fix: Corriger le scraper pour Flickr.

regseb · regseb · commit c6fa8d3de54b · 2023-11-27T10:13:57.000+01:00
diff --git a/src/core/scraper/flickr.js b/src/core/scraper/flickr.js
@@ -38,19 +38,18 @@ const API_URL =
  */
 const action = async function (_url, metadata) {
     const doc = await metadata.html();
-    const video = doc.querySelector("video");
-    if (null === video) {
-        return undefined;
-    }
 
-    const parts = new URL(video.poster).pathname.split(/[./_]/u);
+    const meta = doc.querySelector('meta[property="og:image"]');
+    const parts = new URL(meta.content).pathname.split(/[./_]/u);
     const photoId = parts[2];
     const secret = parts[3];
-    const key = KEY_REGEXP.exec(doc.documentElement.innerHTML).groups.key;
+
+    const apiKey = KEY_REGEXP.exec(doc.documentElement.innerHTML).groups.key;
+
     const response = await fetch(
-        `${API_URL}&photo_id=${photoId}&secret=${secret}&api_key=${key}`,
+        `${API_URL}&photo_id=${photoId}&secret=${secret}&api_key=${apiKey}`,
     );
     const json = await response.json();
-    return json.streams.stream[0]["_content"];
+    return json.streams?.stream[0]["_content"];
 };
 export const extract = matchPattern(action, "*://www.flickr.com/photos/*");
diff --git a/test/unit/core/scraper/flickr.js b/test/unit/core/scraper/flickr.js
@@ -18,26 +18,46 @@ describe("core/scraper/flickr.js", function () {
         });
 
         it("should return undefined when it isn't a video", async function () {
+            const stub = sinon
+                .stub(globalThis, "fetch")
+                .resolves(Response.json({}));
+
             const url = new URL("https://www.flickr.com/photos/foo");
             const metadata = {
                 html: () =>
                     Promise.resolve(
                         new DOMParser().parseFromString(
-                            "<html><body></body></html>",
+                            `<html><head>
+                               <meta property="og:image"
+                                     content="https://live.staticflickr.com` +
+                                `/bar/baz_qux_b.jpg">
+                             </head><body>
+                               <script>
+                                 root.YUI_config.flickr.api.site_key = "quux";
+                               </script>
+                             </body></html>`,
                             "text/html",
                         ),
                     ),
             };
 
             const file = await scraper.extract(url, metadata);
             assert.equal(file, undefined);
+
+            assert.equal(stub.callCount, 1);
+            assert.deepEqual(stub.firstCall.args, [
+                "https://api.flickr.com/services/rest" +
+                    "?method=flickr.video.getStreamInfo&format=json" +
+                    "&nojsoncallback=1&photo_id=baz&secret=qux" +
+                    "&api_key=quux",
+            ]);
         });
 
         it("should return video URL", async function () {
             const stub = sinon.stub(globalThis, "fetch").resolves(
                 Response.json({
                     streams: {
-                        stream: [{ _content: "https://foo.net/bar.mp4" }],
+                        stream: [{ _content: "https://foo.com/bar.mp4" }],
                     },
                 }),
             );
@@ -47,27 +67,29 @@ describe("core/scraper/flickr.js", function () {
                 html: () =>
                     Promise.resolve(
                         new DOMParser().parseFromString(
-                            `<html><body>
+                            `<html><head>
+                               <meta property="og:image"
+                                     content="https://live.staticflickr.com` +
+                                `/qux/quux_corge_b.jpg">
+                             </head><body>
                                <script>
-                                 root.YUI_config.flickr.api.site_key = "qux";
+                                 root.YUI_config.flickr.api.site_key = "grault";
                                </script>
-                               <video poster="//quux.com/corge` +
-                                `/grault_garply.jpg" />
                              </body></html>`,
                             "text/html",
                         ),
                     ),
             };
 
             const file = await scraper.extract(url, metadata);
-            assert.equal(file, "https://foo.net/bar.mp4");
+            assert.equal(file, "https://foo.com/bar.mp4");
 
             assert.equal(stub.callCount, 1);
             assert.deepEqual(stub.firstCall.args, [
                 "https://api.flickr.com/services/rest" +
                     "?method=flickr.video.getStreamInfo&format=json" +
-                    "&nojsoncallback=1&photo_id=grault&secret=garply" +
-                    "&api_key=qux",
+                    "&nojsoncallback=1&photo_id=quux&secret=corge" +
+                    "&api_key=grault",
             ]);
         });
     });
