[bunkr] fix extraction (#3636, #3655)

gallery_dl/extractor/bunkr.py

@@ -9,7 +9,7 @@
 """Extractors for https://bunkr.su/"""
 
 from .lolisafe import LolisafeAlbumExtractor
-from .. import text, util
+from .. import text
 
 
 class BunkrAlbumExtractor(LolisafeAlbumExtractor):
@@ -19,7 +19,7 @@ class BunkrAlbumExtractor(LolisafeAlbumExtractor):
     pattern = r"(?:https?://)?(?:app\.)?bunkr\.(?:[sr]u|is|to)/a/([^/?#]+)"
     test = (
         ("https://bunkr.su/a/Lktg9Keq", {
-            "pattern": r"https://cdn\.bunkr\.su/test-テスト-\"&>-QjgneIQv\.png",
+            "pattern": r"https://cdn\.bunkr\.ru/test-テスト-\"&>-QjgneIQv\.png",
             "content": "0c8768055e4e20e7c7259608b67799171b691140",
             "keyword": {
                 "album_id": "Lktg9Keq",
@@ -40,44 +40,57 @@ class BunkrAlbumExtractor(LolisafeAlbumExtractor):
         ("https://bunkr.is/a/iXTTc1o2", {
             "pattern": r"https://(cdn|media-files)4\.bunkr\.ru/",
             "content": "da29aae371b7adc8c5ef8e6991b66b69823791e8",
+            "keyword": {
+                "album_id": "iXTTc1o2",
+                "album_name": "test2",
+                "album_size": "691.1 KB",
+                "count": 2,
+                "description": "072022",
+                "filename": "re:video-wFO9FtxG|image-sZrQUeOx",
+                "id": "re:wFO9FtxG|sZrQUeOx",
+                "name": "re:video|image",
+                "num": int,
+            },
         }),
         ("https://bunkr.to/a/Lktg9Keq"),
     )
 
     def fetch_album(self, album_id):
-        root = self.root
+        # album metadata
+        page = self.request(self.root + "/a/" + self.album_id).text
+        info = text.split_html(text.extr(
+            page, "<h1", "</div>").partition(">")[2])
+        count, _, size = info[1].split(None, 2)
+
+        # files
+        cdn = None
+        files = []
+        append = files.append
+        headers = {"Referer": self.root.replace("://", "://stream.", 1) + "/"}
 
-        try:
+        pos = page.index('class="grid-images')
-            data = util.json_loads(text.extr(
+        for url in text.extract_iter(page, '<a href="', '"', pos):
-                self.request(root + "/a/" + self.album_id).text,
+            if url.startswith("/"):
-                'id="__NEXT_DATA__" type="application/json">', '<'))
+                if not cdn:
-            album = data["props"]["pageProps"]["album"]
+                    # fetch cdn root from download page
-            files = album["files"]
+                    durl = "{}/d/{}".format(self.root, url[3:])
-        except Exception as exc:
+                    cdn = text.extr(self.request(
-            self.log.debug("%s: %s", exc.__class__.__name__, exc)
+                        durl).text, 'link.href = "', '"')
-            self.log.debug("Falling back to lolisafe API")
+                    cdn = cdn[:cdn.index("/", 8)]
-            self.root = root.replace("://", "://app.", 1)
+                url = cdn + url[2:]
-            files, data = LolisafeAlbumExtractor.fetch_album(self, album_id)
-            # fix file URLs (bunkr..ru -> bunkr.ru) (#3481)
-            for file in files:
-                file["file"] = file["file"].replace("bunkr..", "bunkr.", 1)
-        else:
-            for file in files:
-                file["file"] = file["cdn"] + "/" + file["name"]
-            data = {
-                "album_id"   : self.album_id,
-                "album_name" : text.unescape(album["name"]),
-                "description": text.unescape(album["description"]),
-                "count"      : len(files),
-            }
 
-        headers = {"Referer": root.replace("://", "://stream.", 1) + "/"}
+            url = text.unescape(url)
-        for file in files:
+            if url.endswith((".mp4", ".m4v", ".mov", ".webm", ".mkv", ".ts",
-            if file["file"].endswith(
+                             ".zip", ".rar", ".7z")):
-                    (".mp4", ".m4v", ".mov", ".webm", ".mkv", ".ts",
+                append({"file": url.replace("://cdn", "://media-files", 1),
-                     ".zip", ".rar", ".7z")):
+                        "_http_headers": headers})
-                file["_http_headers"] = headers
+            else:
-                file["file"] = file["file"].replace(
+                append({"file": url})
-                    "://cdn", "://media-files", 1)
 
-        return files, data
+        return files, {
+            "album_id"   : self.album_id,
+            "album_name" : text.unescape(info[0]),
+            "album_size" : size[1:-1],
+            "description": text.unescape(info[2]) if len(info) > 2 else "",
+            "count"      : len(files),
+        }