Leave out characters the filesystem cannot encode (Closes: #264)

This commit is contained in:
Philipp Hagemeister 2012-01-08 17:20:01 +01:00
parent dcb3c22e0b
commit fefb166c52
2 changed files with 147 additions and 132 deletions

View File

@ -0,0 +1,6 @@
# source this file in your shell to get a POSIX locale (which will break many programs, but that's kind of the point)
export LC_ALL=POSIX
export LANG=POSIX
export LANGUAGE=POSIX

View File

@ -259,14 +259,14 @@ def sanitize_open(filename, open_mode):
import msvcrt import msvcrt
msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
return (sys.stdout, filename) return (sys.stdout, filename)
stream = open(filename, open_mode) stream = open(_encodeFilename(filename), open_mode)
return (stream, filename) return (stream, filename)
except (IOError, OSError), err: except (IOError, OSError), err:
# In case of error, try to remove win32 forbidden chars # In case of error, try to remove win32 forbidden chars
filename = re.sub(ur'[/<>:"\|\?\*]', u'#', filename) filename = re.sub(ur'[/<>:"\|\?\*]', u'#', filename)
# An exception here should be caught in the caller # An exception here should be caught in the caller
stream = open(filename, open_mode) stream = open(_encodeFilename(filename), open_mode)
return (stream, filename) return (stream, filename)
@ -291,13 +291,21 @@ def _orderedSet(iterable):
return res return res
def _unescapeHTML(s): def _unescapeHTML(s):
""" """
@param s a string (of type unicode) @param s a string (of type unicode)
""" """
assert type(s) == type(u'') assert type(s) == type(u'')
htmlParser = HTMLParser.HTMLParser() htmlParser = HTMLParser.HTMLParser()
return htmlParser.unescape(s) return htmlParser.unescape(s)
def _encodeFilename(s):
"""
@param s The name of the file (of type unicode)
"""
assert type(s) == type(u'')
return s.encode(sys.getfilesystemencoding(), 'ignore')
class DownloadError(Exception): class DownloadError(Exception):
"""Download Error exception. """Download Error exception.
@ -563,16 +571,17 @@ class FileDownloader(object):
self._pps.append(pp) self._pps.append(pp)
pp.set_downloader(self) pp.set_downloader(self)
def to_screen(self, message, skip_eol=False, ignore_encoding_errors=False): def to_screen(self, message, skip_eol=False):
"""Print message to stdout if not in quiet mode.""" """Print message to stdout if not in quiet mode."""
try: assert type(message) == type(u'')
if not self.params.get('quiet', False): if not self.params.get('quiet', False):
terminator = [u'\n', u''][skip_eol] terminator = [u'\n', u''][skip_eol]
print >>self._screen_file, (u'%s%s' % (message, terminator)).encode(preferredencoding()), output = message + terminator
if 'b' not in self._screen_file.mode or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
output = output.encode(preferredencoding(), 'ignore')
self._screen_file.write(output)
self._screen_file.flush() self._screen_file.flush()
except (UnicodeEncodeError), err:
if not ignore_encoding_errors:
raise
def to_stderr(self, message): def to_stderr(self, message):
"""Print message to stderr.""" """Print message to stderr."""
@ -622,7 +631,7 @@ class FileDownloader(object):
def temp_name(self, filename): def temp_name(self, filename):
"""Returns a temporary filename for the given filename.""" """Returns a temporary filename for the given filename."""
if self.params.get('nopart', False) or filename == u'-' or \ if self.params.get('nopart', False) or filename == u'-' or \
(os.path.exists(filename) and not os.path.isfile(filename)): (os.path.exists(_encodeFilename(filename)) and not os.path.isfile(_encodeFilename(filename))):
return filename return filename
return filename + u'.part' return filename + u'.part'
@ -635,7 +644,7 @@ class FileDownloader(object):
try: try:
if old_filename == new_filename: if old_filename == new_filename:
return return
os.rename(old_filename, new_filename) os.rename(_encodeFilename(old_filename), _encodeFilename(new_filename))
except (IOError, OSError), err: except (IOError, OSError), err:
self.trouble(u'ERROR: unable to rename file') self.trouble(u'ERROR: unable to rename file')
@ -643,7 +652,7 @@ class FileDownloader(object):
"""Try to set the last-modified time of the given file.""" """Try to set the last-modified time of the given file."""
if last_modified_hdr is None: if last_modified_hdr is None:
return return
if not os.path.isfile(filename): if not os.path.isfile(_encodeFilename(filename)):
return return
timestr = last_modified_hdr timestr = last_modified_hdr
if timestr is None: if timestr is None:
@ -659,15 +668,15 @@ class FileDownloader(object):
def report_writedescription(self, descfn): def report_writedescription(self, descfn):
""" Report that the description file is being written """ """ Report that the description file is being written """
self.to_screen(u'[info] Writing video description to: %s' % descfn, ignore_encoding_errors=True) self.to_screen(u'[info] Writing video description to: ' + descfn)
def report_writeinfojson(self, infofn): def report_writeinfojson(self, infofn):
""" Report that the metadata file has been written """ """ Report that the metadata file has been written """
self.to_screen(u'[info] Video description metadata as JSON to: %s' % infofn, ignore_encoding_errors=True) self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
def report_destination(self, filename): def report_destination(self, filename):
"""Report destination filename.""" """Report destination filename."""
self.to_screen(u'[download] Destination: %s' % filename, ignore_encoding_errors=True) self.to_screen(u'[download] Destination: ' + filename)
def report_progress(self, percent_str, data_len_str, speed_str, eta_str): def report_progress(self, percent_str, data_len_str, speed_str, eta_str):
"""Report download progress.""" """Report download progress."""
@ -769,8 +778,8 @@ class FileDownloader(object):
return return
try: try:
dn = os.path.dirname(filename) dn = os.path.dirname(_encodeFilename(filename))
if dn != '' and not os.path.exists(dn): if dn != '' and not os.path.exists(dn): # dn is already encoded
os.makedirs(dn) os.makedirs(dn)
except (OSError, IOError), err: except (OSError, IOError), err:
self.trouble(u'ERROR: unable to create directory ' + unicode(err)) self.trouble(u'ERROR: unable to create directory ' + unicode(err))
@ -778,9 +787,9 @@ class FileDownloader(object):
if self.params.get('writedescription', False): if self.params.get('writedescription', False):
try: try:
descfn = filename + '.description' descfn = filename + u'.description'
self.report_writedescription(descfn) self.report_writedescription(descfn)
descfile = open(descfn, 'wb') descfile = open(_encodeFilename(descfn), 'wb')
try: try:
descfile.write(info_dict['description'].encode('utf-8')) descfile.write(info_dict['description'].encode('utf-8'))
finally: finally:
@ -790,7 +799,7 @@ class FileDownloader(object):
return return
if self.params.get('writeinfojson', False): if self.params.get('writeinfojson', False):
infofn = filename + '.info.json' infofn = filename + u'.info.json'
self.report_writeinfojson(infofn) self.report_writeinfojson(infofn)
try: try:
json.dump json.dump
@ -798,7 +807,7 @@ class FileDownloader(object):
self.trouble(u'ERROR: No JSON encoder found. Update to Python 2.6+, setup a json module, or leave out --write-info-json.') self.trouble(u'ERROR: No JSON encoder found. Update to Python 2.6+, setup a json module, or leave out --write-info-json.')
return return
try: try:
infof = open(infofn, 'wb') infof = open(_encodeFilename(infofn), 'wb')
try: try:
json_info_dict = dict((k,v) for k,v in info_dict.iteritems() if not k in ('urlhandle',)) json_info_dict = dict((k,v) for k,v in info_dict.iteritems() if not k in ('urlhandle',))
json.dump(json_info_dict, infof) json.dump(json_info_dict, infof)
@ -809,7 +818,7 @@ class FileDownloader(object):
return return
if not self.params.get('skip_download', False): if not self.params.get('skip_download', False):
if self.params.get('nooverwrites', False) and os.path.exists(filename): if self.params.get('nooverwrites', False) and os.path.exists(_encodeFilename(filename)):
success = True success = True
else: else:
try: try:
@ -882,11 +891,11 @@ class FileDownloader(object):
basic_args = ['rtmpdump', '-q'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename] basic_args = ['rtmpdump', '-q'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename]
retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)]) retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)])
while retval == 2 or retval == 1: while retval == 2 or retval == 1:
prevsize = os.path.getsize(tmpfilename) prevsize = os.path.getsize(_encodeFilename(tmpfilename))
self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True) self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
time.sleep(5.0) # This seems to be needed time.sleep(5.0) # This seems to be needed
retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1]) retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
cursize = os.path.getsize(tmpfilename) cursize = os.path.getsize(_encodeFilename(tmpfilename))
if prevsize == cursize and retval == 1: if prevsize == cursize and retval == 1:
break break
# Some rtmp streams seem abort after ~ 99.8%. Don't complain for those # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
@ -895,7 +904,7 @@ class FileDownloader(object):
retval = 0 retval = 0
break break
if retval == 0: if retval == 0:
self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(tmpfilename)) self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(_encodeFilename(tmpfilename)))
self.try_rename(tmpfilename, filename) self.try_rename(tmpfilename, filename)
return True return True
else: else:
@ -907,7 +916,7 @@ class FileDownloader(object):
player_url = info_dict.get('player_url', None) player_url = info_dict.get('player_url', None)
# Check file already present # Check file already present
if self.params.get('continuedl', False) and os.path.isfile(filename) and not self.params.get('nopart', False): if self.params.get('continuedl', False) and os.path.isfile(_encodeFilename(filename)) and not self.params.get('nopart', False):
self.report_file_already_downloaded(filename) self.report_file_already_downloaded(filename)
return True return True
@ -924,8 +933,8 @@ class FileDownloader(object):
request = urllib2.Request(url, None, headers) request = urllib2.Request(url, None, headers)
# Establish possible resume length # Establish possible resume length
if os.path.isfile(tmpfilename): if os.path.isfile(_encodeFilename(tmpfilename)):
resume_len = os.path.getsize(tmpfilename) resume_len = os.path.getsize(_encodeFilename(tmpfilename))
else: else:
resume_len = 0 resume_len = 0
@ -3750,17 +3759,17 @@ class MixcloudIE(InfoExtractor):
try: try:
# Process file information # Process file information
self._downloader.process_info({ self._downloader.process_info({
'id': file_id.decode('utf-8'), 'id': file_id.decode('utf-8'),
'url': file_url.decode('utf-8'), 'url': file_url.decode('utf-8'),
'uploader': uploader.decode('utf-8'), 'uploader': uploader.decode('utf-8'),
'upload_date': u'NA', 'upload_date': u'NA',
'title': json_data['name'], 'title': json_data['name'],
'stitle': _simplify_title(json_data['name']), 'stitle': _simplify_title(json_data['name']),
'ext': file_url.split('.')[-1].decode('utf-8'), 'ext': file_url.split('.')[-1].decode('utf-8'),
'format': (format_param is None and u'NA' or format_param.decode('utf-8')), 'format': (format_param is None and u'NA' or format_param.decode('utf-8')),
'thumbnail': json_data['thumbnail_url'], 'thumbnail': json_data['thumbnail_url'],
'description': json_data['description'], 'description': json_data['description'],
'player_url': player_url.decode('utf-8'), 'player_url': player_url.decode('utf-8'),
}) })
except UnavailableVideoError, err: except UnavailableVideoError, err:
self._downloader.trouble(u'ERROR: unable to download file') self._downloader.trouble(u'ERROR: unable to download file')
@ -3885,98 +3894,98 @@ class StanfordOpenClassroomIE(InfoExtractor):
self.extract(entry['url']) self.extract(entry['url'])
class MTVIE(InfoExtractor): class MTVIE(InfoExtractor):
"""Information extractor for MTV.com""" """Information extractor for MTV.com"""
_VALID_URL = r'^(?P<proto>https?://)?(?:www\.)?mtv\.com/videos/[^/]+/(?P<videoid>[0-9]+)/[^/]+$' _VALID_URL = r'^(?P<proto>https?://)?(?:www\.)?mtv\.com/videos/[^/]+/(?P<videoid>[0-9]+)/[^/]+$'
IE_NAME = u'mtv' IE_NAME = u'mtv'
def report_webpage(self, video_id): def report_webpage(self, video_id):
"""Report information extraction.""" """Report information extraction."""
self._downloader.to_screen(u'[%s] %s: Downloading webpage' % (self.IE_NAME, video_id)) self._downloader.to_screen(u'[%s] %s: Downloading webpage' % (self.IE_NAME, video_id))
def report_extraction(self, video_id): def report_extraction(self, video_id):
"""Report information extraction.""" """Report information extraction."""
self._downloader.to_screen(u'[%s] %s: Extracting information' % (self.IE_NAME, video_id)) self._downloader.to_screen(u'[%s] %s: Extracting information' % (self.IE_NAME, video_id))
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
if mobj is None: if mobj is None:
self._downloader.trouble(u'ERROR: invalid URL: %s' % url) self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
return return
if not mobj.group('proto'): if not mobj.group('proto'):
url = 'http://' + url url = 'http://' + url
video_id = mobj.group('videoid') video_id = mobj.group('videoid')
self.report_webpage(video_id) self.report_webpage(video_id)
request = urllib2.Request(url) request = urllib2.Request(url)
try: try:
webpage = urllib2.urlopen(request).read() webpage = urllib2.urlopen(request).read()
except (urllib2.URLError, httplib.HTTPException, socket.error), err: except (urllib2.URLError, httplib.HTTPException, socket.error), err:
self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err)) self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err))
return return
mobj = re.search(r'<meta name="mtv_vt" content="([^"]+)"/>', webpage) mobj = re.search(r'<meta name="mtv_vt" content="([^"]+)"/>', webpage)
if mobj is None: if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract song name') self._downloader.trouble(u'ERROR: unable to extract song name')
return return
song_name = _unescapeHTML(mobj.group(1).decode('iso-8859-1')) song_name = _unescapeHTML(mobj.group(1).decode('iso-8859-1'))
mobj = re.search(r'<meta name="mtv_an" content="([^"]+)"/>', webpage) mobj = re.search(r'<meta name="mtv_an" content="([^"]+)"/>', webpage)
if mobj is None: if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract performer') self._downloader.trouble(u'ERROR: unable to extract performer')
return return
performer = _unescapeHTML(mobj.group(1).decode('iso-8859-1')) performer = _unescapeHTML(mobj.group(1).decode('iso-8859-1'))
video_title = performer + ' - ' + song_name video_title = performer + ' - ' + song_name
mobj = re.search(r'<meta name="mtvn_uri" content="([^"]+)"/>', webpage) mobj = re.search(r'<meta name="mtvn_uri" content="([^"]+)"/>', webpage)
if mobj is None: if mobj is None:
self._downloader.trouble(u'ERROR: unable to mtvn_uri') self._downloader.trouble(u'ERROR: unable to mtvn_uri')
return return
mtvn_uri = mobj.group(1) mtvn_uri = mobj.group(1)
mobj = re.search(r'MTVN.Player.defaultPlaylistId = ([0-9]+);', webpage) mobj = re.search(r'MTVN.Player.defaultPlaylistId = ([0-9]+);', webpage)
if mobj is None: if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract content id') self._downloader.trouble(u'ERROR: unable to extract content id')
return return
content_id = mobj.group(1) content_id = mobj.group(1)
videogen_url = 'http://www.mtv.com/player/includes/mediaGen.jhtml?uri=' + mtvn_uri + '&id=' + content_id + '&vid=' + video_id + '&ref=www.mtvn.com&viewUri=' + mtvn_uri videogen_url = 'http://www.mtv.com/player/includes/mediaGen.jhtml?uri=' + mtvn_uri + '&id=' + content_id + '&vid=' + video_id + '&ref=www.mtvn.com&viewUri=' + mtvn_uri
self.report_extraction(video_id) self.report_extraction(video_id)
request = urllib2.Request(videogen_url) request = urllib2.Request(videogen_url)
try: try:
metadataXml = urllib2.urlopen(request).read() metadataXml = urllib2.urlopen(request).read()
except (urllib2.URLError, httplib.HTTPException, socket.error), err: except (urllib2.URLError, httplib.HTTPException, socket.error), err:
self._downloader.trouble(u'ERROR: unable to download video metadata: %s' % str(err)) self._downloader.trouble(u'ERROR: unable to download video metadata: %s' % str(err))
return return
mdoc = xml.etree.ElementTree.fromstring(metadataXml) mdoc = xml.etree.ElementTree.fromstring(metadataXml)
renditions = mdoc.findall('.//rendition') renditions = mdoc.findall('.//rendition')
# For now, always pick the highest quality. # For now, always pick the highest quality.
rendition = renditions[-1] rendition = renditions[-1]
try: try:
_,_,ext = rendition.attrib['type'].partition('/') _,_,ext = rendition.attrib['type'].partition('/')
format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate'] format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate']
video_url = rendition.find('./src').text video_url = rendition.find('./src').text
except KeyError: except KeyError:
self._downloader.trouble('Invalid rendition field.') self._downloader.trouble('Invalid rendition field.')
return return
self._downloader.increment_downloads() self._downloader.increment_downloads()
info = { info = {
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,
'uploader': performer, 'uploader': performer,
'title': video_title, 'title': video_title,
'stitle': _simplify_title(video_title), 'stitle': _simplify_title(video_title),
'ext': ext, 'ext': ext,
'format': format, 'format': format,
} }
try: try:
self._downloader.process_info(info) self._downloader.process_info(info)
except UnavailableVideoError, err: except UnavailableVideoError, err:
self._downloader.trouble(u'\nERROR: unable to download ' + video_id) self._downloader.trouble(u'\nERROR: unable to download ' + video_id)
class PostProcessor(object): class PostProcessor(object):
@ -4042,7 +4051,7 @@ class FFmpegExtractAudioPP(PostProcessor):
@staticmethod @staticmethod
def get_audio_codec(path): def get_audio_codec(path):
try: try:
cmd = ['ffprobe', '-show_streams', '--', path] cmd = ['ffprobe', '-show_streams', '--', _encodeFilename(path)]
handle = subprocess.Popen(cmd, stderr=file(os.path.devnull, 'w'), stdout=subprocess.PIPE) handle = subprocess.Popen(cmd, stderr=file(os.path.devnull, 'w'), stdout=subprocess.PIPE)
output = handle.communicate()[0] output = handle.communicate()[0]
if handle.wait() != 0: if handle.wait() != 0:
@ -4063,7 +4072,7 @@ class FFmpegExtractAudioPP(PostProcessor):
acodec_opts = [] acodec_opts = []
else: else:
acodec_opts = ['-acodec', codec] acodec_opts = ['-acodec', codec]
cmd = ['ffmpeg', '-y', '-i', path, '-vn'] + acodec_opts + more_opts + ['--', out_path] cmd = ['ffmpeg', '-y', '-i', _encodeFilename(path), '-vn'] + acodec_opts + more_opts + ['--', _encodeFilename(out_path)]
try: try:
p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE) p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
stdout,stderr = p.communicate() stdout,stderr = p.communicate()
@ -4124,9 +4133,9 @@ class FFmpegExtractAudioPP(PostProcessor):
extension = 'wav' extension = 'wav'
more_opts += ['-f', 'wav'] more_opts += ['-f', 'wav']
(prefix, ext) = os.path.splitext(path) prefix, sep, ext = path.rpartition(u'.') # not os.path.splitext, since the latter does not work on unicode in all setups
new_path = prefix + '.' + extension new_path = prefix + sep + extension
self._downloader.to_screen(u'[ffmpeg] Destination: %s' % new_path) self._downloader.to_screen(u'[ffmpeg] Destination: ' + new_path)
try: try:
self.run_ffmpeg(path, new_path, acodec, more_opts) self.run_ffmpeg(path, new_path, acodec, more_opts)
except: except:
@ -4140,13 +4149,13 @@ class FFmpegExtractAudioPP(PostProcessor):
# Try to update the date time for extracted audio file. # Try to update the date time for extracted audio file.
if information.get('filetime') is not None: if information.get('filetime') is not None:
try: try:
os.utime(new_path, (time.time(), information['filetime'])) os.utime(_encodeFilename(new_path), (time.time(), information['filetime']))
except: except:
self._downloader.to_stderr(u'WARNING: Cannot update utime of audio file') self._downloader.to_stderr(u'WARNING: Cannot update utime of audio file')
if not self._keepvideo: if not self._keepvideo:
try: try:
os.remove(path) os.remove(_encodeFilename(path))
except (IOError, OSError): except (IOError, OSError):
self._downloader.to_stderr(u'WARNING: Unable to remove downloaded video file') self._downloader.to_stderr(u'WARNING: Unable to remove downloaded video file')
return None return None
@ -4194,9 +4203,9 @@ def parseOpts():
import optparse import optparse
import shlex import shlex
def _readOptions(filename): def _readOptions(filename_bytes):
try: try:
optionf = open(filename) optionf = open(filename_bytes)
except IOError: except IOError:
return [] # silently skip if file is not present return [] # silently skip if file is not present
try: try: