From df8db1aa2107f204fa14c157d7a536e45ceb65c4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Tue, 26 Feb 2013 23:33:58 +0100
Subject: [PATCH 01/14] Create extract_info method
---
youtube_dl/FileDownloader.py | 85 +++++++++++++++++++-----------------
1 file changed, 44 insertions(+), 41 deletions(-)
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index 192ad37d2..b26c34729 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -377,6 +377,44 @@ class FileDownloader(object):
if re.search(rejecttitle, title, re.IGNORECASE):
return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
return None
+
+ def extract_info(self, url):
+ '''
+ Returns a list with a dictionary for each video we find.
+ '''
+ suitable_found = False
+ for ie in self._ies:
+ # Go to next InfoExtractor if not suitable
+ if not ie.suitable(url):
+ continue
+
+ # Warn if the _WORKING attribute is False
+ if not ie.working():
+ self.to_stderr(u'WARNING: the program functionality for this site has been marked as broken, '
+ u'and will probably not work. If you want to go on, use the -i option.')
+
+ # Suitable InfoExtractor found
+ suitable_found = True
+
+ # Extract information from URL and process it
+ try:
+ videos = ie.extract(url)
+ for video in videos or []:
+ if not 'extractor' in video:
+ #The extractor has already been set somewher else
+ video['extractor'] = ie.IE_NAME
+ return videos
+ except ExtractorError as de: # An error we somewhat expected
+ self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
+ break
+ except Exception as e:
+ if self.params.get('ignoreerrors', False):
+ self.trouble(u'ERROR: ' + compat_str(e), tb=compat_str(traceback.format_exc()))
+ break
+ else:
+ raise
+ if not suitable_found:
+ self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
def process_info(self, info_dict):
"""Process a single dictionary returned by an InfoExtractor."""
@@ -488,49 +526,14 @@ class FileDownloader(object):
raise SameFileError(self.params['outtmpl'])
for url in url_list:
- suitable_found = False
- for ie in self._ies:
- # Go to next InfoExtractor if not suitable
- if not ie.suitable(url):
- continue
+ videos = self.extract_info(url)
- # Warn if the _WORKING attribute is False
- if not ie.working():
- self.to_stderr(u'WARNING: the program functionality for this site has been marked as broken, '
- u'and will probably not work. If you want to go on, use the -i option.')
-
- # Suitable InfoExtractor found
- suitable_found = True
-
- # Extract information from URL and process it
+ for video in videos or []:
try:
- videos = ie.extract(url)
- except ExtractorError as de: # An error we somewhat expected
- self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
- break
- except Exception as e:
- if self.params.get('ignoreerrors', False):
- self.trouble(u'ERROR: ' + compat_str(e), tb=compat_str(traceback.format_exc()))
- break
- else:
- raise
-
- if len(videos or []) > 1 and self.fixed_template():
- raise SameFileError(self.params['outtmpl'])
-
- for video in videos or []:
- video['extractor'] = ie.IE_NAME
- try:
- self.increment_downloads()
- self.process_info(video)
- except UnavailableVideoError:
- self.trouble(u'\nERROR: unable to download video')
-
- # Suitable InfoExtractor had been found; go to next URL
- break
-
- if not suitable_found:
- self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
+ self.increment_downloads()
+ self.process_info(video)
+ except UnavailableVideoError:
+ self.trouble(u'\nERROR: unable to download video')
return self._download_retcode
From 631f73978c0ee851950ac697dfd73f9092abd3c3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Mon, 4 Mar 2013 22:16:42 +0100
Subject: [PATCH 02/14] Add a method for extracting info from a list of urls
---
youtube_dl/FileDownloader.py | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index b26c34729..f668b362b 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -415,6 +415,14 @@ class FileDownloader(object):
raise
if not suitable_found:
self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
+ def extract_info_iterable(self, urls):
+ '''
+ Return the videos founded for the urls
+ '''
+ results = []
+ for url in urls:
+ results.extend(self.extract_info(url))
+ return results
def process_info(self, info_dict):
"""Process a single dictionary returned by an InfoExtractor."""
From 597cc8a45536aa4207c5ffc3e421fcebf2e08fe6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Tue, 5 Mar 2013 11:58:01 +0100
Subject: [PATCH 03/14] Use extract_info in YoutubePlaylist and YoutubeSearch
---
test/test_youtube_lists.py | 16 +++++++++-------
youtube_dl/InfoExtractors.py | 8 +++-----
2 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py
index f4705bc5b..055bf69c8 100644
--- a/test/test_youtube_lists.py
+++ b/test/test_youtube_lists.py
@@ -10,6 +10,7 @@ sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from youtube_dl.InfoExtractors import YoutubeUserIE, YoutubePlaylistIE, YoutubeIE
from youtube_dl.utils import *
+from youtube_dl.FileDownloader import FileDownloader
PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
@@ -22,7 +23,7 @@ proxy_handler = compat_urllib_request.ProxyHandler()
opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
compat_urllib_request.install_opener(opener)
-class FakeDownloader(object):
+class FakeDownloader(FileDownloader):
def __init__(self):
self.result = []
self.params = parameters
@@ -30,15 +31,16 @@ class FakeDownloader(object):
print(s)
def trouble(self, s):
raise Exception(s)
- def download(self, x):
- self.result.append(x)
+ def extract_info(self, url):
+ self.result.append(url)
+ return url
class TestYoutubeLists(unittest.TestCase):
def test_youtube_playlist(self):
dl = FakeDownloader()
ie = YoutubePlaylistIE(dl)
ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
- ytie_results = [YoutubeIE()._extract_id(r[0]) for r in dl.result]
+ ytie_results = [YoutubeIE()._extract_id(url) for url in dl.result]
self.assertEqual(ytie_results, [ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE'])
def test_issue_673(self):
@@ -58,7 +60,7 @@ class TestYoutubeLists(unittest.TestCase):
dl = FakeDownloader()
ie = YoutubePlaylistIE(dl)
ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
- ytie_results = [YoutubeIE()._extract_id(r[0]) for r in dl.result]
+ ytie_results = [YoutubeIE()._extract_id(url) for url in dl.result]
self.assertFalse('pElCt5oNDuI' in ytie_results)
self.assertFalse('KdPEApIVdWM' in ytie_results)
@@ -67,9 +69,9 @@ class TestYoutubeLists(unittest.TestCase):
ie = YoutubePlaylistIE(dl)
# TODO find a > 100 (paginating?) videos course
ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
- self.assertEqual(YoutubeIE()._extract_id(dl.result[0][0]), 'j9WZyLZCBzs')
+ self.assertEqual(YoutubeIE()._extract_id(dl.result[0]), 'j9WZyLZCBzs')
self.assertEqual(len(dl.result), 25)
- self.assertEqual(YoutubeIE()._extract_id(dl.result[-1][0]), 'rYefUsYuEp0')
+ self.assertEqual(YoutubeIE()._extract_id(dl.result[-1]), 'rYefUsYuEp0')
def test_youtube_channel(self):
# I give up, please find a channel that does paginate and test this like test_youtube_playlist_long
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 7ce84fe79..8a7694a76 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -1756,9 +1756,7 @@ class YoutubePlaylistIE(InfoExtractor):
else:
self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(videos)))
- for video in videos:
- self._downloader.download([video])
- return
+ return self._downloader.extract_info_iterable(videos)
class YoutubeChannelIE(InfoExtractor):
@@ -1892,8 +1890,8 @@ class YoutubeUserIE(InfoExtractor):
self._downloader.to_screen(u"[youtube] user %s: Collected %d video ids (downloading %d of them)" %
(username, all_ids_count, len(video_ids)))
- for video_id in video_ids:
- self._downloader.download(['http://www.youtube.com/watch?v=%s' % video_id])
+ urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
+ return self._downloader.extract_info_iterable(urls)
class BlipTVUserIE(InfoExtractor):
From f6e6da9525150487476d4990693eedf73acffab1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Tue, 5 Mar 2013 12:26:18 +0100
Subject: [PATCH 04/14] Use extract_info in BlipTV User and Youtube Channel
---
youtube_dl/InfoExtractors.py | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 8a7694a76..d79f6068f 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -1806,9 +1806,8 @@ class YoutubeChannelIE(InfoExtractor):
self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
- for id in video_ids:
- self._downloader.download(['http://www.youtube.com/watch?v=%s' % id])
- return
+ urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
+ return self._downloader.extract_info_iterable(urls)
class YoutubeUserIE(InfoExtractor):
@@ -1981,8 +1980,8 @@ class BlipTVUserIE(InfoExtractor):
self._downloader.to_screen(u"[%s] user %s: Collected %d video ids (downloading %d of them)" %
(self.IE_NAME, username, all_ids_count, len(video_ids)))
- for video_id in video_ids:
- self._downloader.download([u'http://blip.tv/'+video_id])
+ urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
+ return self._downloader.extract_info_iterable(urls)
class DepositFilesIE(InfoExtractor):
From 6ac7f082c469b3b2153735ae8475e1d0fc8b5439 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Tue, 5 Mar 2013 20:14:32 +0100
Subject: [PATCH 05/14] `extract_info` now expects `ie.extract` to return a
list in the format proposed in issue 608.
Each element should have a '_type' key specifying if it's a video, an url or a playlist.
`extract_info` will process each element to get the full info
---
youtube_dl/FileDownloader.py | 32 ++++++++++++++++++++++++++------
1 file changed, 26 insertions(+), 6 deletions(-)
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index 9b630c123..68fad11bc 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -410,12 +410,9 @@ class FileDownloader(object):
# Extract information from URL and process it
try:
- videos = ie.extract(url)
- for video in videos or []:
- if not 'extractor' in video:
- #The extractor has already been set somewher else
- video['extractor'] = ie.IE_NAME
- return videos
+ ie_results = ie.extract(url)
+ results = self.process_ie_results(ie_results, ie)
+ return results
except ExtractorError as de: # An error we somewhat expected
self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
break
@@ -435,6 +432,29 @@ class FileDownloader(object):
for url in urls:
results.extend(self.extract_info(url))
return results
+
+ def process_ie_results(self, ie_results, ie):
+ """
+ Take the results of the ie and return a list of videos.
+ For url elements it will seartch the suitable ie and get the videos
+ For playlist elements it will process each of the elements of the 'entries' key
+ """
+ results = []
+ for result in ie_results or []:
+ result_type = result.get('_type', 'video') #If not given we suppose it's a video, support the dafault old system
+ if result_type == 'video':
+ if not 'extractor' in result:
+ #The extractor has already been set somewhere else
+ result['extractor'] = ie.IE_NAME
+ results.append(result)
+ elif result_type == 'url':
+ #We get the videos pointed by the url
+ results.extend(self.extract_info(result['url']))
+ elif result_type == 'playlist':
+ #We process each entry in the playlist
+ entries_result = self.process_ie_results(result['entries'], ie)
+ results.extend(entries_result)
+ return results
def process_info(self, info_dict):
"""Process a single dictionary returned by an InfoExtractor."""
From 8a38a194fb08a253986cdbafa02cf699ef76c9a5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Tue, 5 Mar 2013 20:55:48 +0100
Subject: [PATCH 06/14] Add auxiliary methods to InfoExtractor to set the
'_type' key and use them for some playlist IEs
---
test/test_youtube_lists.py | 35 +++++++++++++++++++++--------------
youtube_dl/InfoExtractors.py | 30 ++++++++++++++++++++++++++----
2 files changed, 47 insertions(+), 18 deletions(-)
diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py
index 055bf69c8..9e91484f8 100644
--- a/test/test_youtube_lists.py
+++ b/test/test_youtube_lists.py
@@ -36,31 +36,37 @@ class FakeDownloader(FileDownloader):
return url
class TestYoutubeLists(unittest.TestCase):
+ def assertIsPlaylist(self,info):
+ """Make sure the info has '_type' set to 'playlist'"""
+ self.assertEqual(info['_type'], 'playlist')
+
def test_youtube_playlist(self):
dl = FakeDownloader()
ie = YoutubePlaylistIE(dl)
- ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
- ytie_results = [YoutubeIE()._extract_id(url) for url in dl.result]
+ result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')[0]
+ self.assertIsPlaylist(result)
+ ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
self.assertEqual(ytie_results, [ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE'])
def test_issue_673(self):
dl = FakeDownloader()
ie = YoutubePlaylistIE(dl)
- ie.extract('PLBB231211A4F62143')
- self.assertTrue(len(dl.result) > 40)
+ result = ie.extract('PLBB231211A4F62143')[0]
+ self.assertTrue(len(result['entries']) > 40)
def test_youtube_playlist_long(self):
dl = FakeDownloader()
ie = YoutubePlaylistIE(dl)
- ie.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
- self.assertTrue(len(dl.result) >= 799)
+ result = ie.extract('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')[0]
+ self.assertIsPlaylist(result)
+ self.assertTrue(len(result['entries']) >= 799)
def test_youtube_playlist_with_deleted(self):
#651
dl = FakeDownloader()
ie = YoutubePlaylistIE(dl)
- ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
- ytie_results = [YoutubeIE()._extract_id(url) for url in dl.result]
+ result = ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')[0]
+ ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
self.assertFalse('pElCt5oNDuI' in ytie_results)
self.assertFalse('KdPEApIVdWM' in ytie_results)
@@ -68,10 +74,11 @@ class TestYoutubeLists(unittest.TestCase):
dl = FakeDownloader()
ie = YoutubePlaylistIE(dl)
# TODO find a > 100 (paginating?) videos course
- ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
- self.assertEqual(YoutubeIE()._extract_id(dl.result[0]), 'j9WZyLZCBzs')
- self.assertEqual(len(dl.result), 25)
- self.assertEqual(YoutubeIE()._extract_id(dl.result[-1]), 'rYefUsYuEp0')
+ result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')[0]
+ entries = result['entries']
+ self.assertEqual(YoutubeIE()._extract_id(entries[0]['url']), 'j9WZyLZCBzs')
+ self.assertEqual(len(entries), 25)
+ self.assertEqual(YoutubeIE()._extract_id(entries[-1]['url']), 'rYefUsYuEp0')
def test_youtube_channel(self):
# I give up, please find a channel that does paginate and test this like test_youtube_playlist_long
@@ -80,8 +87,8 @@ class TestYoutubeLists(unittest.TestCase):
def test_youtube_user(self):
dl = FakeDownloader()
ie = YoutubeUserIE(dl)
- ie.extract('https://www.youtube.com/user/TheLinuxFoundation')
- self.assertTrue(len(dl.result) >= 320)
+ result = ie.extract('https://www.youtube.com/user/TheLinuxFoundation')[0]
+ self.assertTrue(len(result['entries']) >= 320)
if __name__ == '__main__':
unittest.main()
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index d79f6068f..895658f49 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -128,6 +128,24 @@ class InfoExtractor(object):
urlh = self._request_webpage(url_or_request, video_id, note, errnote)
webpage_bytes = urlh.read()
return webpage_bytes.decode('utf-8', 'replace')
+
+ #Methods for following #608
+ #They set the correct value of the '_type' key
+ def video_result(self, video_info):
+ """Returns a video"""
+ video_info['_type'] = 'video'
+ return video_info
+ def url_result(self, url, ie=None):
+ """Returns a url that points to a page that should be processed"""
+ #TODO: ie should be the class used for getting the info
+ video_info = {'_type': 'url',
+ 'url': url}
+ return video_info
+ def playlist_result(self, entries):
+ """Returns a playlist"""
+ video_info = {'_type': 'playlist',
+ 'entries': entries}
+ return video_info
class YoutubeIE(InfoExtractor):
@@ -1756,7 +1774,8 @@ class YoutubePlaylistIE(InfoExtractor):
else:
self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(videos)))
- return self._downloader.extract_info_iterable(videos)
+ url_results = [self.url_result(url) for url in videos]
+ return [self.playlist_result(url_results)]
class YoutubeChannelIE(InfoExtractor):
@@ -1807,7 +1826,8 @@ class YoutubeChannelIE(InfoExtractor):
self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
- return self._downloader.extract_info_iterable(urls)
+ url_entries = [self.url_result(url) for url in urls]
+ return [self.playlist_result(url_entries)]
class YoutubeUserIE(InfoExtractor):
@@ -1890,7 +1910,8 @@ class YoutubeUserIE(InfoExtractor):
(username, all_ids_count, len(video_ids)))
urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
- return self._downloader.extract_info_iterable(urls)
+ url_results = [self.url_result(url) for url in urls]
+ return [self.playlist_result(url_results)]
class BlipTVUserIE(InfoExtractor):
@@ -1981,7 +2002,8 @@ class BlipTVUserIE(InfoExtractor):
(self.IE_NAME, username, all_ids_count, len(video_ids)))
urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
- return self._downloader.extract_info_iterable(urls)
+ url_entries = [self.url_result(url) for url in urls]
+ return [self.playlist_result(url_entries)]
class DepositFilesIE(InfoExtractor):
From a0d6fe7b924697c089ed7ae37df0ca590ac38a96 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Tue, 5 Mar 2013 22:33:32 +0100
Subject: [PATCH 07/14] When a redirect is found return the new url using the
new style
---
youtube_dl/InfoExtractors.py | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 895658f49..e714fa6b0 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -1311,7 +1311,7 @@ class GenericIE(InfoExtractor):
self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url)
def _test_redirect(self, url):
- """Check if it is a redirect, like url shorteners, in case restart chain."""
+ """Check if it is a redirect, like url shorteners, in case return the new url."""
class HeadRequest(compat_urllib_request.Request):
def get_method(self):
return "HEAD"
@@ -1362,11 +1362,11 @@ class GenericIE(InfoExtractor):
return False
self.report_following_redirect(new_url)
- self._downloader.download([new_url])
- return True
+ return new_url
def _real_extract(self, url):
- if self._test_redirect(url): return
+ new_url = self._test_redirect(url)
+ if new_url: return [self.url_result(new_url)]
video_id = url.split('/')[-1]
request = compat_urllib_request.Request(url)
From d2c690828a8297c014d8053fbdee4e26fe11586a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Thu, 28 Mar 2013 13:39:00 +0100
Subject: [PATCH 08/14] Add title and id to playlist results
Not all IE give both. They are not used yet.
---
youtube_dl/InfoExtractors.py | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index dd4a776e4..6053d14ec 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -147,10 +147,14 @@ class InfoExtractor(object):
video_info = {'_type': 'url',
'url': url}
return video_info
- def playlist_result(self, entries):
+ def playlist_result(self, entries, playlist_id=None, playlist_title=None):
"""Returns a playlist"""
video_info = {'_type': 'playlist',
'entries': entries}
+ if playlist_id:
+ video_info['id'] = playlist_id
+ if playlist_title:
+ video_info['title'] = playlist_title
return video_info
@@ -1808,7 +1812,7 @@ class YoutubePlaylistIE(InfoExtractor):
self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(videos)))
url_results = [self.url_result(url) for url in videos]
- return [self.playlist_result(url_results)]
+ return [self.playlist_result(url_results, playlist_id)]
class YoutubeChannelIE(InfoExtractor):
@@ -1860,7 +1864,7 @@ class YoutubeChannelIE(InfoExtractor):
urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
url_entries = [self.url_result(url) for url in urls]
- return [self.playlist_result(url_entries)]
+ return [self.playlist_result(url_entries, channel_id)]
class YoutubeUserIE(InfoExtractor):
@@ -1944,7 +1948,7 @@ class YoutubeUserIE(InfoExtractor):
urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
url_results = [self.url_result(url) for url in urls]
- return [self.playlist_result(url_results)]
+ return [self.playlist_result(url_results, playlist_title = username)]
class BlipTVUserIE(InfoExtractor):
@@ -2036,7 +2040,7 @@ class BlipTVUserIE(InfoExtractor):
urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
url_entries = [self.url_result(url) for url in urls]
- return [self.playlist_result(url_entries)]
+ return [self.playlist_result(url_entries, playlist_title = username)]
class DepositFilesIE(InfoExtractor):
From 7eab8dc7504cf1f5f1dd03eb62e266ce24948b93 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaimemf93@gmail.com>
Date: Fri, 29 Mar 2013 12:32:42 +0100
Subject: [PATCH 09/14] Pass the playlist info_dict to process_info
the playlist value can be used in the output template
---
README.md | 1 +
youtube_dl/FileDownloader.py | 11 ++++++++++-
2 files changed, 11 insertions(+), 1 deletion(-)
diff --git a/README.md b/README.md
index 7c09d0c0d..1f3422ef8 100644
--- a/README.md
+++ b/README.md
@@ -138,6 +138,7 @@ The `-o` option allows users to indicate a template for the output file names. T
- `ext`: The sequence will be replaced by the appropriate extension (like flv or mp4).
- `epoch`: The sequence will be replaced by the Unix epoch when creating the file.
- `autonumber`: The sequence will be replaced by a five-digit number that will be increased with each download, starting at zero.
+ - `playlist`: The name or the id of the playlist that contains the video.
The current default template is `%(id)s.%(ext)s`, but that will be switchted to `%(title)s-%(id)s.%(ext)s` (which can be requested with `-t` at the moment).
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index 6af2acbee..d2b9be9ef 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -460,12 +460,21 @@ class FileDownloader(object):
elif result_type == 'playlist':
#We process each entry in the playlist
entries_result = self.process_ie_results(result['entries'], ie)
- results.extend(entries_result)
+ result['entries'] = entries_result
+ results.extend([result])
return results
def process_info(self, info_dict):
"""Process a single dictionary returned by an InfoExtractor."""
+ if info_dict.get('_type','video') == 'playlist':
+ playlist = info_dict.get('title', None) or info_dict.get('id', None)
+ self.to_screen(u'[download] Downloading playlist: %s' % playlist)
+ for video in info_dict['entries']:
+ video['playlist'] = playlist
+ self.process_info(video)
+ return
+
# Keep for backwards compatibility
info_dict['stitle'] = info_dict['title']
From d39919c03e45b3e8f804c23f78fae33cb4adc7df Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Fri, 5 Apr 2013 13:01:59 +0200
Subject: [PATCH 10/14] Add progress counter for playlists
Closes #276
---
youtube_dl/FileDownloader.py | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index 2237d355d..ba3277577 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -492,8 +492,10 @@ class FileDownloader(object):
if info_dict.get('_type','video') == 'playlist':
playlist = info_dict.get('title', None) or info_dict.get('id', None)
self.to_screen(u'[download] Downloading playlist: %s' % playlist)
- for video in info_dict['entries']:
+ n_videos = len(info_dict['entries'])
+ for i,video in enumerate(info_dict['entries'],1):
video['playlist'] = playlist
+ self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_videos))
self.process_info(video)
return
From 146c12a2dafdb9ff0e5138aa0f9da38bddca6c8a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 10 Apr 2013 00:05:04 +0200
Subject: [PATCH 11/14] Change the order for extracting/downloading
Now it gets a video info and directly downloads it, the it pass to the next video founded.
---
youtube_dl/FileDownloader.py | 103 ++++++++++++++++++-----------------
1 file changed, 53 insertions(+), 50 deletions(-)
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index ba3277577..58be5caee 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -419,9 +419,10 @@ class FileDownloader(object):
return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
return None
- def extract_info(self, url):
+ def extract_info(self, url, download = True):
'''
Returns a list with a dictionary for each video we find.
+ If 'download', also downloads the videos.
'''
suitable_found = False
for ie in self._ies:
@@ -440,7 +441,12 @@ class FileDownloader(object):
# Extract information from URL and process it
try:
ie_results = ie.extract(url)
- results = self.process_ie_results(ie_results, ie)
+ results = []
+ for ie_result in ie_results:
+ if not 'extractor' in ie_result:
+ #The extractor has already been set somewhere else
+ ie_result['extractor'] = ie.IE_NAME
+ results.append(self.process_ie_result(ie_result, download))
return results
except ExtractorError as de: # An error we somewhat expected
self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
@@ -453,51 +459,51 @@ class FileDownloader(object):
raise
if not suitable_found:
self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
- def extract_info_iterable(self, urls):
- '''
- Return the videos founded for the urls
- '''
- results = []
- for url in urls:
- results.extend(self.extract_info(url))
- return results
- def process_ie_results(self, ie_results, ie):
+ def process_ie_result(self, ie_result, download = True):
"""
- Take the results of the ie and return a list of videos.
- For url elements it will seartch the suitable ie and get the videos
+ Take the result of the ie and return a list of videos.
+ For url elements it will search the suitable ie and get the videos
For playlist elements it will process each of the elements of the 'entries' key
+
+ It will also download the videos if 'download'.
"""
- results = []
- for result in ie_results or []:
- result_type = result.get('_type', 'video') #If not given we suppose it's a video, support the dafault old system
- if result_type == 'video':
- if not 'extractor' in result:
- #The extractor has already been set somewhere else
- result['extractor'] = ie.IE_NAME
- results.append(result)
- elif result_type == 'url':
- #We get the videos pointed by the url
- results.extend(self.extract_info(result['url']))
- elif result_type == 'playlist':
- #We process each entry in the playlist
- entries_result = self.process_ie_results(result['entries'], ie)
- result['entries'] = entries_result
- results.extend([result])
- return results
+ result_type = ie_result.get('_type', 'video') #If not given we suppose it's a video, support the dafault old system
+ if result_type == 'video':
+ if 'playlist' not in ie_result:
+ #It isn't part of a playlist
+ ie_result['playlist'] = None
+ if download:
+ #Do the download:
+ self.process_info(ie_result)
+ return ie_result
+ elif result_type == 'url':
+ #We get the video pointed by the url
+ result = self.extract_info(ie_result['url'], download)[0]
+ return result
+ elif result_type == 'playlist':
+ #We process each entry in the playlist
+ playlist = ie_result.get('title', None) or ie_result.get('id', None)
+ self.to_screen(u'[download] Downloading playlist: %s' % playlist)
+ n_videos = len(ie_result['entries'])
+ playlist_results = []
+ for i,entry in enumerate(ie_result['entries'],1):
+ self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_videos))
+ entry_result = self.process_ie_result(entry, False)
+ entry_result['playlist'] = playlist
+ #We must do the download here to correctly set the 'playlist' key
+ if download:
+ self.process_info(entry_result)
+ playlist_results.append(entry_result)
+ result = ie_result.copy()
+ result['entries'] = playlist_results
+ return result
def process_info(self, info_dict):
"""Process a single dictionary returned by an InfoExtractor."""
- if info_dict.get('_type','video') == 'playlist':
- playlist = info_dict.get('title', None) or info_dict.get('id', None)
- self.to_screen(u'[download] Downloading playlist: %s' % playlist)
- n_videos = len(info_dict['entries'])
- for i,video in enumerate(info_dict['entries'],1):
- video['playlist'] = playlist
- self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_videos))
- self.process_info(video)
- return
+ #We increment the download the download count here to match the previous behaviour.
+ self.increment_downloads()
# Keep for backwards compatibility
info_dict['stitle'] = info_dict['title']
@@ -633,17 +639,14 @@ class FileDownloader(object):
raise SameFileError(self.params['outtmpl'])
for url in url_list:
- videos = self.extract_info(url)
-
- for video in videos or []:
- try:
- self.increment_downloads()
- self.process_info(video)
- except UnavailableVideoError:
- self.trouble(u'\nERROR: unable to download video')
- except MaxDownloadsReached:
- self.to_screen(u'[info] Maximum number of downloaded files reached.')
- raise
+ try:
+ #It also downloads the videos
+ videos = self.extract_info(url)
+ except UnavailableVideoError:
+ self.trouble(u'\nERROR: unable to download video')
+ except MaxDownloadsReached:
+ self.to_screen(u'[info] Maximum number of downloaded files reached.')
+ raise
return self._download_retcode
From 532d797824a1ec48480f1d10075e66a90aa53449 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 10 Apr 2013 00:06:03 +0200
Subject: [PATCH 12/14] In MetacafeIE return a url if YoutubeIE should do the
job
---
youtube_dl/InfoExtractors.py | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 81eaddc72..b7371365a 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -723,8 +723,7 @@ class MetacafeIE(InfoExtractor):
# Check if video comes from YouTube
mobj2 = re.match(r'^yt-(.*)$', video_id)
if mobj2 is not None:
- self._downloader.download(['http://www.youtube.com/watch?v=%s' % mobj2.group(1)])
- return
+ return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1))]
# Retrieve video webpage to extract further information
request = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)
From bce878a7c1678ac698ecd556b2c77a1e2f2306df Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 10 Apr 2013 14:32:03 +0200
Subject: [PATCH 13/14] Implement the playlist/start options in FileDownloader
It makes it available for all the InfoExtractors
---
youtube_dl/FileDownloader.py | 21 ++++++++++++++++++---
youtube_dl/InfoExtractors.py | 34 ----------------------------------
2 files changed, 18 insertions(+), 37 deletions(-)
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index 58be5caee..5a5141ba5 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -485,10 +485,25 @@ class FileDownloader(object):
#We process each entry in the playlist
playlist = ie_result.get('title', None) or ie_result.get('id', None)
self.to_screen(u'[download] Downloading playlist: %s' % playlist)
- n_videos = len(ie_result['entries'])
+
playlist_results = []
- for i,entry in enumerate(ie_result['entries'],1):
- self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_videos))
+
+ n_all_entries = len(ie_result['entries'])
+ playliststart = self.params.get('playliststart', 1) - 1
+ playlistend = self.params.get('playlistend', -1)
+
+ if playlistend == -1:
+ entries = ie_result['entries'][playliststart:]
+ else:
+ entries = ie_result['entries'][playliststart:playlistend]
+
+ n_entries = len(entries)
+
+ self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
+ (ie_result['extractor'], playlist, n_all_entries, n_entries))
+
+ for i,entry in enumerate(entries,1):
+ self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_entries))
entry_result = self.process_ie_result(entry, False)
entry_result['playlist'] = playlist
#We must do the download here to correctly set the 'playlist' key
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index b7371365a..a7fdf1607 100755
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -1806,19 +1806,6 @@ class YoutubePlaylistIE(InfoExtractor):
page_num += 1
videos = [v[1] for v in sorted(videos)]
- total = len(videos)
-
- playliststart = self._downloader.params.get('playliststart', 1) - 1
- playlistend = self._downloader.params.get('playlistend', -1)
- if playlistend == -1:
- videos = videos[playliststart:]
- else:
- videos = videos[playliststart:playlistend]
-
- if len(videos) == total:
- self._downloader.to_screen(u'[youtube] PL %s: Found %i videos' % (playlist_id, total))
- else:
- self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(videos)))
url_results = [self.url_result(url) for url in videos]
return [self.playlist_result(url_results, playlist_id)]
@@ -1943,18 +1930,6 @@ class YoutubeUserIE(InfoExtractor):
pagenum += 1
- all_ids_count = len(video_ids)
- playliststart = self._downloader.params.get('playliststart', 1) - 1
- playlistend = self._downloader.params.get('playlistend', -1)
-
- if playlistend == -1:
- video_ids = video_ids[playliststart:]
- else:
- video_ids = video_ids[playliststart:playlistend]
-
- self._downloader.to_screen(u"[youtube] user %s: Collected %d video ids (downloading %d of them)" %
- (username, all_ids_count, len(video_ids)))
-
urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
url_results = [self.url_result(url) for url in urls]
return [self.playlist_result(url_results, playlist_title = username)]
@@ -2035,15 +2010,6 @@ class BlipTVUserIE(InfoExtractor):
pagenum += 1
- all_ids_count = len(video_ids)
- playliststart = self._downloader.params.get('playliststart', 1) - 1
- playlistend = self._downloader.params.get('playlistend', -1)
-
- if playlistend == -1:
- video_ids = video_ids[playliststart:]
- else:
- video_ids = video_ids[playliststart:playlistend]
-
self._downloader.to_screen(u"[%s] user %s: Collected %d video ids (downloading %d of them)" %
(self.IE_NAME, username, all_ids_count, len(video_ids)))
From d281274bf250065f876bb4f75fb6f711e1a26eba Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Tue, 16 Apr 2013 15:13:29 +0200
Subject: [PATCH 14/14] Add a playlist_index key to the info_dict, can be used
in the output template
---
README.md | 1 +
youtube_dl/FileDownloader.py | 4 ++++
2 files changed, 5 insertions(+)
diff --git a/README.md b/README.md
index c8d28db3c..e2958a9b0 100644
--- a/README.md
+++ b/README.md
@@ -145,6 +145,7 @@ The `-o` option allows users to indicate a template for the output file names. T
- `epoch`: The sequence will be replaced by the Unix epoch when creating the file.
- `autonumber`: The sequence will be replaced by a five-digit number that will be increased with each download, starting at zero.
- `playlist`: The name or the id of the playlist that contains the video.
+ - `playlist_index`: The index of the video in the playlist, a five-digit number.
The current default template is `%(id)s.%(ext)s`, but that will be switchted to `%(title)s-%(id)s.%(ext)s` (which can be requested with `-t` at the moment).
diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py
index 5a5141ba5..4dabbb440 100644
--- a/youtube_dl/FileDownloader.py
+++ b/youtube_dl/FileDownloader.py
@@ -389,6 +389,8 @@ class FileDownloader(object):
template_dict['epoch'] = int(time.time())
template_dict['autonumber'] = u'%05d' % self._num_downloads
+ if template_dict['playlist_index'] is not None:
+ template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index']
sanitize = lambda k,v: sanitize_filename(
u'NA' if v is None else compat_str(v),
@@ -473,6 +475,7 @@ class FileDownloader(object):
if 'playlist' not in ie_result:
#It isn't part of a playlist
ie_result['playlist'] = None
+ ie_result['playlist_index'] = None
if download:
#Do the download:
self.process_info(ie_result)
@@ -506,6 +509,7 @@ class FileDownloader(object):
self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_entries))
entry_result = self.process_ie_result(entry, False)
entry_result['playlist'] = playlist
+ entry_result['playlist_index'] = i + playliststart
#We must do the download here to correctly set the 'playlist' key
if download:
self.process_info(entry_result)