[vk] improve extractor (#474)

- fetch all photos
- add 'metadata' option
- fix extracting photos without '?' in URL

docs/supportedsites.md

@@ -694,7 +694,7 @@ Consider all sites to be NSFW unless otherwise known.
 <tr>
     <td>VK</td>
     <td>https://vk.com/</td>
-    <td>Albums</td>
+    <td>Photos</td>
     <td></td>
 </tr>
 <tr>

gallery_dl/extractor/vk.py

@@ -8,20 +8,21 @@
 
 """Extractors for https://vk.com/"""
 
-from .common import GalleryExtractor
+from .common import Extractor, Message
 from .. import text
 import re
 
 
-class VkAlbumExtractor(GalleryExtractor):
+class VkPhotosExtractor(Extractor):
-    """Extractor for vkontakte albums"""
+    """Extractor for photos from a vk user"""
     category = "vk"
-    subcategory = "album"
+    subcategory = "photos"
-    directory_fmt = ("{category}", "{album_id}")
+    directory_fmt = ("{category}", "{user[id]}")
     filename_fmt = "{id}.{extension}"
     archive_fmt = "{id}"
-    root = "https://vk.com/"
+    root = "https://vk.com"
-    pattern = r"(?:https://)?(?:www\.|m\.)?vk\.com/(?:albums|id)(\d+)"
+    request_interval = 1.0
+    pattern = r"(?:https://)?(?:www\.|m\.)?vk\.com/(?:albums|photos|id)(\d+)"
     test = (
         ("https://vk.com/id398982326", {
             "pattern": r"https://sun\d+-\d+\.userapi\.com/c\d+/v\d+"
@@ -29,27 +30,59 @@ class VkAlbumExtractor(GalleryExtractor):
             "count": ">= 35",
         }),
         ("https://m.vk.com/albums398982326"),
-        ("https://www.vk.com/id398982326"),
+        ("https://www.vk.com/id398982326?profile=1"),
     )
 
     def __init__(self, match):
-        self.album_id = match.group(1)
+        Extractor.__init__(self, match)
-        url = "{}/albums{}".format(self.root, self.album_id)
+        self.user_id = match.group(1)
-        GalleryExtractor.__init__(self, match, url)
 
-    def metadata(self, page):
+    def items(self):
-        return {
+        user_id = self.user_id
-            "album_id": self.album_id,
+
+        if self.config("metadata"):
+            url = "{}/id{}".format(self.root, user_id)
+            extr = text.extract_from(self.request(url).text)
+            data = {"user": {
+                "id"  : user_id,
+                "nick": text.unescape(extr(
+                    "<title>", " | VK<")),
+                "name": text.unescape(extr(
+                    '<h1 class="page_name">', "<")).replace("  ", " "),
+                "info": text.unescape(text.remove_html(extr(
+                    '<span class="current_text">', '</span')))
+            }}
+        else:
+            data = {"user": {"id": user_id}}
+
+        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",
         }
 
-    def images(self, page):
+        yield Message.Directory, data
-        results = []
         sub = re.compile(r"/imp[fg]/").sub
-        needle = 'data-id="{}_'.format(self.album_id)
+        needle = 'data-id="{}_'.format(user_id)
 
-        for photo in text.extract_iter(page, needle, '?'):
+        while True:
-            photo_id = photo.partition('"')[0]
+            offset, html = self.request(
-            url = sub("/", photo.rpartition("(")[2])
+                photos_url, method="POST", headers=headers, data=params
-            results.append((url, {"id": photo_id}))
+            ).json()["payload"][1]
 
-        return results
+            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