[vk] add 'album' extractor (#474, fixes #1952)

todo: better metadata for albums

docs/supportedsites.md

@@ -778,7 +778,7 @@ Consider all sites to be NSFW unless otherwise known.
 <tr>
     <td>VK</td>
     <td>https://vk.com/</td>
-    <td>Photos</td>
+    <td>Albums, Photos</td>
     <td></td>
 </tr>
 <tr>

gallery_dl/extractor/vk.py

@@ -12,18 +12,67 @@ from .common import Extractor, Message
 from .. import text
 import re
 
+BASE_PATTERN = r"(?:https://)?(?:www\.|m\.)?vk\.com"
 
-class VkPhotosExtractor(Extractor):
+
-    """Extractor for photos from a vk user"""
+class VkExtractor(Extractor):
+    """Base class for vk extractors"""
     category = "vk"
-    subcategory = "photos"
     directory_fmt = ("{category}", "{user[name]|user[id]}")
     filename_fmt = "{id}.{extension}"
     archive_fmt = "{id}"
     root = "https://vk.com"
     request_interval = 1.0
-    pattern = (r"(?:https://)?(?:www\.|m\.)?vk\.com/(?:"
+
-               r"(?:albums|photos|id)(-?\d+)|([^/?#]+))")
+    def items(self):
+        data = self.metadata()
+        yield Message.Directory, data
+        for photo in self.photos():
+            photo.update(data)
+            yield Message.Url, photo["url"], photo
+
+    def _pagination(self, photos_url, user_id):
+        sub = re.compile(r"/imp[fg]/").sub
+        needle = 'data-id="{}_'.format(user_id)
+        cnt = 0
+
+        headers = {
+            "X-Requested-With": "XMLHttpRequest",
+            "Origin"          : self.root,
+            "Referer"         : photos_url,
+        }
+        params = {
+            "al"    : "1",
+            "al_ad" : "0",
+            "offset": 0,
+            "part"  : "1",
+        }
+
+        while True:
+            payload = self.request(
+                photos_url, method="POST", headers=headers, data=params
+            ).json()["payload"][1]
+
+            offset = payload[0]
+            html = payload[1]
+
+            for cnt, photo in enumerate(text.extract_iter(html, needle, ')')):
+                pid = photo[:photo.find('"')]
+                url = photo[photo.rindex("(")+1:]
+                url = sub("/", url.partition("?")[0])
+                yield text.nameext_from_url(url, {"url": url, "id": pid})
+
+            if cnt <= 20 or offset == params["offset"]:
+                return
+            params["offset"] = offset
+
+
+class VkPhotosExtractor(VkExtractor):
+    """Extractor for photos from a vk user"""
+    subcategory = "photos"
+    pattern = (BASE_PATTERN + r"/(?:"
+               r"(?:albums|photos|id)(-?\d+)"
+               r"|(?!album-?\d+_)([^/?#]+))")
     test = (
         ("https://vk.com/id398982326", {
             "pattern": r"https://sun\d+-\d+\.userapi\.com/c\d+/v\d+"
@@ -58,10 +107,14 @@ class VkPhotosExtractor(Extractor):
     )
 
     def __init__(self, match):
-        Extractor.__init__(self, match)
+        VkExtractor.__init__(self, match)
         self.user_id, self.user_name = match.groups()
 
-    def items(self):
+    def photos(self):
+        url = "{}/photos{}".format(self.root, self.user_id)
+        return self._pagination(url, self.user_id)
+
+    def metadata(self):
         if self.user_id:
             user_id = self.user_id
             prefix = "public" if user_id[0] == "-" else "id"
@@ -70,40 +123,8 @@ class VkPhotosExtractor(Extractor):
         else:
             url = "{}/{}".format(self.root, self.user_name)
             data = self._extract_profile(url)
-            user_id = data["user"]["id"]
+            self.user_id = data["user"]["id"]
-
+        return data
-        photos_url = "{}/photos{}".format(self.root, user_id)
-        headers = {
-            "X-Requested-With": "XMLHttpRequest",
-            "Origin"          : self.root,
-            "Referer"         : photos_url,
-        }
-        params = {
-            "al"    : "1",
-            "al_ad" : "0",
-            "offset": 0,
-            "part"  : "1",
-        }
-
-        yield Message.Directory, data
-        sub = re.compile(r"/imp[fg]/").sub
-        needle = 'data-id="{}_'.format(user_id)
-        cnt = 0
-
-        while True:
-            offset, html = self.request(
-                photos_url, method="POST", headers=headers, data=params
-            ).json()["payload"][1]
-
-            for cnt, photo in enumerate(text.extract_iter(html, needle, ')')):
-                data["id"] = photo[:photo.find('"')]
-                url = photo[photo.rindex("(")+1:]
-                url = sub("/", url.partition("?")[0])
-                yield Message.Url, url, text.nameext_from_url(url, data)
-
-            if cnt <= 40 or offset == params["offset"]:
-                return
-            params["offset"] = offset
 
     def _extract_profile(self, url):
         extr = text.extract_from(self.request(url).text)
@@ -116,3 +137,32 @@ class VkPhotosExtractor(Extractor):
                 '<span class="current_text">', '</span'))),
             "id"  : extr('<a href="/albums', '"'),
         }}
+
+
+class VkAlbumExtractor(VkExtractor):
+    """Extractor for a vk album"""
+    subcategory = "album"
+    directory_fmt = ("{category}", "{user[id]}", "{album[id]}")
+    pattern = BASE_PATTERN + r"/album(-?\d+)_(\d+)$"
+    test = (
+        ("https://vk.com/album221469416_0", {
+            "count": 3,
+        }),
+        ("https://vk.com/album-165740836_281339889", {
+            "count": 12,
+        }),
+    )
+
+    def __init__(self, match):
+        VkExtractor.__init__(self, match)
+        self.user_id, self.album_id = match.groups()
+
+    def photos(self):
+        url = "{}/album{}_{}".format(self.root, self.user_id, self.album_id)
+        return self._pagination(url, self.user_id)
+
+    def metadata(self):
+        return {
+            "user": {"id": self.user_id},
+            "album": {"id": self.album_id},
+        }