release 2013.10.07

Conflicts:
	youtube_dl/YoutubeDL.py
	youtube_dl/__init__.py

README.md

@@ -52,6 +52,9 @@ which means you can modify it, redistribute it or use it however you like.
     --datebefore DATE          download only videos uploaded before this date
     --dateafter DATE           download only videos uploaded after this date
     --no-playlist              download only the currently playing video
+    --age-limit YEARS          download only videos suitable for the given age
+    --download-archive FILE    Download only videos not present in the archive
+                               file. Record all downloaded videos in it.
 
 ## Download Options:
     -r, --rate-limit LIMIT     maximum download rate (e.g. 50k or 44.6m)

test/test_age_restriction.py

@@ -0,0 +1,53 @@
+#!/usr/bin/env python
+
+import sys
+import unittest
+
+# Allow direct execution
+import os
+sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl import YoutubeDL
+from helper import try_rm
+
+
+def _download_restricted(url, filename, age):
+    """ Returns true iff the file has been downloaded """
+
+    params = {
+        'age_limit': age,
+        'skip_download': True,
+        'writeinfojson': True,
+        "outtmpl": "%(id)s.%(ext)s",
+    }
+    ydl = YoutubeDL(params)
+    ydl.add_default_info_extractors()
+    json_filename = filename + '.info.json'
+    try_rm(json_filename)
+    ydl.download([url])
+    res = os.path.exists(json_filename)
+    try_rm(json_filename)
+    return res
+
+
+class TestAgeRestriction(unittest.TestCase):
+    def _assert_restricted(self, url, filename, age, old_age=None):
+        self.assertTrue(_download_restricted(url, filename, old_age))
+        self.assertFalse(_download_restricted(url, filename, age))
+
+    def test_youtube(self):
+        self._assert_restricted('07FYdnEawAQ', '07FYdnEawAQ.mp4', 10)
+
+    def test_youporn(self):
+        self._assert_restricted(
+            'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
+            '505835.mp4', 2, old_age=25)
+
+    def test_pornotube(self):
+        self._assert_restricted(
+            'http://pornotube.com/c/173/m/1689755/Marilyn-Monroe-Bathing',
+            '1689755.flv', 13)
+
+
+if __name__ == '__main__':
+    unittest.main()

youtube_dl/YoutubeDL.py

@@ -3,6 +3,7 @@
 
 from __future__ import absolute_import
 
+import errno
 import io
 import os
 import re
@@ -84,6 +85,11 @@ class YoutubeDL(object):
     cachedir:          Location of the cache files in the filesystem.
                        None to disable filesystem cache.
     noplaylist:        Download single video instead of a playlist if in doubt.
+    age_limit:         An integer representing the user's age in years.
+                       Unsuitable videos for the given age are skipped.
+    downloadarchive:   File name of a file where all downloads are recorded.
+                       Videos already present in the file are not downloaded
+                       again.
     
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
@@ -309,6 +315,13 @@ class YoutubeDL(object):
             dateRange = self.params.get('daterange', DateRange())
             if date not in dateRange:
                 return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
+        age_limit = self.params.get('age_limit')
+        if age_limit is not None:
+            if age_limit < info_dict.get('age_limit', 0):
+                return u'Skipping "' + title + '" because it is age restricted'
+        if self.in_download_archive(info_dict):
+            return (u'%(title)s has already been recorded in archive'
+                    % info_dict)
         return None
         
     def extract_info(self, url, download=True, ie_key=None, extra_info={}):
@@ -578,6 +591,8 @@ class YoutubeDL(object):
                     self.report_error(u'postprocessing: %s' % str(err))
                     return
 
+        self.record_download_archive(info_dict)
+
     def download(self, url_list):
         """Download a given list of URLs."""
         if len(url_list) > 1 and self.fixed_template():
@@ -617,3 +632,26 @@ class YoutubeDL(object):
                 os.remove(encodeFilename(filename))
             except (IOError, OSError):
                 self.report_warning(u'Unable to remove downloaded video file')
+
+    def in_download_archive(self, info_dict):
+        fn = self.params.get('download_archive')
+        if fn is None:
+            return False
+        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+        try:
+            with locked_file(fn, 'r', encoding='utf-8') as archive_file:
+                for line in archive_file:
+                    if line.strip() == vid_id:
+                        return True
+        except IOError as ioe:
+            if ioe.errno != errno.ENOENT:
+                raise
+        return False
+
+    def record_download_archive(self, info_dict):
+        fn = self.params.get('download_archive')
+        if fn is None:
+            return
+        vid_id = info_dict['extractor'] + u' ' + info_dict['id']
+        with locked_file(fn, 'a', encoding='utf-8') as archive_file:
+            archive_file.write(vid_id + u'\n')

youtube_dl/__init__.py

@@ -188,6 +188,12 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--datebefore', metavar='DATE', dest='datebefore', help='download only videos uploaded before this date', default=None)
     selection.add_option('--dateafter', metavar='DATE', dest='dateafter', help='download only videos uploaded after this date', default=None)
     selection.add_option('--no-playlist', action='store_true', dest='noplaylist', help='download only the currently playing video', default=False)
+    selection.add_option('--age-limit', metavar='YEARS', dest='age_limit',
+                         help='download only videos suitable for the given age',
+                         default=None, type=int)
+    selection.add_option('--download-archive', metavar='FILE',
+                         dest='download_archive',
+                         help='Download only videos not present in the archive file. Record all downloaded videos in it.')
 
 
     authentication.add_option('-u', '--username',
@@ -478,6 +484,8 @@ def _real_main(argv=None):
             if not ie._WORKING:
                 continue
             desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
+            if desc is False:
+                continue
             if hasattr(ie, 'SEARCH_KEY'):
                 _SEARCHES = (u'cute kittens', u'slithering pythons', u'falling cat', u'angry poodle', u'purple fish', u'running tortoise')
                 _COUNTS = (u'', u'5', u'10', u'all')
@@ -631,6 +639,8 @@ def _real_main(argv=None):
         'daterange': date,
         'cachedir': opts.cachedir,
         'youtube_print_sig_code': opts.youtube_print_sig_code,
+        'age_limit': opts.age_limit,
+        'download_archive': opts.download_archive,
         })
 
     if opts.verbose:

youtube_dl/extractor/__init__.py

@@ -141,6 +141,7 @@ from .youtube import (
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
     YoutubeRecommendedIE,
+    YoutubeTruncatedURLIE,
     YoutubeWatchLaterIE,
     YoutubeFavouritesIE,
 )

youtube_dl/extractor/common.py

@@ -54,6 +54,7 @@ class InfoExtractor(object):
     view_count:     How many users have watched the video on the platform.
     urlhandle:      [internal] The urlHandle to be used to download the file,
                     like returned by urllib.request.urlopen
+    age_limit:      Age restriction for the video, as an integer (years)
     formats:        A list of dictionaries for each format available, it must
                     be ordered from worst to best quality. Potential fields:
                     * url       Mandatory. The URL of the video file
@@ -318,6 +319,15 @@ class InfoExtractor(object):
                                         self._og_regex('video')],
                                        html, name, **kargs)
 
+    def _rta_search(self, html):
+        # See http://www.rtalabel.org/index.php?content=howtofaq#single
+        if re.search(r'(?ix)<meta\s+name="rating"\s+'
+                     r'     content="RTA-5042-1996-1400-1577-RTA"',
+                     html):
+            return 18
+        return 0
+
+
 class SearchInfoExtractor(InfoExtractor):
     """
     Base class for paged search queries extractors.

youtube_dl/extractor/generic.py

@@ -117,7 +117,7 @@ class GenericIE(InfoExtractor):
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Failed to download URL: %s' % url)
 
         self.report_extraction(video_id)
         # Look for BrightCove:
@@ -149,12 +149,12 @@ class GenericIE(InfoExtractor):
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Unsupported URL: %s' % url)
 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError(u'Did not find a valid video URL at %s' % url)
 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)

youtube_dl/extractor/jeuxvideo.py

@@ -6,6 +6,7 @@ import xml.etree.ElementTree
 
 from .common import InfoExtractor
 
+
 class JeuxVideoIE(InfoExtractor):
     _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)-\d+\.htm'
 
@@ -23,25 +24,29 @@ class JeuxVideoIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         title = re.match(self._VALID_URL, url).group(1)
         webpage = self._download_webpage(url, title)
-        m_download = re.search(r'<param name="flashvars" value="config=(.*?)" />', webpage)
-
-        xml_link = m_download.group(1)
+        xml_link = self._html_search_regex(
+            r'<param name="flashvars" value="config=(.*?)" />',
+            webpage, u'config URL')
         
-        id = re.search(r'http://www.jeuxvideo.com/config/\w+/0011/(.*?)/\d+_player\.xml', xml_link).group(1)
+        video_id = self._search_regex(
+            r'http://www\.jeuxvideo\.com/config/\w+/\d+/(.*?)/\d+_player\.xml',
+            xml_link, u'video ID')
 
-        xml_config = self._download_webpage(xml_link, title,
-                                                  'Downloading XML config')
+        xml_config = self._download_webpage(
+            xml_link, title, u'Downloading XML config')
         config = xml.etree.ElementTree.fromstring(xml_config.encode('utf-8'))
-        info = re.search(r'<format\.json>(.*?)</format\.json>',
-                         xml_config, re.MULTILINE|re.DOTALL).group(1)
-        info = json.loads(info)['versions'][0]
+        info_json = self._search_regex(
+            r'(?sm)<format\.json>(.*?)</format\.json>',
+            xml_config, u'JSON information')
+        info = json.loads(info_json)['versions'][0]
         
         video_url = 'http://video720.jeuxvideo.com/' + info['file']
 
-        return {'id': id,
-                'title' : config.find('titre_video').text,
-                'ext' : 'mp4',
-                'url' : video_url,
-                'description': self._og_search_description(webpage),
-                'thumbnail': config.find('image').text,
-                }
+        return {
+            'id': video_id,
+            'title': config.find('titre_video').text,
+            'ext': 'mp4',
+            'url': video_url,
+            'description': self._og_search_description(webpage),
+            'thumbnail': config.find('image').text,
+        }

youtube_dl/extractor/pornotube.py

@@ -38,6 +38,7 @@ class PornotubeIE(InfoExtractor):
         VIDEO_UPLOADED_RE = r'<div class="video_added_by">Added (?P<date>[0-9\/]+) by'
         upload_date = self._html_search_regex(VIDEO_UPLOADED_RE, webpage, u'upload date', fatal=False)
         if upload_date: upload_date = unified_strdate(upload_date)
+        age_limit = self._rta_search(webpage)
 
         info = {'id': video_id,
                 'url': video_url,
@@ -45,6 +46,7 @@ class PornotubeIE(InfoExtractor):
                 'upload_date': upload_date,
                 'title': video_title,
                 'ext': 'flv',
-                'format': 'flv'}
+                'format': 'flv',
+                'age_limit': age_limit}
 
         return [info]

youtube_dl/extractor/redtube.py

@@ -30,9 +30,14 @@ class RedTubeIE(InfoExtractor):
             r'<h1 class="videoTitle slidePanelMovable">(.+?)</h1>',
             webpage, u'title')
 
+        # No self-labeling, but they describe themselves as
+        # "Home of Videos Porno"
+        age_limit = 18
+
         return {
-            'id':       video_id,
-            'url':      video_url,
-            'ext':      video_extension,
-            'title':    video_title,
+            'id':        video_id,
+            'url':       video_url,
+            'ext':       video_extension,
+            'title':     video_title,
+            'age_limit': age_limit,
         }

youtube_dl/extractor/vimeo.py

@@ -17,7 +17,7 @@ class VimeoIE(InfoExtractor):
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs
-    _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)(?:[?].*)?$'
+    _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)/?(?:[?].*)?$'
     _NETRC_MACHINE = 'vimeo'
     IE_NAME = u'vimeo'
     _TESTS = [

youtube_dl/extractor/youporn.py

@@ -51,6 +51,7 @@ class YouPornIE(InfoExtractor):
         req = compat_urllib_request.Request(url)
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
+        age_limit = self._rta_search(webpage)
 
         # Get JSON parameters
         json_params = self._search_regex(r'var currentVideo = new Video\((.*)\);', webpage, u'JSON parameters')
@@ -115,7 +116,8 @@ class YouPornIE(InfoExtractor):
                 'ext': extension,
                 'format': format,
                 'thumbnail': thumbnail,
-                'description': video_description
+                'description': video_description,
+                'age_limit': age_limit,
             })
 
         if self._downloader.params.get('listformats', None):

youtube_dl/extractor/youtube.py

@@ -1250,9 +1250,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         return url_map
 
     def _real_extract(self, url):
-        if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
-            self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like  youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply  youtube-dl BaW_jenozKc  ).')
-
         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
         mobj = re.search(self._NEXT_URL_RE, url)
         if mobj:
@@ -1495,7 +1492,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'description':  video_description,
                 'player_url':   player_url,
                 'subtitles':    video_subtitles,
-                'duration':     video_duration
+                'duration':     video_duration,
+                'age_limit':    18 if age_gate else 0,
             })
         return results
 
@@ -1636,7 +1634,7 @@ class YoutubeChannelIE(InfoExtractor):
 
 class YoutubeUserIE(InfoExtractor):
     IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
-    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?)|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
+    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!watch(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
     _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
     _GDATA_PAGE_SIZE = 50
     _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
@@ -1829,3 +1827,18 @@ class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
         webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
         playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
         return self.url_result(playlist_id, 'YoutubePlaylist')
+
+
+class YoutubeTruncatedURLIE(InfoExtractor):
+    IE_NAME = 'youtube:truncated_url'
+    IE_DESC = False  # Do not list
+    _VALID_URL = r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$'
+
+    def _real_extract(self, url):
+        raise ExtractorError(
+            u'Did you forget to quote the URL? Remember that & is a meta '
+            u'character in most shells, so you want to put the URL in quotes, '
+            u'like  youtube-dl '
+            u'\'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\''
+            u' (or simply  youtube-dl BaW_jenozKc  ).',
+            expected=True)

youtube_dl/utils.py

@@ -830,3 +830,99 @@ def get_cachedir(params={}):
     cache_root = os.environ.get('XDG_CACHE_HOME',
                                 os.path.expanduser('~/.cache'))
     return params.get('cachedir', os.path.join(cache_root, 'youtube-dl'))
+
+
+# Cross-platform file locking
+if sys.platform == 'win32':
+    import ctypes.wintypes
+    import msvcrt
+
+    class OVERLAPPED(ctypes.Structure):
+        _fields_ = [
+            ('Internal', ctypes.wintypes.LPVOID),
+            ('InternalHigh', ctypes.wintypes.LPVOID),
+            ('Offset', ctypes.wintypes.DWORD),
+            ('OffsetHigh', ctypes.wintypes.DWORD),
+            ('hEvent', ctypes.wintypes.HANDLE),
+        ]
+
+    kernel32 = ctypes.windll.kernel32
+    LockFileEx = kernel32.LockFileEx
+    LockFileEx.argtypes = [
+        ctypes.wintypes.HANDLE,     # hFile
+        ctypes.wintypes.DWORD,      # dwFlags
+        ctypes.wintypes.DWORD,      # dwReserved
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockLow
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockHigh
+        ctypes.POINTER(OVERLAPPED)  # Overlapped
+    ]
+    LockFileEx.restype = ctypes.wintypes.BOOL
+    UnlockFileEx = kernel32.UnlockFileEx
+    UnlockFileEx.argtypes = [
+        ctypes.wintypes.HANDLE,     # hFile
+        ctypes.wintypes.DWORD,      # dwReserved
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockLow
+        ctypes.wintypes.DWORD,      # nNumberOfBytesToLockHigh
+        ctypes.POINTER(OVERLAPPED)  # Overlapped
+    ]
+    UnlockFileEx.restype = ctypes.wintypes.BOOL
+    whole_low = 0xffffffff
+    whole_high = 0x7fffffff
+
+    def _lock_file(f, exclusive):
+        overlapped = OVERLAPPED()
+        overlapped.Offset = 0
+        overlapped.OffsetHigh = 0
+        overlapped.hEvent = 0
+        f._lock_file_overlapped_p = ctypes.pointer(overlapped)
+        handle = msvcrt.get_osfhandle(f.fileno())
+        if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0,
+                          whole_low, whole_high, f._lock_file_overlapped_p):
+            raise OSError('Locking file failed: %r' % ctypes.FormatError())
+
+    def _unlock_file(f):
+        assert f._lock_file_overlapped_p
+        handle = msvcrt.get_osfhandle(f.fileno())
+        if not UnlockFileEx(handle, 0,
+                            whole_low, whole_high, f._lock_file_overlapped_p):
+            raise OSError('Unlocking file failed: %r' % ctypes.FormatError())
+
+else:
+    import fcntl
+
+    def _lock_file(f, exclusive):
+        fcntl.lockf(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
+
+    def _unlock_file(f):
+        fcntl.lockf(f, fcntl.LOCK_UN)
+
+
+class locked_file(object):
+    def __init__(self, filename, mode, encoding=None):
+        assert mode in ['r', 'a', 'w']
+        self.f = io.open(filename, mode, encoding=encoding)
+        self.mode = mode
+
+    def __enter__(self):
+        exclusive = self.mode != 'r'
+        try:
+            _lock_file(self.f, exclusive)
+        except IOError:
+            self.f.close()
+            raise
+        return self
+
+    def __exit__(self, etype, value, traceback):
+        try:
+            _unlock_file(self.f)
+        finally:
+            self.f.close()
+
+    def __iter__(self):
+        return iter(self.f)
+
+    def write(self, *args):
+        return self.f.write(*args)
+
+    def read(self, *args):
+        return self.f.read(*args)

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2013.10.06'
+__version__ = '2013.10.07'