Merge branch 'master' into use-other-downloaders

.gitignore

@@ -17,3 +17,4 @@ youtube-dl.tar.gz
 .coverage
 cover/
 updates_key.pem
+*.egg-info

Makefile

@@ -7,6 +7,7 @@ PREFIX=/usr/local
 BINDIR=$(PREFIX)/bin
 MANDIR=$(PREFIX)/man
 SYSCONFDIR=/etc
+PYTHON=/usr/bin/env python
 
 install: youtube-dl youtube-dl.1 youtube-dl.bash-completion
 	install -d $(DESTDIR)$(BINDIR)
@@ -27,7 +28,7 @@ tar: youtube-dl.tar.gz
 youtube-dl: youtube_dl/*.py
 	zip --quiet youtube-dl youtube_dl/*.py
 	zip --quiet --junk-paths youtube-dl youtube_dl/__main__.py
-	echo '#!/usr/bin/env python' > youtube-dl
+	echo '#!$(PYTHON)' > youtube-dl
 	cat youtube-dl.zip >> youtube-dl
 	rm youtube-dl.zip
 	chmod a+x youtube-dl

README.md

@@ -38,6 +38,10 @@ which means you can modify it, redistribute it or use it however you like.
     --reject-title REGEX     skip download for matching titles (regex or
                              caseless sub-string)
     --max-downloads NUMBER   Abort after downloading NUMBER files
+    --min-filesize SIZE      Do not download any videos smaller than SIZE (e.g.
+                             50k or 44.6m)
+    --max-filesize SIZE      Do not download any videos larger than SIZE (e.g.
+                             50k or 44.6m)
 
 ## Filesystem Options:
     -t, --title              use title in file name

devscripts/release.sh

@@ -83,4 +83,7 @@ ROOT=$(pwd)
 )
 rm -rf build
 
+echo "Uploading to PyPi ..."
+python setup.py sdist upload
+
 echo "\n### DONE!"

setup.py

@@ -2,10 +2,14 @@
 # -*- coding: utf-8 -*-
 
 from __future__ import print_function
-from distutils.core import setup
 import pkg_resources
 import sys
 
+try:
+    from setuptools import setup
+except ImportError:
+    from distutils.core import setup
+
 try:
     import py2exe
     """This will create an exe that needs Microsoft Visual C++ 2008 Redistributable Package"""

test/test_download.py

@@ -98,7 +98,7 @@ def generator(test_case):
 
             for tc in test_cases:
                 if not test_case.get('params', {}).get('skip_download', False):
-                    self.assertTrue(os.path.exists(tc['file']))
+                    self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
                     self.assertTrue(tc['file'] in finished_hook_called)
                 self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
                 if 'md5' in tc:
@@ -107,11 +107,7 @@ def generator(test_case):
                 with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
                     info_dict = json.load(infof)
                 for (info_field, value) in tc.get('info_dict', {}).items():
-                    if value.startswith('md5:'):
+                    self.assertEqual(value, info_dict.get(info_field))
-                        md5_info_value = hashlib.md5(info_dict.get(info_field, '')).hexdigest()
-                        self.assertEqual(value[3:], md5_info_value)
-                    else:
-                        self.assertEqual(value, info_dict.get(info_field))
         finally:
             for tc in test_cases:
                 _try_rm(tc['file'])

test/tests.json

@@ -76,7 +76,8 @@
     "name": "StanfordOpenClassroom",
     "md5":  "544a9468546059d4e80d76265b0443b8",
     "url":  "http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100",
-    "file":  "PracticalUnix_intro-environment.mp4"
+    "file":  "PracticalUnix_intro-environment.mp4",
+    "skip": "Currently offline"
   },
   {
     "name": "XNXX",
@@ -181,37 +182,12 @@
   },
   {
     "name": "ComedyCentral",
-    "url": "http://www.thedailyshow.com/full-episodes/thu-december-13-2012-kristen-stewart",
+    "url": "http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart",
-    "playlist": [
+    "file": "422212.mp4",
-      {
+    "md5": "4e2f5cb088a83cd8cdb7756132f9739d",
-        "file": "422204.mp4",
+    "info_dict": {
-        "md5": "7a7abe068b31ff03e7b8a37596e72380",
+        "title": "thedailyshow-kristen-stewart part 1"
-        "info_dict": {
+    }
-            "title": "thedailyshow-thu-december-13-2012-kristen-stewart part 1"
-        }
-      },
-      {
-        "file": "422205.mp4",
-        "md5": "30552b7274c94dbb933f64600eadddd2",
-        "info_dict": {
-            "title": "thedailyshow-thu-december-13-2012-kristen-stewart part 2"
-        }
-      },
-      {
-        "file": "422206.mp4",
-        "md5": "1f4c0664b352cb8e8fe85d5da4fbee91",
-        "info_dict": {
-            "title": "thedailyshow-thu-december-13-2012-kristen-stewart part 3"
-        }
-      },
-      {
-        "file": "422207.mp4",
-        "md5": "f61ee8a4e6bd1308438e03badad78554",
-        "info_dict": {
-            "title": "thedailyshow-thu-december-13-2012-kristen-stewart part 4"
-        }
-      }
-    ]
   },
   {
     "name": "RBMARadio",
@@ -225,5 +201,80 @@
         "uploader_id": "ford-lopatin",
         "location": "Spain"
     }
+  },
+  {
+    "name": "Facebook",
+    "url": "https://www.facebook.com/photo.php?v=120708114770723",
+    "file": "120708114770723.mp4",
+    "md5": "48975a41ccc4b7a581abd68651c1a5a8",
+    "info_dict": {
+      "title": "PEOPLE ARE AWESOME 2013",
+      "duration": 279
+    }
+  },
+  {
+    "name": "EightTracks",
+    "url": "http://8tracks.com/ytdl/youtube-dl-test-tracks-a",
+    "playlist": [
+      {
+        "file": "11885610.m4a",
+        "md5": "96ce57f24389fc8734ce47f4c1abcc55",
+        "info_dict": {
+          "title": "youtue-dl project<>\"' - youtube-dl test track 1 \"'/\\\u00e4\u21ad",
+          "uploader_id": "ytdl"
+        }
+      },
+      {
+        "file": "11885608.m4a",
+        "md5": "4ab26f05c1f7291ea460a3920be8021f",
+        "info_dict": {
+          "title": "youtube-dl project - youtube-dl test track 2 \"'/\\\u00e4\u21ad",
+          "uploader_id": "ytdl"
+
+        }
+      },
+      {
+        "file": "11885679.m4a",
+        "md5": "d30b5b5f74217410f4689605c35d1fd7",
+        "info_dict": {
+          "title": "youtube-dl project as well - youtube-dl test track 3 \"'/\\\u00e4\u21ad"
+        }
+      },
+      {
+        "file": "11885680.m4a",
+        "md5": "4eb0a669317cd725f6bbd336a29f923a",
+        "info_dict": {
+          "title": "youtube-dl project as well - youtube-dl test track 4 \"'/\\\u00e4\u21ad"
+        }
+      },
+      {
+        "file": "11885682.m4a",
+        "md5": "1893e872e263a2705558d1d319ad19e8",
+        "info_dict": {
+          "title": "PH - youtube-dl test track 5 \"'/\\\u00e4\u21ad"
+        }
+      },
+      {
+        "file": "11885683.m4a",
+        "md5": "b673c46f47a216ab1741ae8836af5899",
+        "info_dict": {
+          "title": "PH - youtube-dl test track 6 \"'/\\\u00e4\u21ad"
+        }
+      },
+      {
+        "file": "11885684.m4a",
+        "md5": "1d74534e95df54986da7f5abf7d842b7",
+        "info_dict": {
+          "title": "phihag - youtube-dl test track 7 \"'/\\\u00e4\u21ad"
+        }
+      },
+      {
+        "file": "11885685.m4a",
+        "md5": "f081f47af8f6ae782ed131d38b9cd1c0",
+        "info_dict": {
+          "title": "phihag - youtube-dl test track 8 \"'/\\\u00e4\u21ad"
+        }
+      }
+    ]
   }
 ]

youtube_dl/FileDownloader.py

@@ -82,6 +82,8 @@ class FileDownloader(object):
     subtitleslang:     Language of the subtitles to download
     test:              Download only first bytes to test the downloader.
     keepvideo:         Keep the video file after post-processing
+    min_filesize:      Skip files smaller than this size
+    max_filesize:      Skip files larger than this size
     """
 
     params = None
@@ -767,6 +769,15 @@ class FileDownloader(object):
         data_len = data.info().get('Content-length', None)
         if data_len is not None:
             data_len = int(data_len) + resume_len
+            min_data_len = self.params.get("min_filesize", None)
+            max_data_len =  self.params.get("max_filesize", None)
+            if min_data_len is not None and data_len < min_data_len:
+                self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
+                return False
+            if max_data_len is not None and data_len > max_data_len:
+                self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
+                return False
+
         data_len_str = self.format_bytes(data_len)
         byte_counter = 0 + resume_len
         block_size = self.params.get('buffersize', 1024)

youtube_dl/InfoExtractors.py

@@ -5,6 +5,7 @@ from __future__ import absolute_import
 
 import base64
 import datetime
+import itertools
 import netrc
 import os
 import re
@@ -263,13 +264,18 @@ class YoutubeIE(InfoExtractor):
             srt_lang = list(srt_lang_list.keys())[0]
         if not srt_lang in srt_lang_list:
             return (u'WARNING: no closed captions found in the specified language', None)
-        request = compat_urllib_request.Request('http://www.youtube.com/api/timedtext?lang=%s&name=%s&v=%s' % (srt_lang, srt_lang_list[srt_lang], video_id))
+        params = compat_urllib_parse.urlencode({
+            'lang': srt_lang,
+            'name': srt_lang_list[srt_lang].encode('utf-8'),
+            'v': video_id,
+        })
+        url = 'http://www.youtube.com/api/timedtext?' + params
         try:
-            srt_xml = compat_urllib_request.urlopen(request).read().decode('utf-8')
+            srt_xml = compat_urllib_request.urlopen(url).read().decode('utf-8')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             return (u'WARNING: unable to download video subtitles: %s' % compat_str(err), None)
         if not srt_xml:
-            return (u'WARNING: unable to download video subtitles', None)
+            return (u'WARNING: Did not fetch video subtitles', None)
         return (None, self._closed_captions_xml_to_srt(srt_xml))
 
     def _print_formats(self, formats):
@@ -972,7 +978,7 @@ class VimeoIE(InfoExtractor):
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs
-    _VALID_URL = r'(?:https?://)?(?:(?:www|player).)?vimeo\.com/(?:(?:groups|album)/[^/]+/)?(?:videos?/)?([0-9]+)'
+    _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo\.com/(?:(?:groups|album)/[^/]+/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)'
     IE_NAME = u'vimeo'
 
     def __init__(self, downloader=None):
@@ -993,7 +999,11 @@ class VimeoIE(InfoExtractor):
             self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
             return
 
-        video_id = mobj.group(1)
+        video_id = mobj.group('id')
+        if not mobj.group('proto'):
+            url = 'https://' + url
+        if mobj.group('direct_link'):
+            url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
         request = compat_urllib_request.Request(url, None, std_headers)
@@ -1980,62 +1990,14 @@ class DepositFilesIE(InfoExtractor):
 class FacebookIE(InfoExtractor):
     """Information Extractor for Facebook"""
 
-    _WORKING = False
     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
     _LOGIN_URL = 'https://login.facebook.com/login.php?m&next=http%3A%2F%2Fm.facebook.com%2Fhome.php&'
     _NETRC_MACHINE = 'facebook'
-    _available_formats = ['video', 'highqual', 'lowqual']
-    _video_extensions = {
-        'video': 'mp4',
-        'highqual': 'mp4',
-        'lowqual': 'mp4',
-    }
     IE_NAME = u'facebook'
 
-    def __init__(self, downloader=None):
-        InfoExtractor.__init__(self, downloader)
-
-    def _reporter(self, message):
-        """Add header and report message."""
-        self._downloader.to_screen(u'[facebook] %s' % message)
-
     def report_login(self):
         """Report attempt to log in."""
-        self._reporter(u'Logging in')
+        self._downloader.to_screen(u'[%s] Logging in' % self.IE_NAME)
-
-    def report_video_webpage_download(self, video_id):
-        """Report attempt to download video webpage."""
-        self._reporter(u'%s: Downloading video webpage' % video_id)
-
-    def report_information_extraction(self, video_id):
-        """Report attempt to extract video information."""
-        self._reporter(u'%s: Extracting video information' % video_id)
-
-    def _parse_page(self, video_webpage):
-        """Extract video information from page"""
-        # General data
-        data = {'title': r'\("video_title", "(.*?)"\)',
-            'description': r'<div class="datawrap">(.*?)</div>',
-            'owner': r'\("video_owner_name", "(.*?)"\)',
-            'thumbnail':  r'\("thumb_url", "(?P<THUMB>.*?)"\)',
-            }
-        video_info = {}
-        for piece in data.keys():
-            mobj = re.search(data[piece], video_webpage)
-            if mobj is not None:
-                video_info[piece] = compat_urllib_parse.unquote_plus(mobj.group(1).decode("unicode_escape"))
-
-        # Video urls
-        video_urls = {}
-        for fmt in self._available_formats:
-            mobj = re.search(r'\("%s_src\", "(.+?)"\)' % fmt, video_webpage)
-            if mobj is not None:
-                # URL is in a Javascript segment inside an escaped Unicode format within
-                # the generally utf-8 page
-                video_urls[fmt] = compat_urllib_parse.unquote_plus(mobj.group(1).decode("unicode_escape"))
-        video_info['video_urls'] = video_urls
-
-        return video_info
 
     def _real_initialize(self):
         if self._downloader is None:
@@ -2088,100 +2050,35 @@ class FacebookIE(InfoExtractor):
             return
         video_id = mobj.group('ID')
 
-        # Get video webpage
+        url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
-        self.report_video_webpage_download(video_id)
+        webpage = self._download_webpage(url, video_id)
-        request = compat_urllib_request.Request('https://www.facebook.com/video/video.php?v=%s' % video_id)
-        try:
-            page = compat_urllib_request.urlopen(request)
-            video_webpage = page.read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % compat_str(err))
-            return
 
-        # Start extracting information
+        BEFORE = '[["allowFullScreen","true"],["allowScriptAccess","always"],["salign","tl"],["scale","noscale"],["wmode","opaque"]].forEach(function(param) {swf.addParam(param[0], param[1]);});\n'
-        self.report_information_extraction(video_id)
+        AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
+        m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
+        if not m:
+            raise ExtractorError(u'Cannot parse data')
+        data = dict(json.loads(m.group(1)))
+        params_raw = compat_urllib_parse.unquote(data['params'])
+        params = json.loads(params_raw)
+        video_url = params['hd_src']
+        video_duration = int(params['video_duration'])
 
-        # Extract information
+        m = re.search('<h2 class="uiHeaderTitle">([^<]+)</h2>', webpage)
-        video_info = self._parse_page(video_webpage)
+        if not m:
+            raise ExtractorError(u'Cannot find title in webpage')
+        video_title = unescapeHTML(m.group(1))
 
-        # uploader
+        info = {
-        if 'owner' not in video_info:
+            'id': video_id,
-            self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
+            'title': video_title,
-            return
+            'url': video_url,
-        video_uploader = video_info['owner']
+            'ext': 'mp4',
+            'duration': video_duration,
+            'thumbnail': params['thumbnail_src'],
+        }
+        return [info]
 
-        # title
-        if 'title' not in video_info:
-            self._downloader.trouble(u'ERROR: unable to extract video title')
-            return
-        video_title = video_info['title']
-        video_title = video_title.decode('utf-8')
-
-        # thumbnail image
-        if 'thumbnail' not in video_info:
-            self._downloader.trouble(u'WARNING: unable to extract video thumbnail')
-            video_thumbnail = ''
-        else:
-            video_thumbnail = video_info['thumbnail']
-
-        # upload date
-        upload_date = None
-        if 'upload_date' in video_info:
-            upload_time = video_info['upload_date']
-            timetuple = email.utils.parsedate_tz(upload_time)
-            if timetuple is not None:
-                try:
-                    upload_date = time.strftime('%Y%m%d', timetuple[0:9])
-                except:
-                    pass
-
-        # description
-        video_description = video_info.get('description', 'No description available.')
-
-        url_map = video_info['video_urls']
-        if url_map:
-            # Decide which formats to download
-            req_format = self._downloader.params.get('format', None)
-            format_limit = self._downloader.params.get('format_limit', None)
-
-            if format_limit is not None and format_limit in self._available_formats:
-                format_list = self._available_formats[self._available_formats.index(format_limit):]
-            else:
-                format_list = self._available_formats
-            existing_formats = [x for x in format_list if x in url_map]
-            if len(existing_formats) == 0:
-                self._downloader.trouble(u'ERROR: no known formats available for video')
-                return
-            if req_format is None:
-                video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
-            elif req_format == 'worst':
-                video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality
-            elif req_format == '-1':
-                video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
-            else:
-                # Specific format
-                if req_format not in url_map:
-                    self._downloader.trouble(u'ERROR: requested format not available')
-                    return
-                video_url_list = [(req_format, url_map[req_format])] # Specific format
-
-        results = []
-        for format_param, video_real_url in video_url_list:
-            # Extension
-            video_extension = self._video_extensions.get(format_param, 'mp4')
-
-            results.append({
-                'id':       video_id.decode('utf-8'),
-                'url':      video_real_url.decode('utf-8'),
-                'uploader': video_uploader.decode('utf-8'),
-                'upload_date':  upload_date,
-                'title':    video_title,
-                'ext':      video_extension.decode('utf-8'),
-                'format':   (format_param is None and u'NA' or format_param.decode('utf-8')),
-                'thumbnail':    video_thumbnail.decode('utf-8'),
-                'description':  video_description.decode('utf-8'),
-            })
-        return results
 
 class BlipTVIE(InfoExtractor):
     """Information extractor for blip.tv"""
@@ -2983,8 +2880,7 @@ class StanfordOpenClassroomIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
-            self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
+            raise ExtractorError(u'Invalid URL: %s' % url)
-            return
 
         if mobj.group('course') and mobj.group('video'): # A specific video
             course = mobj.group('course')
@@ -3021,12 +2917,9 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 'upload_date': None,
             }
 
-            self.report_download_webpage(info['id'])
+            coursepage = self._download_webpage(url, info['id'],
-            try:
+                                        note='Downloading course info page',
-                coursepage = compat_urllib_request.urlopen(url).read()
+                                        errnote='Unable to download course info page')
-            except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                self._downloader.trouble(u'ERROR: unable to download course info page: ' + compat_str(err))
-                return
 
             m = re.search('<h1>([^<]+)</h1>', coursepage)
             if m:
@@ -3050,7 +2943,6 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 assert entry['type'] == 'reference'
                 results += self.extract(entry['url'])
             return results
-
         else: # Root page
             info = {
                 'id': 'Stanford OpenClassroom',
@@ -3927,8 +3819,6 @@ class PornotubeIE(InfoExtractor):
 
         return [info]
 
-
-
 class YouJizzIE(InfoExtractor):
     """Information extractor for youjizz.com."""
     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youjizz\.com/videos/(?P<videoid>[^.]+).html$'
@@ -3975,6 +3865,49 @@ class YouJizzIE(InfoExtractor):
 
         return [info]
 
+class EightTracksIE(InfoExtractor):
+    IE_NAME = '8tracks'
+    _VALID_URL = r'https?://8tracks.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        if mobj is None:
+            raise ExtractorError(u'Invalid URL: %s' % url)
+        playlist_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        m = re.search(r"new TRAX.Mix\((.*?)\);\n*\s*TRAX.initSearchAutocomplete\('#search'\);", webpage, flags=re.DOTALL)
+        if not m:
+            raise ExtractorError(u'Cannot find trax information')
+        json_like = m.group(1)
+        data = json.loads(json_like)
+
+        session = str(random.randint(0, 1000000000))
+        mix_id = data['id']
+        track_count = data['tracks_count']
+        first_url = 'http://8tracks.com/sets/%s/play?player=sm&mix_id=%s&format=jsonh' % (session, mix_id)
+        next_url = first_url
+        res = []
+        for i in itertools.count():
+            api_json = self._download_webpage(next_url, playlist_id,
+                note=u'Downloading song information %s/%s' % (str(i+1), track_count),
+                errnote=u'Failed to download song information')
+            api_data = json.loads(api_json)
+            track_data = api_data[u'set']['track']
+            info = {
+                'id': track_data['id'],
+                'url': track_data['track_file_stream_url'],
+                'title': track_data['performer'] + u' - ' + track_data['name'],
+                'raw_title': track_data['name'],
+                'uploader_id': data['user']['login'],
+                'ext': 'm4a',
+            }
+            res.append(info)
+            if api_data['set']['at_last_track']:
+                break
+            next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id'])
+        return res
 
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
@@ -4021,6 +3954,7 @@ def gen_extractors():
         SteamIE(),
         UstreamIE(),
         RBMARadioIE(),
+        EightTracksIE(),
         GenericIE()
     ]
 

youtube_dl/PostProcessor.py

@@ -143,10 +143,10 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
 
         more_opts = []
         if self._preferredcodec == 'best' or self._preferredcodec == filecodec or (self._preferredcodec == 'm4a' and filecodec == 'aac'):
-            if self._preferredcodec == 'm4a' and filecodec == 'aac':
+            if filecodec == 'aac' and self._preferredcodec in ['m4a', 'best']:
                 # Lossless, but in another container
                 acodec = 'copy'
-                extension = self._preferredcodec
+                extension = 'm4a'
                 more_opts = [self._exes['avconv'] and '-bsf:a' or '-absf', 'aac_adtstoasc']
             elif filecodec in ['aac', 'mp3', 'vorbis', 'opus']:
                 # Lossless if possible

youtube_dl/__init__.py

@@ -150,6 +150,9 @@ def parseOpts():
     selection.add_option('--match-title', dest='matchtitle', metavar='REGEX',help='download only matching titles (regex or caseless sub-string)')
     selection.add_option('--reject-title', dest='rejecttitle', metavar='REGEX',help='skip download for matching titles (regex or caseless sub-string)')
     selection.add_option('--max-downloads', metavar='NUMBER', dest='max_downloads', help='Abort after downloading NUMBER files', default=None)
+    selection.add_option('--min-filesize', metavar='SIZE', dest='min_filesize', help="Do not download any videos smaller than SIZE (e.g. 50k or 44.6m)", default=None)
+    selection.add_option('--max-filesize', metavar='SIZE', dest='max_filesize', help="Do not download any videos larger than SIZE (e.g. 50k or 44.6m)", default=None)
+
 
     authentication.add_option('-u', '--username',
             dest='username', metavar='USERNAME', help='account username')
@@ -290,10 +293,13 @@ def _real_main():
     else:
         try:
             jar = compat_cookiejar.MozillaCookieJar(opts.cookiefile)
-            if os.path.isfile(opts.cookiefile) and os.access(opts.cookiefile, os.R_OK):
+            if os.access(opts.cookiefile, os.R_OK):
                 jar.load()
         except (IOError, OSError) as err:
-            sys.exit(u'ERROR: unable to open cookie file')
+            if opts.verbose:
+                traceback.print_exc()
+            sys.stderr.write(u'ERROR: unable to open cookie file\n')
+            sys.exit(101)
     # Set user agent
     if opts.user_agent is not None:
         std_headers['User-Agent'] = opts.user_agent
@@ -353,6 +359,16 @@ def _real_main():
         if numeric_limit is None:
             parser.error(u'invalid rate limit specified')
         opts.ratelimit = numeric_limit
+    if opts.min_filesize is not None:
+        numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
+        if numeric_limit is None:
+            parser.error(u'invalid min_filesize specified')
+        opts.min_filesize = numeric_limit
+    if opts.max_filesize is not None:
+        numeric_limit = FileDownloader.parse_bytes(opts.max_filesize)
+        if numeric_limit is None:
+            parser.error(u'invalid max_filesize specified')
+        opts.max_filesize = numeric_limit
     if opts.retries is not None:
         try:
             opts.retries = int(opts.retries)
@@ -442,6 +458,8 @@ def _real_main():
         'verbose': opts.verbose,
         'test': opts.test,
         'keepvideo': opts.keepvideo,
+        'min_filesize': opts.min_filesize,
+        'max_filesize': opts.max_filesize
         })
 
     if opts.verbose:

youtube_dl/utils.py

@@ -280,6 +280,12 @@ class AttrParser(compat_html_parser.HTMLParser):
             lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]]
         lines[-1] = lines[-1][:self.result[2][1]]
         return '\n'.join(lines).strip()
+# Hack for https://github.com/rg3/youtube-dl/issues/662
+if sys.version_info < (2, 7, 3):
+    AttrParser.parse_endtag = (lambda self, i:
+        i + len("</scr'+'ipt>")
+        if self.rawdata[i:].startswith("</scr'+'ipt>")
+        else compat_html_parser.HTMLParser.parse_endtag(self, i))
 
 def get_element_by_id(id, html):
     """Return the content of the tag with the specified ID in the passed HTML document"""

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2013.01.13'
+__version__ = '2013.02.02'