[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / vidme.py
index 296e00423c0288418cf958f9e1e06a8fe5500a26..174e69cd6b1e8ce1e5bf829781140ea756bdc6bf 100644 (file)
@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
 
+import itertools
+
 from .common import InfoExtractor
 from ..compat import compat_HTTPError
 from ..utils import (
@@ -7,20 +9,22 @@ from ..utils import (
     int_or_none,
     float_or_none,
     parse_iso8601,
+    url_or_none,
 )
 
 
 class VidmeIE(InfoExtractor):
-    _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)'
+    IE_NAME = 'vidme'
+    _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]{,5})(?:[^\da-zA-Z]|$)'
     _TESTS = [{
         'url': 'https://vid.me/QNB',
-        'md5': 'c62f1156138dc3323902188c5b5a8bd6',
+        'md5': 'f42d05e7149aeaec5c037b17e5d3dc82',
         'info_dict': {
             'id': 'QNB',
             'ext': 'mp4',
             'title': 'Fishing for piranha - the easy way',
             'description': 'source: https://www.facebook.com/photo.php?v=312276045600871',
-            'thumbnail': 're:^https?://.*\.jpg',
+            'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1406313244,
             'upload_date': '20140725',
             'age_limit': 0,
@@ -36,7 +40,7 @@ class VidmeIE(InfoExtractor):
             'id': 'Gc6M',
             'ext': 'mp4',
             'title': 'O Mere Dil ke chain - Arnav and Khushi VM',
-            'thumbnail': 're:^https?://.*\.jpg',
+            'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1441211642,
             'upload_date': '20150902',
             'uploader': 'SunshineM',
@@ -58,7 +62,7 @@ class VidmeIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'The Carver',
             'description': 'md5:e9c24870018ae8113be936645b93ba3c',
-            'thumbnail': 're:^https?://.*\.jpg',
+            'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1433203629,
             'upload_date': '20150602',
             'uploader': 'Thomas',
@@ -79,7 +83,7 @@ class VidmeIE(InfoExtractor):
             'id': 'Wmur',
             'ext': 'mp4',
             'title': 'naked smoking & stretching',
-            'thumbnail': 're:^https?://.*\.jpg',
+            'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1430931613,
             'upload_date': '20150506',
             'uploader': 'naked-yogi',
@@ -101,6 +105,10 @@ class VidmeIE(InfoExtractor):
         # suspended
         'url': 'https://vid.me/Ox3G',
         'only_matching': True,
+    }, {
+        # deleted
+        'url': 'https://vid.me/KTPm',
+        'only_matching': True,
     }, {
         # no formats in the API response
         'url': 'https://vid.me/e5g',
@@ -108,7 +116,7 @@ class VidmeIE(InfoExtractor):
             'id': 'e5g',
             'ext': 'mp4',
             'title': 'Video upload (e5g)',
-            'thumbnail': 're:^https?://.*\.jpg',
+            'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1401480195,
             'upload_date': '20140530',
             'uploader': None,
@@ -143,19 +151,39 @@ class VidmeIE(InfoExtractor):
 
         video = response['video']
 
+        if video.get('state') == 'deleted':
+            raise ExtractorError(
+                'Vidme said: Sorry, this video has been deleted.',
+                expected=True)
+
         if video.get('state') in ('user-disabled', 'suspended'):
             raise ExtractorError(
                 'Vidme said: This video has been suspended either due to a copyright claim, '
                 'or for violating the terms of use.',
                 expected=True)
 
-        formats = [{
-            'format_id': f.get('type'),
-            'url': f['uri'],
-            'width': int_or_none(f.get('width')),
-            'height': int_or_none(f.get('height')),
-            'preference': 0 if f.get('type', '').endswith('clip') else 1,
-        } for f in video.get('formats', []) if f.get('uri')]
+        formats = []
+        for f in video.get('formats', []):
+            format_url = url_or_none(f.get('uri'))
+            if not format_url:
+                continue
+            format_type = f.get('type')
+            if format_type == 'dash':
+                formats.extend(self._extract_mpd_formats(
+                    format_url, video_id, mpd_id='dash', fatal=False))
+            elif format_type == 'hls':
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'format_id': f.get('type'),
+                    'url': format_url,
+                    'width': int_or_none(f.get('width')),
+                    'height': int_or_none(f.get('height')),
+                    'preference': 0 if f.get('type', '').endswith(
+                        'clip') else 1,
+                })
 
         if not formats and video.get('complete_url'):
             formats.append({
@@ -193,3 +221,75 @@ class VidmeIE(InfoExtractor):
             'comment_count': comment_count,
             'formats': formats,
         }
+
+
+class VidmeListBaseIE(InfoExtractor):
+    # Max possible limit according to https://docs.vid.me/#api-Videos-List
+    _LIMIT = 100
+
+    def _entries(self, user_id, user_name):
+        for page_num in itertools.count(1):
+            page = self._download_json(
+                'https://api.vid.me/videos/%s?user=%s&limit=%d&offset=%d'
+                % (self._API_ITEM, user_id, self._LIMIT, (page_num - 1) * self._LIMIT),
+                user_name, 'Downloading user %s page %d' % (self._API_ITEM, page_num))
+
+            videos = page.get('videos', [])
+            if not videos:
+                break
+
+            for video in videos:
+                video_url = video.get('full_url') or video.get('embed_url')
+                if video_url:
+                    yield self.url_result(video_url, VidmeIE.ie_key())
+
+            total = int_or_none(page.get('page', {}).get('total'))
+            if total and self._LIMIT * page_num >= total:
+                break
+
+    def _real_extract(self, url):
+        user_name = self._match_id(url)
+
+        user_id = self._download_json(
+            'https://api.vid.me/userByUsername?username=%s' % user_name,
+            user_name)['user']['user_id']
+
+        return self.playlist_result(
+            self._entries(user_id, user_name), user_id,
+            '%s - %s' % (user_name, self._TITLE))
+
+
+class VidmeUserIE(VidmeListBaseIE):
+    IE_NAME = 'vidme:user'
+    _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z_-]{6,})(?!/likes)(?:[^\da-zA-Z_-]|$)'
+    _API_ITEM = 'list'
+    _TITLE = 'Videos'
+    _TESTS = [{
+        'url': 'https://vid.me/MasakoX',
+        'info_dict': {
+            'id': '16112341',
+            'title': 'MasakoX - %s' % _TITLE,
+        },
+        'playlist_mincount': 191,
+    }, {
+        'url': 'https://vid.me/unsQuare_netWork',
+        'only_matching': True,
+    }]
+
+
+class VidmeUserLikesIE(VidmeListBaseIE):
+    IE_NAME = 'vidme:user:likes'
+    _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z_-]{6,})/likes'
+    _API_ITEM = 'likes'
+    _TITLE = 'Likes'
+    _TESTS = [{
+        'url': 'https://vid.me/ErinAlexis/likes',
+        'info_dict': {
+            'id': '6483530',
+            'title': 'ErinAlexis - %s' % _TITLE,
+        },
+        'playlist_mincount': 415,
+    }, {
+        'url': 'https://vid.me/Kaleidoscope-Ish/likes',
+        'only_matching': True,
+    }]