Change the order for extracting/downloading

Now it gets a video info and directly downloads it, the it pass to the next video founded.
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-04-10 00:05:04 +02:00
parent d39919c03e
commit 146c12a2da

View file

@ -419,9 +419,10 @@ def _match_entry(self, info_dict):
return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
return None 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. Returns a list with a dictionary for each video we find.
If 'download', also downloads the videos.
''' '''
suitable_found = False suitable_found = False
for ie in self._ies: for ie in self._ies:
@ -440,7 +441,12 @@ def extract_info(self, url):
# Extract information from URL and process it # Extract information from URL and process it
try: try:
ie_results = ie.extract(url) 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 return results
except ExtractorError as de: # An error we somewhat expected except ExtractorError as de: # An error we somewhat expected
self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback()) self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
@ -453,51 +459,51 @@ def extract_info(self, url):
raise raise
if not suitable_found: if not suitable_found:
self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url) 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. Take the result of the ie and return a list of videos.
For url elements it will seartch the suitable ie and get the 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 For playlist elements it will process each of the elements of the 'entries' key
It will also download the videos if 'download'.
""" """
results = [] result_type = ie_result.get('_type', 'video') #If not given we suppose it's a video, support the dafault old system
for result in ie_results or []: if result_type == 'video':
result_type = result.get('_type', 'video') #If not given we suppose it's a video, support the dafault old system if 'playlist' not in ie_result:
if result_type == 'video': #It isn't part of a playlist
if not 'extractor' in result: ie_result['playlist'] = None
#The extractor has already been set somewhere else if download:
result['extractor'] = ie.IE_NAME #Do the download:
results.append(result) self.process_info(ie_result)
elif result_type == 'url': return ie_result
#We get the videos pointed by the url elif result_type == 'url':
results.extend(self.extract_info(result['url'])) #We get the video pointed by the url
elif result_type == 'playlist': result = self.extract_info(ie_result['url'], download)[0]
#We process each entry in the playlist return result
entries_result = self.process_ie_results(result['entries'], ie) elif result_type == 'playlist':
result['entries'] = entries_result #We process each entry in the playlist
results.extend([result]) playlist = ie_result.get('title', None) or ie_result.get('id', None)
return results 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): def process_info(self, info_dict):
"""Process a single dictionary returned by an InfoExtractor.""" """Process a single dictionary returned by an InfoExtractor."""
if info_dict.get('_type','video') == 'playlist': #We increment the download the download count here to match the previous behaviour.
playlist = info_dict.get('title', None) or info_dict.get('id', None) self.increment_downloads()
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
# Keep for backwards compatibility # Keep for backwards compatibility
info_dict['stitle'] = info_dict['title'] info_dict['stitle'] = info_dict['title']
@ -633,17 +639,14 @@ def download(self, url_list):
raise SameFileError(self.params['outtmpl']) raise SameFileError(self.params['outtmpl'])
for url in url_list: for url in url_list:
videos = self.extract_info(url) try:
#It also downloads the videos
for video in videos or []: videos = self.extract_info(url)
try: except UnavailableVideoError:
self.increment_downloads() self.trouble(u'\nERROR: unable to download video')
self.process_info(video) except MaxDownloadsReached:
except UnavailableVideoError: self.to_screen(u'[info] Maximum number of downloaded files reached.')
self.trouble(u'\nERROR: unable to download video') raise
except MaxDownloadsReached:
self.to_screen(u'[info] Maximum number of downloaded files reached.')
raise
return self._download_retcode return self._download_retcode