[vimeo:likes] Support large like lists (Fixes #3847)

test/test_utils.py

@@ -22,7 +22,8 @@
     fix_xml_ampersands,
     get_meta_content,
     orderedSet,
-    PagedList,
+    OnDemandPagedList,
+    InAdvancePagedList,
     parse_duration,
     read_batch_urls,
     sanitize_filename,
@@ -246,10 +247,14 @@ def get_page(pagenum):
                 for i in range(firstid, upto):
                     yield i
 
-            pl = PagedList(get_page, pagesize)
+            pl = OnDemandPagedList(get_page, pagesize)
             got = pl.getslice(*sliceargs)
             self.assertEqual(got, expected)
 
+            iapl = InAdvancePagedList(get_page, size // pagesize + 1, pagesize)
+            got = iapl.getslice(*sliceargs)
+            self.assertEqual(got, expected)
+
         testPL(5, 2, (), [0, 1, 2, 3, 4])
         testPL(5, 2, (1,), [1, 2, 3, 4])
         testPL(5, 2, (2,), [2, 3, 4])

youtube_dl/extractor/vimeo.py

@@ -8,18 +8,19 @@
 from .common import InfoExtractor
 from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
+    clean_html,
     compat_HTTPError,
     compat_urllib_parse,
     compat_urllib_request,
-    clean_html,
-    get_element_by_attribute,
+    compat_urlparse,
     ExtractorError,
+    get_element_by_attribute,
+    InAdvancePagedList,
+    int_or_none,
     RegexNotFoundError,
-    smuggle_url,
     std_headers,
     unsmuggle_url,
     urlencode_postdata,
-    int_or_none,
 )
 
 
@@ -533,32 +534,55 @@ def _real_extract(self, url):
 
 
 class VimeoLikesIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes(?:$|[?#])'
+    _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
     IE_NAME = 'vimeo:likes'
     IE_DESC = 'Vimeo user likes'
     _TEST = {
-        'url': 'https://vimeo.com/user20132939/likes',
-        'playlist_mincount': 4,
-        'add_ies': ['Generic'],
+        'url': 'https://vimeo.com/user755559/likes/',
+        'playlist_mincount': 293,
         "info_dict": {
-            "description": "Videos Philipp Hagemeister likes on Vimeo.",
-            "title": "Vimeo / Philipp Hagemeister's likes",
-        },
-        'params': {
-            'extract_flat': False,
+            "description": "See all the videos urza likes",
+            "title": 'Videos urza likes',
         },
     }
 
     def _real_extract(self, url):
         user_id = self._match_id(url)
-        rss_url = '%s//vimeo.com/user%s/likes/rss' % (
-            self.http_scheme(), user_id)
-        surl = smuggle_url(rss_url, {
-            'force_videoid': '%s_likes' % user_id,
-            'to_generic': True,
-        })
+        webpage = self._download_webpage(url, user_id)
+        page_count = self._int(
+            self._search_regex(
+                r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
+                    .*?</a></li>\s*<li\s+class="pagination_next">
+                ''', webpage, 'page count'),
+            'page count', fatal=True)
+        PAGE_SIZE = 12
+        title = self._html_search_regex(
+            r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
+        description = self._html_search_meta('description', webpage)
+
+        def _get_page(idx):
+            page_url = '%s//vimeo.com/user%s/likes/page:%d/sort:date' % (
+                self.http_scheme(), user_id, idx + 1)
+            webpage = self._download_webpage(
+                page_url, user_id,
+                note='Downloading page %d/%d' % (idx + 1, page_count))
+            video_list = self._search_regex(
+                r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
+                webpage, 'video content')
+            paths = re.findall(
+                r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
+            for path in paths:
+                yield {
+                    '_type': 'url',
+                    'url': compat_urlparse.urljoin(page_url, path),
+                }
+
+        pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 
         return {
-            '_type': 'url',
-            'url': surl,
+            '_type': 'playlist',
+            'id': 'user%s_likes' % user_id,
+            'title': title,
+            'description': description,
+            'entries': pl,
         }

youtube_dl/extractor/youtube.py

@@ -26,7 +26,7 @@
     get_element_by_attribute,
     ExtractorError,
     int_or_none,
-    PagedList,
+    OnDemandPagedList,
     unescapeHTML,
     unified_strdate,
     orderedSet,
@@ -1341,7 +1341,7 @@ def download_page(pagenum):
                     'id': video_id,
                     'title': title,
                 }
-        url_results = PagedList(download_page, self._GDATA_PAGE_SIZE)
+        url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
 
         return self.playlist_result(url_results, playlist_title=username)
 

youtube_dl/utils.py

@@ -1384,14 +1384,16 @@ def check_executable(exe, args=[]):
 
 
 class PagedList(object):
-    def __init__(self, pagefunc, pagesize):
-        self._pagefunc = pagefunc
-        self._pagesize = pagesize
-
     def __len__(self):
         # This is only useful for tests
         return len(self.getslice())
 
+
+class OnDemandPagedList(PagedList):
+    def __init__(self, pagefunc, pagesize):
+        self._pagefunc = pagefunc
+        self._pagesize = pagesize
+
     def getslice(self, start=0, end=None):
         res = []
         for pagenum in itertools.count(start // self._pagesize):
@@ -1430,6 +1432,35 @@ def getslice(self, start=0, end=None):
         return res
 
 
+class InAdvancePagedList(PagedList):
+    def __init__(self, pagefunc, pagecount, pagesize):
+        self._pagefunc = pagefunc
+        self._pagecount = pagecount
+        self._pagesize = pagesize
+
+    def getslice(self, start=0, end=None):
+        res = []
+        start_page = start // self._pagesize
+        end_page = (
+            self._pagecount if end is None else (end // self._pagesize + 1))
+        skip_elems = start - start_page * self._pagesize
+        only_more = None if end is None else end - start
+        for pagenum in range(start_page, end_page):
+            page = list(self._pagefunc(pagenum))
+            if skip_elems:
+                page = page[skip_elems:]
+                skip_elems = None
+            if only_more is not None:
+                if len(page) < only_more:
+                    only_more -= len(page)
+                else:
+                    page = page[:only_more]
+                    res.extend(page)
+                    break
+            res.extend(page)
+        return res
+
+
 def uppercase_escape(s):
     unicode_escape = codecs.getdecoder('unicode_escape')
     return re.sub(