merge #6443: [bilibili] add support for articles (#2824)

docs/supportedsites.md

@@ -133,6 +133,12 @@ Consider all listed sites to potentially be NSFW.
     <td>Collections, Galleries, User Profiles</td>
     <td></td>
 </tr>
+<tr>
+    <td>Bilibili</td>
+    <td>https://www.bilibili.com/</td>
+    <td>Articles, User Articles</td>
+    <td></td>
+</tr>
 <tr>
     <td>Bluesky</td>
     <td>https://bsky.app/</td>

gallery_dl/extractor/__init__.py

@@ -30,6 +30,7 @@ modules = [
     "batoto",
     "bbc",
     "behance",
+    "bilibili",
     "blogger",
     "bluesky",
     "boosty",

gallery_dl/extractor/bilibili.py

@@ -0,0 +1,107 @@
+# -*- coding: utf-8 -*-
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License version 2 as
+# published by the Free Software Foundation.
+
+"""Extractors for https://www.bilibili.com/"""
+
+from .common import Extractor, Message
+from .. import text, util, exception
+
+
+class BilibiliExtractor(Extractor):
+    """Base class for bilibili extractors"""
+    category = "bilibili"
+    root = "https://www.bilibili.com"
+
+    def _init(self):
+        self.api = BilibiliAPI(self)
+
+
+class BilibiliUserArticlesExtractor(BilibiliExtractor):
+    """Extractor for a bilibili user's articles"""
+    subcategory = "user-articles"
+    pattern = r"(?:https?://)?space\.bilibili\.com/(\d+)/article"
+    example = "https://space.bilibili.com/12345/article"
+
+    def items(self):
+        for article in self.api.user_articles(self.groups[0]):
+            article["_extractor"] = BilibiliArticleExtractor
+            url = "{}/opus/{}".format(self.root, article["opus_id"])
+            yield Message.Queue, url, article
+
+
+class BilibiliArticleExtractor(BilibiliExtractor):
+    """Extractor for a bilibili article"""
+    subcategory = "article"
+    pattern = (r"(?:https?://)?"
+               r"(?:t\.bilibili\.com|(?:www\.)?bilibili.com/opus)/(\d+)")
+    example = "https://www.bilibili.com/opus/12345"
+    directory_fmt = ("{category}", "{username}")
+    filename_fmt = "{id}_{num}.{extension}"
+    archive_fmt = "{id}_{num}"
+
+    def items(self):
+        article = self.api.article(self.groups[0])
+
+        # Flatten modules list
+        modules = {}
+        for module in article["detail"]["modules"]:
+            del module['module_type']
+            modules.update(module)
+        article["detail"]["modules"] = modules
+
+        article["username"] = modules["module_author"]["name"]
+
+        pics = []
+        for paragraph in modules['module_content']['paragraphs']:
+            if "pic" not in paragraph:
+                continue
+
+            try:
+                pics.extend(paragraph["pic"]["pics"])
+            except Exception:
+                pass
+
+        article["count"] = len(pics)
+        yield Message.Directory, article
+        for article["num"], pic in enumerate(pics, 1):
+            url = pic["url"]
+            article.update(pic)
+            yield Message.Url, url, text.nameext_from_url(url, article)
+
+
+class BilibiliAPI():
+    def __init__(self, extractor):
+        self.extractor = extractor
+
+    def _call(self, endpoint, params):
+        url = "https://api.bilibili.com/x/polymer/web-dynamic/v1" + endpoint
+        data = self.extractor.request(url, params=params).json()
+
+        if data["code"] != 0:
+            self.extractor.log.debug("Server response: %s", data)
+            raise exception.StopExtraction("API request failed")
+
+        return data
+
+    def user_articles(self, user_id):
+        endpoint = "/opus/feed/space"
+        params = {"host_mid": user_id}
+
+        while True:
+            data = self._call(endpoint, params)
+
+            for item in data["data"]["items"]:
+                params["offset"] = item["opus_id"]
+                yield item
+
+            if not data["data"]["has_more"]:
+                break
+
+    def article(self, article_id):
+        url = "https://www.bilibili.com/opus/" + article_id
+        response = self.extractor.request(url)
+        return util.json_loads(text.extr(
+            response.text, "window.__INITIAL_STATE__=", "};") + "}")

scripts/supportedsites.py

@@ -195,6 +195,9 @@ SUBCATEGORY_MAP = {
         "artwork": "Artwork Listings",
         "collections": "",
     },
+    "bilibili": {
+        "user-articles": "User Articles",
+    },
     "bluesky": {
         "posts": "",
     },

test/results/bilibili.py

@@ -0,0 +1,45 @@
+# -*- coding: utf-8 -*-
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License version 2 as
+# published by the Free Software Foundation.
+
+from gallery_dl.extractor import bilibili
+
+
+__tests__ = (
+{
+    "#url"  : "https://www.bilibili.com/opus/988425412565532689",
+    "#class": bilibili.BilibiliArticleExtractor,
+    "#urls" : (
+        "http://i0.hdslb.com/bfs/new_dyn/311264c4dcf45261f7d7a7fe451b05b9405279279.png",
+        "http://i0.hdslb.com/bfs/new_dyn/b60d8bc6996529613d617443a12c0a93405279279.png",
+        "http://i0.hdslb.com/bfs/new_dyn/d4494543210d9eee5310e11dc62581e4405279279.png",
+        "http://i0.hdslb.com/bfs/new_dyn/45268e63086b2d99811b2e6490130937405279279.png",
+    ),
+
+    "count"    : 4,
+    "detail"   : dict,
+    "extension": "png",
+    "filename" : str,
+    "height"   : 800,
+    "id"       : "988425412565532689",
+    "isClient" : False,
+    "isPreview": False,
+    "num"      : range(1, 4),
+    "size"     : float,
+    "theme"    : str,
+    "themeMode": "light",
+    "url"      : str,
+    "username" : "平平出击",
+    "width"    : 800,
+},
+
+{
+    "#url"    : "https://space.bilibili.com/405279279/article",
+    "#class"  : bilibili.BilibiliUserArticlesExtractor,
+    "#pattern": bilibili.BilibiliArticleExtractor.pattern,
+    "#count"  : range(50, 100),
+},
+
+)