From 79981f039ba14cd4b8e1be1c723ad9ee188fe2c4 Mon Sep 17 00:00:00 2001 From: codesparkle Date: Sat, 17 May 2014 14:22:24 +1000 Subject: [PATCH 1/2] Fixed test failure in test_all_urls: test_no_duplicates: BandcampAlbumIE inappropriately matched non-album bandcamp links as well. BandcampIE changed to report full-accuracy duration instead of unnecessarily rounding it to the nearest integer. Simplified conditionals and parsing a bit. Fixed typos. --- youtube_dl/extractor/bandcamp.py | 38 ++++++++++++++------------------ 1 file changed, 16 insertions(+), 22 deletions(-) diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 929aafdff..be4123d85 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -19,7 +19,7 @@ class BandcampIE(InfoExtractor): 'md5': 'c557841d5e50261777a6585648adf439', 'info_dict': { "title": "youtube-dl \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad", - "duration": 10, + "duration": 9.8485, }, '_skip': 'There is a limit of 200 free downloads / month for the test song' }] @@ -28,36 +28,32 @@ class BandcampIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) title = mobj.group('title') webpage = self._download_webpage(url, title) - # We get the link to the free download page m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage) - if m_download is None: + if not m_download: m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage) if m_trackinfo: json_code = m_trackinfo.group(1) - data = json.loads(json_code) - d = data[0] + data = json.loads(json_code)[0] - duration = int(round(d['duration'])) formats = [] - for format_id, format_url in d['file'].items(): - ext, _, abr_str = format_id.partition('-') - + for format_id, format_url in data['file'].items(): + ext, abr_str = format_id.split('-', maxsplit=1) formats.append({ 'format_id': format_id, 'url': format_url, - 'ext': format_id.partition('-')[0], + 'ext': ext, 'vcodec': 'none', - 'acodec': format_id.partition('-')[0], - 'abr': int(format_id.partition('-')[2]), + 'acodec': ext, + 'abr': int(abr_str), }) self._sort_formats(formats) return { - 'id': compat_str(d['id']), - 'title': d['title'], + 'id': compat_str(data['id']), + 'title': data['title'], 'formats': formats, - 'duration': duration, + 'duration': float(data['duration']), } else: raise ExtractorError('No free songs found') @@ -67,11 +63,9 @@ class BandcampIE(InfoExtractor): r'var TralbumData = {(.*?)id: (?P\d*?)$', webpage, re.MULTILINE | re.DOTALL).group('id') - download_webpage = self._download_webpage(download_link, video_id, - 'Downloading free downloads page') - # We get the dictionary of the track from some javascrip code - info = re.search(r'items: (.*?),$', - download_webpage, re.MULTILINE).group(1) + download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page') + # We get the dictionary of the track from some javascript code + info = re.search(r'items: (.*?),$', download_webpage, re.MULTILINE).group(1) info = json.loads(info)[0] # We pick mp3-320 for now, until format selection can be easily implemented. mp3_info = info['downloads']['mp3-320'] @@ -100,7 +94,7 @@ class BandcampIE(InfoExtractor): class BandcampAlbumIE(InfoExtractor): IE_NAME = 'Bandcamp:album' - _VALID_URL = r'https?://(?:(?P[^.]+)\.)?bandcamp\.com(?:/album/(?P[^?#]+))?' + _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<title>[^?#]+))' _TEST = { 'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1', @@ -123,7 +117,7 @@ class BandcampAlbumIE(InfoExtractor): 'params': { 'playlistend': 2 }, - 'skip': 'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test' + 'skip': 'Bandcamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test' } def _real_extract(self, url): From 2902d44f994ec8140595855f8d5b85601fae164f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= <dstftw@gmail.com> Date: Tue, 20 May 2014 19:44:42 +0700 Subject: [PATCH 2/2] [bandcamp] Replace maxsplit keyword argument with regular one Named arguments are not supported by methods implemented in native C (see http://bugs.python.org/issue1176) --- youtube_dl/extractor/bandcamp.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index be4123d85..dcbbdef43 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -37,7 +37,7 @@ class BandcampIE(InfoExtractor): formats = [] for format_id, format_url in data['file'].items(): - ext, abr_str = format_id.split('-', maxsplit=1) + ext, abr_str = format_id.split('-', 1) formats.append({ 'format_id': format_id, 'url': format_url,