bytes.NewReader(out)
}

82
vendor/modules.txt vendored
View File

@@ -1,33 +1,29 @@
# github.com/99designs/gqlgen v0.9.0
github.com/99designs/gqlgen
github.com/99designs/gqlgen/graphql
github.com/99designs/gqlgen/handler
github.com/99designs/gqlgen/graphql/introspection
github.com/99designs/gqlgen/cmd
github.com/99designs/gqlgen/complexity
github.com/99designs/gqlgen/api
github.com/99designs/gqlgen/codegen/config
github.com/99designs/gqlgen/plugin/servergen
github.com/99designs/gqlgen/cmd
github.com/99designs/gqlgen/codegen
github.com/99designs/gqlgen/codegen/config
github.com/99designs/gqlgen/codegen/templates
github.com/99designs/gqlgen/complexity
github.com/99designs/gqlgen/graphql
github.com/99designs/gqlgen/graphql/introspection
github.com/99designs/gqlgen/handler
github.com/99designs/gqlgen/internal/code
github.com/99designs/gqlgen/internal/imports
github.com/99designs/gqlgen/plugin
github.com/99designs/gqlgen/plugin/modelgen
github.com/99designs/gqlgen/plugin/resolvergen
github.com/99designs/gqlgen/codegen/templates
github.com/99designs/gqlgen/internal/code
github.com/99designs/gqlgen/internal/imports
github.com/99designs/gqlgen/plugin/servergen
# github.com/BurntSushi/toml v0.3.1
github.com/BurntSushi/toml
# github.com/PuerkitoBio/goquery v1.5.0
github.com/PuerkitoBio/goquery
# github.com/agnivade/levenshtein v1.0.1
github.com/agnivade/levenshtein
# github.com/andybalholm/cascadia v1.0.0
github.com/andybalholm/cascadia
# github.com/antchfx/htmlquery v1.2.3
github.com/antchfx/htmlquery
# github.com/antchfx/xpath v1.1.6
github.com/antchfx/xpath
# github.com/bmatcuk/doublestar v1.1.5
# github.com/bmatcuk/doublestar v1.3.1
github.com/bmatcuk/doublestar
# github.com/davecgh/go-spew v1.1.1
github.com/davecgh/go-spew/spew
@@ -46,8 +42,8 @@ github.com/gobuffalo/envy
# github.com/gobuffalo/events v1.2.0
github.com/gobuffalo/events
# github.com/gobuffalo/flect v0.1.0
github.com/gobuffalo/flect/name
github.com/gobuffalo/flect
github.com/gobuffalo/flect/name
# github.com/gobuffalo/genny v0.0.0-20190219203444-c95082806342
github.com/gobuffalo/genny
# github.com/gobuffalo/gogen v0.0.0-20190224213239-1c6076128bbc
@@ -64,19 +60,19 @@ github.com/gobuffalo/meta
github.com/gobuffalo/packd
# github.com/gobuffalo/packr/v2 v2.0.2
github.com/gobuffalo/packr/v2
github.com/gobuffalo/packr/v2/file/resolver
github.com/gobuffalo/packr/v2/file
github.com/gobuffalo/packr/v2/file/resolver
github.com/gobuffalo/packr/v2/file/resolver/encoding/hex
github.com/gobuffalo/packr/v2/jam/parser
github.com/gobuffalo/packr/v2/plog
github.com/gobuffalo/packr/v2/file/resolver/encoding/hex
# github.com/gobuffalo/syncx v0.0.0-20181120194010-558ac7de985f
github.com/gobuffalo/syncx
# github.com/golang-migrate/migrate/v4 v4.3.1
github.com/golang-migrate/migrate/v4/database/sqlite3
github.com/golang-migrate/migrate/v4/source/file
github.com/golang-migrate/migrate/v4
github.com/golang-migrate/migrate/v4/source
github.com/golang-migrate/migrate/v4/database
github.com/golang-migrate/migrate/v4/database/sqlite3
github.com/golang-migrate/migrate/v4/source
github.com/golang-migrate/migrate/v4/source/file
# github.com/golang/groupcache v0.0.0-20200121045136-8c9f03a8e57e
github.com/golang/groupcache/lru
# github.com/gorilla/securecookie v1.1.1
@@ -87,9 +83,9 @@ github.com/gorilla/sessions
github.com/gorilla/websocket
# github.com/h2non/filetype v1.0.8
github.com/h2non/filetype
github.com/h2non/filetype/types
github.com/h2non/filetype/matchers
github.com/h2non/filetype/matchers/isobmff
github.com/h2non/filetype/types
# github.com/hashicorp/errwrap v1.0.0
github.com/hashicorp/errwrap
# github.com/hashicorp/go-multierror v1.0.0
@@ -99,13 +95,13 @@ github.com/hashicorp/golang-lru
github.com/hashicorp/golang-lru/simplelru
# github.com/hashicorp/hcl v1.0.0
github.com/hashicorp/hcl
github.com/hashicorp/hcl/hcl/printer
github.com/hashicorp/hcl/hcl/ast
github.com/hashicorp/hcl/hcl/parser
github.com/hashicorp/hcl/hcl/token
github.com/hashicorp/hcl/json/parser
github.com/hashicorp/hcl/hcl/printer
github.com/hashicorp/hcl/hcl/scanner
github.com/hashicorp/hcl/hcl/strconv
github.com/hashicorp/hcl/hcl/token
github.com/hashicorp/hcl/json/parser
github.com/hashicorp/hcl/json/scanner
github.com/hashicorp/hcl/json/token
# github.com/inconshreveable/mousetrap v1.0.0
@@ -179,27 +175,27 @@ github.com/vektah/dataloaden/pkg/generator
github.com/vektah/gqlparser
github.com/vektah/gqlparser/ast
github.com/vektah/gqlparser/gqlerror
github.com/vektah/gqlparser/lexer
github.com/vektah/gqlparser/parser
github.com/vektah/gqlparser/validator
github.com/vektah/gqlparser/validator/rules
github.com/vektah/gqlparser/lexer
# golang.org/x/crypto v0.0.0-20190701094942-4def268fd1a4
golang.org/x/crypto/bcrypt
golang.org/x/crypto/blowfish
golang.org/x/crypto/ssh/terminal
# golang.org/x/image v0.0.0-20190118043309-183bebdce1b2
golang.org/x/image/webp
golang.org/x/image/bmp
golang.org/x/image/tiff
golang.org/x/image/riff
golang.org/x/image/tiff
golang.org/x/image/tiff/lzw
golang.org/x/image/vp8
golang.org/x/image/vp8l
golang.org/x/image/tiff/lzw
golang.org/x/image/webp
# golang.org/x/net v0.0.0-20200421231249-e086a090c8fd
golang.org/x/net/html
golang.org/x/net/html/charset
golang.org/x/net/context/ctxhttp
golang.org/x/net/html
golang.org/x/net/html/atom
golang.org/x/net/html/charset
# golang.org/x/sys v0.0.0-20200323222414-85ca7c5b95cd
golang.org/x/sys/unix
golang.org/x/sys/windows
@@ -207,31 +203,31 @@ golang.org/x/sys/windows
golang.org/x/text/encoding
golang.org/x/text/encoding/charmap
golang.org/x/text/encoding/htmlindex
golang.org/x/text/transform
golang.org/x/text/unicode/norm
golang.org/x/text/encoding/internal/identifier
golang.org/x/text/encoding/internal
golang.org/x/text/encoding/internal/identifier
golang.org/x/text/encoding/japanese
golang.org/x/text/encoding/korean
golang.org/x/text/encoding/simplifiedchinese
golang.org/x/text/encoding/traditionalchinese
golang.org/x/text/encoding/unicode
golang.org/x/text/language
golang.org/x/text/internal/utf8internal
golang.org/x/text/runes
golang.org/x/text/internal/language
golang.org/x/text/internal/language/compact
golang.org/x/text/internal/tag
golang.org/x/text/internal/utf8internal
golang.org/x/text/language
golang.org/x/text/runes
golang.org/x/text/transform
golang.org/x/text/unicode/norm
# golang.org/x/tools v0.0.0-20190515012406-7d7faa4812bd
golang.org/x/tools/go/packages
golang.org/x/tools/imports
golang.org/x/tools/go/ast/astutil
golang.org/x/tools/go/gcexportdata
golang.org/x/tools/go/internal/packagesdriver
golang.org/x/tools/internal/gopathwalk
golang.org/x/tools/internal/semver
golang.org/x/tools/internal/module
golang.org/x/tools/go/internal/gcimporter
golang.org/x/tools/go/internal/packagesdriver
golang.org/x/tools/go/packages
golang.org/x/tools/imports
golang.org/x/tools/internal/fastwalk
golang.org/x/tools/internal/gopathwalk
golang.org/x/tools/internal/module
golang.org/x/tools/internal/semver
# gopkg.in/yaml.v2 v2.2.2
gopkg.in/yaml.v2