Add http headers support to scraper (#1273)

pkg/scraper/config.go

@@ -175,11 +175,17 @@ type clickOptions struct {
 	Sleep int    `yaml:"sleep"`
 }
 
+type header struct {
+	Key   string `yaml:"Key"`
+	Value string `yaml:"Value"`
+}
+
 type scraperDriverOptions struct {
 	UseCDP  bool             `yaml:"useCDP"`
 	Sleep   int              `yaml:"sleep"`
 	Clicks  []*clickOptions  `yaml:"clicks"`
 	Cookies []*cookieOptions `yaml:"cookies"`
+	Headers []*header        `yaml:"headers"`
 }
 
 func loadScraperFromYAML(id string, reader io.Reader) (*config, error) {

pkg/scraper/url.go

@@ -74,12 +74,21 @@ func loadURL(url string, scraperConfig config, globalConfig GlobalConfig) (io.Re
 		req.Header.Set("User-Agent", userAgent)
 	}
 
+	if driverOptions != nil { // setting the Headers after the UA allows us to override it from inside the scraper
+		for _, h := range driverOptions.Headers {
+			if h.Key != "" {
+				req.Header.Set(h.Key, h.Value)
+				logger.Debugf("[scraper] adding header <%s:%s>", h.Key, h.Value)
+			}
+		}
+	}
+
 	resp, err := client.Do(req)
 	if err != nil {
 		return nil, err
 	}
 	if resp.StatusCode >= 400 {
-		return nil, fmt.Errorf("http error %d", resp.StatusCode)
+		return nil, fmt.Errorf("http error %d:%s", resp.StatusCode, http.StatusText(resp.StatusCode))
 	}
 
 	defer resp.Body.Close()
@@ -156,10 +165,13 @@ func urlFromCDP(url string, driverOptions scraperDriverOptions, globalConfig Glo
 	defer cancel()
 
 	var res string
+	headers := cdpHeaders(driverOptions)
+
 	err := chromedp.Run(ctx,
 		network.Enable(),
 		setCDPCookies(driverOptions),
 		printCDPCookies(driverOptions, "Cookies found"),
+		network.SetExtraHTTPHeaders(network.Headers(headers)),
 		chromedp.Navigate(url),
 		chromedp.Sleep(sleepDuration),
 		setCDPClicks(driverOptions),
@@ -241,3 +253,16 @@ func cdpNetwork(enable bool) chromedp.Action {
 		return nil
 	})
 }
+
+func cdpHeaders(driverOptions scraperDriverOptions) map[string]interface{} {
+	headers := map[string]interface{}{}
+	if driverOptions.Headers != nil {
+		for _, h := range driverOptions.Headers {
+			if h.Key != "" {
+				headers[h.Key] = h.Value
+				logger.Debugf("[scraper] adding header <%s:%s>", h.Key, h.Value)
+			}
+		}
+	}
+	return headers
+}

ui/v2.5/src/components/Changelog/versions/v070.md

@@ -4,6 +4,7 @@
 * Added scene queue.
 
 ### 🎨 Improvements
+* Support http request headers in scrapers.
 * Sort performers by gender in scene/image/gallery cards and details.
 * Add popover buttons for scenes/images/galleries on performer/studio/tag cards.
 * Add slideshow to image wall view.

ui/v2.5/src/docs/en/Scraping.md

@@ -544,6 +544,24 @@ When developing a scraper you can have a look at the cookies set by a site by ad
 
 and having a look at the log / console in debug mode.
 
+### Headers
+
+Sending request headers is possible when using a scraper.
+Headers can be set in the `driver` section and are supported for plain, CDP enabled and JSON scrapers.
+They consist of a Key and a Value. If the the Key is empty or not defined then the header is ignored.
+
+```yaml
+driver:
+  headers:
+    - Key: User-Agent
+      Value: My Stash Scraper
+    - Key: Authorization
+      Value: Bearer ds3sdfcFdfY17p4qBkTVF03zscUU2glSjWF17bZyoe8
+```
+
+* headers are set after stash's `User-Agent` configuration option is applied.
+This means setting a `User-Agent` header from the scraper overrides the one in the configuration settings.
+
 ### XPath scraper example
 
 A performer and scene xpath scraper is shown as an example below:
@@ -614,31 +632,42 @@ A performer and scene scraper for ThePornDB is shown below:
 name: ThePornDB
 performerByName:
   action: scrapeJson
-  queryURL: https://metadataapi.net/api/performers?q={}
+  queryURL: https://api.metadataapi.net/performers?q={}
   scraper: performerSearch
 performerByURL:
   - action: scrapeJson
     url:
-      - https://metadataapi.net/api/performers/
+      - https://api.metadataapi.net/performers/
     scraper: performerScraper
 sceneByURL:
   - action: scrapeJson
     url:
-      - https://metadataapi.net/api/scenes/
+      - https://api.metadataapi.net/scenes/
     scraper: sceneScraper
 sceneByFragment:
   action: scrapeJson
-  queryURL: https://metadataapi.net/api/scenes?parse={filename}&limit=1
+  queryURL: https://api.metadataapi.net/scenes?parse={filename}&hash={oshash}&limit=1
   scraper: sceneQueryScraper
+  queryURLReplace:
+    filename:
+      - regex: "[^a-zA-Z\\d\\-._~]" # clean filename so that it can contruct a valid url
+        with: "." # "%20"
+      - regex: HEVC
+        with:
+      - regex: x265
+        with:
+      - regex: \.+
+        with: "."
 jsonScrapers:
   performerSearch:
     performer:
       Name: data.#.name
       URL:
         selector: data.#.id
-        replace:
+        postProcess:
+          - replace:
               - regex: ^
-            with: https://metadataapi.net/api/performers/
+                with: https://api.metadataapi.net/performers/
 
   performerScraper:
     common:
@@ -648,7 +677,12 @@ jsonScrapers:
       Gender: $extras.gender
       Birthdate: $extras.birthday
       Ethnicity: $extras.ethnicity
-      Height: $extras.height
+      Height:
+        selector: $extras.height
+        postProcess:
+          - replace:
+              - regex: cm
+                with:
       Measurements: $extras.measurements
       Tattoos: $extras.tattoos
       Piercings: $extras.piercings
@@ -687,6 +721,13 @@ jsonScrapers:
         Name: $data.site.name
       Tags:
         Name: $data.tags.#.tag
+driver:
+  headers:
+    - Key: User-Agent
+      Value: Stash JSON Scraper
+    - Key: Authorization
+      Value: Bearer lPdwFdfY17p4qBkTVF03zscUU2glSjdf17bZyoe  # use an actual API Key here
+# Last Updated April 7, 2021
 ```
 
 ## Object fields