2 # -*- coding: utf-8 -*-
4 from __future__
import absolute_import
22 class FileDownloader(object):
23 """File Downloader class.
25 File downloader objects are the ones responsible of downloading the
26 actual video file and writing it to disk if the user has requested
27 it, among some other tasks. In most cases there should be one per
28 program. As, given a video URL, the downloader doesn't know how to
29 extract all the needed information, task that InfoExtractors do, it
30 has to pass the URL to one of them.
32 For this, file downloader objects have a method that allows
33 InfoExtractors to be registered in a given order. When it is passed
34 a URL, the file downloader handles it to the first InfoExtractor it
35 finds that reports being able to handle it. The InfoExtractor extracts
36 all the information about the video or videos the URL refers to, and
37 asks the FileDownloader to process the video information, possibly
38 downloading the video.
40 File downloaders accept a lot of parameters. In order not to saturate
41 the object constructor with arguments, it receives a dictionary of
42 options instead. These options are available through the params
43 attribute for the InfoExtractors to use. The FileDownloader also
44 registers itself as the downloader in charge for the InfoExtractors
45 that are added to it, so this is a "mutual registration".
49 username: Username for authentication purposes.
50 password: Password for authentication purposes.
51 usenetrc: Use netrc for authentication instead.
52 quiet: Do not print messages to stdout.
53 forceurl: Force printing final URL.
54 forcetitle: Force printing title.
55 forcethumbnail: Force printing thumbnail URL.
56 forcedescription: Force printing description.
57 forcefilename: Force printing final filename.
58 simulate: Do not download the video files.
59 format: Video format code.
60 format_limit: Highest quality format to try.
61 outtmpl: Template for output names.
62 restrictfilenames: Do not allow "&" and spaces in file names
63 ignoreerrors: Do not stop on download errors.
64 ratelimit: Download speed limit, in bytes/sec.
65 nooverwrites: Prevent overwriting files.
66 retries: Number of times to retry for HTTP error 5xx
67 buffersize: Size of download buffer in bytes.
68 noresizebuffer: Do not automatically resize the download buffer.
69 continuedl: Try to continue downloads if possible.
70 noprogress: Do not print the progress bar.
71 playliststart: Playlist item to start at.
72 playlistend: Playlist item to end at.
73 matchtitle: Download only matching titles.
74 rejecttitle: Reject downloads for matching titles.
75 logtostderr: Log messages to stderr instead of stdout.
76 consoletitle: Display progress in console window's titlebar.
77 nopart: Do not use temporary .part files.
78 updatetime: Use the Last-modified header to set output file timestamps.
79 writedescription: Write the video description to a .description file
80 writeinfojson: Write the video description to a .info.json file
81 writesubtitles: Write the video subtitles to a file
82 onlysubtitles: Downloads only the subtitles of the video
83 allsubtitles: Downloads all the subtitles of the video
84 listsubtitles: Lists all available subtitles for the video
85 subtitlesformat: Subtitle format [sbv/srt] (default=srt)
86 subtitleslang: Language of the subtitles to download
87 test: Download only first bytes to test the downloader.
88 keepvideo: Keep the video file after post-processing
89 min_filesize: Skip files smaller than this size
90 max_filesize: Skip files larger than this size
96 _download_retcode
= None
100 def __init__(self
, params
):
101 """Create a FileDownloader object with the given options."""
104 self
._progress
_hooks
= []
105 self
._download
_retcode
= 0
106 self
._num
_downloads
= 0
107 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
110 if '%(stitle)s' in self
.params
['outtmpl']:
111 self
.report_warning(u
'%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
114 def format_bytes(bytes):
117 if type(bytes) is str:
122 exponent
= int(math
.log(bytes, 1024.0))
123 suffix
= 'bkMGTPEZY'[exponent
]
124 converted
= float(bytes) / float(1024 ** exponent
)
125 return '%.2f%s' % (converted
, suffix
)
128 def calc_percent(byte_counter
, data_len
):
131 return '%6s' % ('%3.1f%%' % (float(byte_counter
) / float(data_len
) * 100.0))
134 def calc_eta(start
, now
, total
, current
):
138 if current
== 0 or dif
< 0.001: # One millisecond
140 rate
= float(current
) / dif
141 eta
= int((float(total
) - float(current
)) / rate
)
142 (eta_mins
, eta_secs
) = divmod(eta
, 60)
145 return '%02d:%02d' % (eta_mins
, eta_secs
)
148 def calc_speed(start
, now
, bytes):
150 if bytes == 0 or dif
< 0.001: # One millisecond
151 return '%10s' % '---b/s'
152 return '%10s' % ('%s/s' % FileDownloader
.format_bytes(float(bytes) / dif
))
155 def best_block_size(elapsed_time
, bytes):
156 new_min
= max(bytes / 2.0, 1.0)
157 new_max
= min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
158 if elapsed_time
< 0.001:
160 rate
= bytes / elapsed_time
168 def parse_bytes(bytestr
):
169 """Parse a string indicating a byte quantity into an integer."""
170 matchobj
= re
.match(r
'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr
)
173 number
= float(matchobj
.group(1))
174 multiplier
= 1024.0 ** 'bkmgtpezy'.index(matchobj
.group(2).lower())
175 return int(round(number
* multiplier
))
177 def add_info_extractor(self
, ie
):
178 """Add an InfoExtractor object to the end of the list."""
180 ie
.set_downloader(self
)
182 def add_post_processor(self
, pp
):
183 """Add a PostProcessor object to the end of the chain."""
185 pp
.set_downloader(self
)
187 def to_screen(self
, message
, skip_eol
=False):
188 """Print message to stdout if not in quiet mode."""
189 assert type(message
) == type(u
'')
190 if not self
.params
.get('quiet', False):
191 terminator
= [u
'\n', u
''][skip_eol
]
192 output
= message
+ terminator
193 if 'b' in getattr(self
._screen
_file
, 'mode', '') or sys
.version_info
[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
194 output
= output
.encode(preferredencoding(), 'ignore')
195 self
._screen
_file
.write(output
)
196 self
._screen
_file
.flush()
198 def to_stderr(self
, message
):
199 """Print message to stderr."""
200 assert type(message
) == type(u
'')
201 output
= message
+ u
'\n'
202 if 'b' in getattr(self
._screen
_file
, 'mode', '') or sys
.version_info
[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr
203 output
= output
.encode(preferredencoding())
204 sys
.stderr
.write(output
)
206 def to_cons_title(self
, message
):
207 """Set console/terminal window title to message."""
208 if not self
.params
.get('consoletitle', False):
210 if os
.name
== 'nt' and ctypes
.windll
.kernel32
.GetConsoleWindow():
211 # c_wchar_p() might not be necessary if `message` is
212 # already of type unicode()
213 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
214 elif 'TERM' in os
.environ
:
215 self
.to_screen('\033]0;%s\007' % message
, skip_eol
=True)
217 def fixed_template(self
):
218 """Checks if the output template is fixed."""
219 return (re
.search(u
'(?u)%\\(.+?\\)s', self
.params
['outtmpl']) is None)
221 def trouble(self
, message
=None, tb
=None):
222 """Determine action to take when a download problem appears.
224 Depending on if the downloader has been configured to ignore
225 download errors or not, this method may throw an exception or
226 not when errors are found, after printing the message.
228 tb, if given, is additional traceback information.
230 if message
is not None:
231 self
.to_stderr(message
)
232 if self
.params
.get('verbose'):
234 tb_data
= traceback
.format_list(traceback
.extract_stack())
235 tb
= u
''.join(tb_data
)
237 if not self
.params
.get('ignoreerrors', False):
238 raise DownloadError(message
)
239 self
._download
_retcode
= 1
241 def report_warning(self
, message
):
243 Print the message to stderr, it will be prefixed with 'WARNING:'
244 If stderr is a tty file the 'WARNING:' will be colored
246 if sys
.stderr
.isatty():
247 _msg_header
=u
'\033[0;33mWARNING:\033[0m'
249 _msg_header
=u
'WARNING:'
250 warning_message
=u
'%s %s' % (_msg_header
,message
)
251 self
.to_stderr(warning_message
)
253 def slow_down(self
, start_time
, byte_counter
):
254 """Sleep if the download speed is over the rate limit."""
255 rate_limit
= self
.params
.get('ratelimit', None)
256 if rate_limit
is None or byte_counter
== 0:
259 elapsed
= now
- start_time
262 speed
= float(byte_counter
) / elapsed
263 if speed
> rate_limit
:
264 time
.sleep((byte_counter
- rate_limit
* (now
- start_time
)) / rate_limit
)
266 def temp_name(self
, filename
):
267 """Returns a temporary filename for the given filename."""
268 if self
.params
.get('nopart', False) or filename
== u
'-' or \
269 (os
.path
.exists(encodeFilename(filename
)) and not os
.path
.isfile(encodeFilename(filename
))):
271 return filename
+ u
'.part'
273 def undo_temp_name(self
, filename
):
274 if filename
.endswith(u
'.part'):
275 return filename
[:-len(u
'.part')]
278 def try_rename(self
, old_filename
, new_filename
):
280 if old_filename
== new_filename
:
282 os
.rename(encodeFilename(old_filename
), encodeFilename(new_filename
))
283 except (IOError, OSError) as err
:
284 self
.trouble(u
'ERROR: unable to rename file')
286 def try_utime(self
, filename
, last_modified_hdr
):
287 """Try to set the last-modified time of the given file."""
288 if last_modified_hdr
is None:
290 if not os
.path
.isfile(encodeFilename(filename
)):
292 timestr
= last_modified_hdr
295 filetime
= timeconvert(timestr
)
299 os
.utime(filename
, (time
.time(), filetime
))
304 def report_writedescription(self
, descfn
):
305 """ Report that the description file is being written """
306 self
.to_screen(u
'[info] Writing video description to: ' + descfn
)
308 def report_writesubtitles(self
, sub_filename
):
309 """ Report that the subtitles file is being written """
310 self
.to_screen(u
'[info] Writing video subtitles to: ' + sub_filename
)
312 def report_writeinfojson(self
, infofn
):
313 """ Report that the metadata file has been written """
314 self
.to_screen(u
'[info] Video description metadata as JSON to: ' + infofn
)
316 def report_destination(self
, filename
):
317 """Report destination filename."""
318 self
.to_screen(u
'[download] Destination: ' + filename
)
320 def report_progress(self
, percent_str
, data_len_str
, speed_str
, eta_str
):
321 """Report download progress."""
322 if self
.params
.get('noprogress', False):
324 if self
.params
.get('progress_with_newline', False):
325 self
.to_screen(u
'[download] %s of %s at %s ETA %s' %
326 (percent_str
, data_len_str
, speed_str
, eta_str
))
328 self
.to_screen(u
'\r[download] %s of %s at %s ETA %s' %
329 (percent_str
, data_len_str
, speed_str
, eta_str
), skip_eol
=True)
330 self
.to_cons_title(u
'youtube-dl - %s of %s at %s ETA %s' %
331 (percent_str
.strip(), data_len_str
.strip(), speed_str
.strip(), eta_str
.strip()))
333 def report_resuming_byte(self
, resume_len
):
334 """Report attempt to resume at given byte."""
335 self
.to_screen(u
'[download] Resuming download at byte %s' % resume_len
)
337 def report_retry(self
, count
, retries
):
338 """Report retry in case of HTTP error 5xx"""
339 self
.to_screen(u
'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count
, retries
))
341 def report_file_already_downloaded(self
, file_name
):
342 """Report file has already been fully downloaded."""
344 self
.to_screen(u
'[download] %s has already been downloaded' % file_name
)
345 except (UnicodeEncodeError) as err
:
346 self
.to_screen(u
'[download] The file has already been downloaded')
348 def report_unable_to_resume(self
):
349 """Report it was impossible to resume download."""
350 self
.to_screen(u
'[download] Unable to resume')
352 def report_finish(self
):
353 """Report download finished."""
354 if self
.params
.get('noprogress', False):
355 self
.to_screen(u
'[download] Download completed')
359 def increment_downloads(self
):
360 """Increment the ordinal that assigns a number to each file."""
361 self
._num
_downloads
+= 1
363 def prepare_filename(self
, info_dict
):
364 """Generate the output filename."""
366 template_dict
= dict(info_dict
)
368 template_dict
['epoch'] = int(time
.time())
369 template_dict
['autonumber'] = u
'%05d' % self
._num
_downloads
371 sanitize
= lambda k
,v
: sanitize_filename(
372 u
'NA' if v
is None else compat_str(v
),
373 restricted
=self
.params
.get('restrictfilenames'),
375 template_dict
= dict((k
, sanitize(k
, v
)) for k
,v
in template_dict
.items())
377 filename
= self
.params
['outtmpl'] % template_dict
379 except KeyError as err
:
380 self
.trouble(u
'ERROR: Erroneous output template')
382 except ValueError as err
:
383 self
.trouble(u
'ERROR: Insufficient system charset ' + repr(preferredencoding()))
386 def _match_entry(self
, info_dict
):
387 """ Returns None iff the file should be downloaded """
389 title
= info_dict
['title']
390 matchtitle
= self
.params
.get('matchtitle', False)
392 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
393 return u
'[download] "' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
394 rejecttitle
= self
.params
.get('rejecttitle', False)
396 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
397 return u
'"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
400 def process_info(self
, info_dict
):
401 """Process a single dictionary returned by an InfoExtractor."""
403 # Keep for backwards compatibility
404 info_dict
['stitle'] = info_dict
['title']
406 if not 'format' in info_dict
:
407 info_dict
['format'] = info_dict
['ext']
409 reason
= self
._match
_entry
(info_dict
)
410 if reason
is not None:
411 self
.to_screen(u
'[download] ' + reason
)
414 max_downloads
= self
.params
.get('max_downloads')
415 if max_downloads
is not None:
416 if self
._num
_downloads
> int(max_downloads
):
417 raise MaxDownloadsReached()
419 filename
= self
.prepare_filename(info_dict
)
422 if self
.params
.get('forcetitle', False):
423 compat_print(info_dict
['title'])
424 if self
.params
.get('forceurl', False):
425 compat_print(info_dict
['url'])
426 if self
.params
.get('forcethumbnail', False) and 'thumbnail' in info_dict
:
427 compat_print(info_dict
['thumbnail'])
428 if self
.params
.get('forcedescription', False) and 'description' in info_dict
:
429 compat_print(info_dict
['description'])
430 if self
.params
.get('forcefilename', False) and filename
is not None:
431 compat_print(filename
)
432 if self
.params
.get('forceformat', False):
433 compat_print(info_dict
['format'])
435 # Do nothing else if in simulate mode
436 if self
.params
.get('simulate', False):
443 dn
= os
.path
.dirname(encodeFilename(filename
))
444 if dn
!= '' and not os
.path
.exists(dn
): # dn is already encoded
446 except (OSError, IOError) as err
:
447 self
.trouble(u
'ERROR: unable to create directory ' + compat_str(err
))
450 if self
.params
.get('writedescription', False):
452 descfn
= filename
+ u
'.description'
453 self
.report_writedescription(descfn
)
454 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
455 descfile
.write(info_dict
['description'])
456 except (OSError, IOError):
457 self
.trouble(u
'ERROR: Cannot write description file ' + descfn
)
460 if self
.params
.get('writesubtitles', False) and 'subtitles' in info_dict
and info_dict
['subtitles']:
461 # subtitles download errors are already managed as troubles in relevant IE
462 # that way it will silently go on when used with unsupporting IE
463 subtitle
= info_dict
['subtitles'][0]
464 (sub_error
, sub_lang
, sub
) = subtitle
465 sub_format
= self
.params
.get('subtitlesformat')
467 sub_filename
= filename
.rsplit('.', 1)[0] + u
'.' + sub_lang
+ u
'.' + sub_format
468 self
.report_writesubtitles(sub_filename
)
469 with io
.open(encodeFilename(sub_filename
), 'w', encoding
='utf-8') as subfile
:
471 except (OSError, IOError):
472 self
.trouble(u
'ERROR: Cannot write subtitles file ' + descfn
)
474 if self
.params
.get('onlysubtitles', False):
477 if self
.params
.get('allsubtitles', False) and 'subtitles' in info_dict
and info_dict
['subtitles']:
478 subtitles
= info_dict
['subtitles']
479 sub_format
= self
.params
.get('subtitlesformat')
480 for subtitle
in subtitles
:
481 (sub_error
, sub_lang
, sub
) = subtitle
483 sub_filename
= filename
.rsplit('.', 1)[0] + u
'.' + sub_lang
+ u
'.' + sub_format
484 self
.report_writesubtitles(sub_filename
)
485 with io
.open(encodeFilename(sub_filename
), 'w', encoding
='utf-8') as subfile
:
487 except (OSError, IOError):
488 self
.trouble(u
'ERROR: Cannot write subtitles file ' + descfn
)
490 if self
.params
.get('onlysubtitles', False):
493 if self
.params
.get('writeinfojson', False):
494 infofn
= filename
+ u
'.info.json'
495 self
.report_writeinfojson(infofn
)
497 json_info_dict
= dict((k
, v
) for k
,v
in info_dict
.items() if not k
in ['urlhandle'])
498 write_json_file(json_info_dict
, encodeFilename(infofn
))
499 except (OSError, IOError):
500 self
.trouble(u
'ERROR: Cannot write metadata to JSON file ' + infofn
)
503 if not self
.params
.get('skip_download', False):
504 if self
.params
.get('nooverwrites', False) and os
.path
.exists(encodeFilename(filename
)):
508 success
= self
._do
_download
(filename
, info_dict
)
509 except (OSError, IOError) as err
:
510 raise UnavailableVideoError()
511 except (compat_urllib_error
.URLError
, compat_http_client
.HTTPException
, socket
.error
) as err
:
512 self
.trouble(u
'ERROR: unable to download video data: %s' % str(err
))
514 except (ContentTooShortError
, ) as err
:
515 self
.trouble(u
'ERROR: content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
520 self
.post_process(filename
, info_dict
)
521 except (PostProcessingError
) as err
:
522 self
.trouble(u
'ERROR: postprocessing: %s' % str(err
))
525 def download(self
, url_list
):
526 """Download a given list of URLs."""
527 if len(url_list
) > 1 and self
.fixed_template():
528 raise SameFileError(self
.params
['outtmpl'])
531 suitable_found
= False
533 # Go to next InfoExtractor if not suitable
534 if not ie
.suitable(url
):
537 # Warn if the _WORKING attribute is False
539 self
.report_warning(u
'the program functionality for this site has been marked as broken, '
540 u
'and will probably not work. If you want to go on, use the -i option.')
542 # Suitable InfoExtractor found
543 suitable_found
= True
545 # Extract information from URL and process it
547 videos
= ie
.extract(url
)
548 except ExtractorError
as de
: # An error we somewhat expected
549 self
.trouble(u
'ERROR: ' + compat_str(de
), de
.format_traceback())
551 except MaxDownloadsReached
:
552 self
.to_screen(u
'[info] Maximum number of downloaded files reached.')
554 except Exception as e
:
555 if self
.params
.get('ignoreerrors', False):
556 self
.trouble(u
'ERROR: ' + compat_str(e
), tb
=compat_str(traceback
.format_exc()))
561 if len(videos
or []) > 1 and self
.fixed_template():
562 raise SameFileError(self
.params
['outtmpl'])
564 for video
in videos
or []:
565 video
['extractor'] = ie
.IE_NAME
567 self
.increment_downloads()
568 self
.process_info(video
)
569 except UnavailableVideoError
:
570 self
.trouble(u
'\nERROR: unable to download video')
572 # Suitable InfoExtractor had been found; go to next URL
575 if not suitable_found
:
576 self
.trouble(u
'ERROR: no suitable InfoExtractor: %s' % url
)
578 return self
._download
_retcode
580 def post_process(self
, filename
, ie_info
):
581 """Run all the postprocessors on the given file."""
583 info
['filepath'] = filename
587 keep_video_wish
,new_info
= pp
.run(info
)
588 if keep_video_wish
is not None:
590 keep_video
= keep_video_wish
591 elif keep_video
is None:
592 # No clear decision yet, let IE decide
593 keep_video
= keep_video_wish
594 except PostProcessingError
as e
:
595 self
.to_stderr(u
'ERROR: ' + e
.msg
)
596 if keep_video
is False and not self
.params
.get('keepvideo', False):
598 self
.to_screen(u
'Deleting original file %s (pass -k to keep)' % filename
)
599 os
.remove(encodeFilename(filename
))
600 except (IOError, OSError):
601 self
.report_warning(u
'Unable to remove downloaded video file')
603 def _download_with_rtmpdump(self
, filename
, url
, player_url
, page_url
):
604 self
.report_destination(filename
)
605 tmpfilename
= self
.temp_name(filename
)
607 # Check for rtmpdump first
609 subprocess
.call(['rtmpdump', '-h'], stdout
=(open(os
.path
.devnull
, 'w')), stderr
=subprocess
.STDOUT
)
610 except (OSError, IOError):
611 self
.trouble(u
'ERROR: RTMP download detected but "rtmpdump" could not be run')
614 # Download using rtmpdump. rtmpdump returns exit code 2 when
615 # the connection was interrumpted and resuming appears to be
616 # possible. This is part of rtmpdump's normal usage, AFAIK.
617 basic_args
= ['rtmpdump', '-q', '-r', url
, '-o', tmpfilename
]
618 if player_url
is not None:
619 basic_args
+= ['-W', player_url
]
620 if page_url
is not None:
621 basic_args
+= ['--pageUrl', page_url
]
622 args
= basic_args
+ [[], ['-e', '-k', '1']][self
.params
.get('continuedl', False)]
623 if self
.params
.get('verbose', False):
626 shell_quote
= lambda args
: ' '.join(map(pipes
.quote
, args
))
629 self
.to_screen(u
'[debug] rtmpdump command line: ' + shell_quote(args
))
630 retval
= subprocess
.call(args
)
631 while retval
== 2 or retval
== 1:
632 prevsize
= os
.path
.getsize(encodeFilename(tmpfilename
))
633 self
.to_screen(u
'\r[rtmpdump] %s bytes' % prevsize
, skip_eol
=True)
634 time
.sleep(5.0) # This seems to be needed
635 retval
= subprocess
.call(basic_args
+ ['-e'] + [[], ['-k', '1']][retval
== 1])
636 cursize
= os
.path
.getsize(encodeFilename(tmpfilename
))
637 if prevsize
== cursize
and retval
== 1:
639 # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
640 if prevsize
== cursize
and retval
== 2 and cursize
> 1024:
641 self
.to_screen(u
'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
645 fsize
= os
.path
.getsize(encodeFilename(tmpfilename
))
646 self
.to_screen(u
'\r[rtmpdump] %s bytes' % fsize
)
647 self
.try_rename(tmpfilename
, filename
)
648 self
._hook
_progress
({
649 'downloaded_bytes': fsize
,
650 'total_bytes': fsize
,
651 'filename': filename
,
652 'status': 'finished',
656 self
.trouble(u
'\nERROR: rtmpdump exited with code %d' % retval
)
659 def _do_download(self
, filename
, info_dict
):
660 url
= info_dict
['url']
662 # Check file already present
663 if self
.params
.get('continuedl', False) and os
.path
.isfile(encodeFilename(filename
)) and not self
.params
.get('nopart', False):
664 self
.report_file_already_downloaded(filename
)
665 self
._hook
_progress
({
666 'filename': filename
,
667 'status': 'finished',
671 # Attempt to download using rtmpdump
672 if url
.startswith('rtmp'):
673 return self
._download
_with
_rtmpdump
(filename
, url
,
674 info_dict
.get('player_url', None),
675 info_dict
.get('page_url', None))
677 tmpfilename
= self
.temp_name(filename
)
680 # Do not include the Accept-Encoding header
681 headers
= {'Youtubedl-no-compression': 'True'}
682 if 'user_agent' in info_dict
:
683 headers
['Youtubedl-user-agent'] = info_dict
['user_agent']
684 basic_request
= compat_urllib_request
.Request(url
, None, headers
)
685 request
= compat_urllib_request
.Request(url
, None, headers
)
687 if self
.params
.get('test', False):
688 request
.add_header('Range','bytes=0-10240')
690 # Establish possible resume length
691 if os
.path
.isfile(encodeFilename(tmpfilename
)):
692 resume_len
= os
.path
.getsize(encodeFilename(tmpfilename
))
698 if self
.params
.get('continuedl', False):
699 self
.report_resuming_byte(resume_len
)
700 request
.add_header('Range','bytes=%d-' % resume_len
)
706 retries
= self
.params
.get('retries', 0)
707 while count
<= retries
:
708 # Establish connection
710 if count
== 0 and 'urlhandle' in info_dict
:
711 data
= info_dict
['urlhandle']
712 data
= compat_urllib_request
.urlopen(request
)
714 except (compat_urllib_error
.HTTPError
, ) as err
:
715 if (err
.code
< 500 or err
.code
>= 600) and err
.code
!= 416:
716 # Unexpected HTTP error
718 elif err
.code
== 416:
719 # Unable to resume (requested range not satisfiable)
721 # Open the connection again without the range header
722 data
= compat_urllib_request
.urlopen(basic_request
)
723 content_length
= data
.info()['Content-Length']
724 except (compat_urllib_error
.HTTPError
, ) as err
:
725 if err
.code
< 500 or err
.code
>= 600:
728 # Examine the reported length
729 if (content_length
is not None and
730 (resume_len
- 100 < int(content_length
) < resume_len
+ 100)):
731 # The file had already been fully downloaded.
732 # Explanation to the above condition: in issue #175 it was revealed that
733 # YouTube sometimes adds or removes a few bytes from the end of the file,
734 # changing the file size slightly and causing problems for some users. So
735 # I decided to implement a suggested change and consider the file
736 # completely downloaded if the file size differs less than 100 bytes from
737 # the one in the hard drive.
738 self
.report_file_already_downloaded(filename
)
739 self
.try_rename(tmpfilename
, filename
)
740 self
._hook
_progress
({
741 'filename': filename
,
742 'status': 'finished',
746 # The length does not match, we start the download over
747 self
.report_unable_to_resume()
753 self
.report_retry(count
, retries
)
756 self
.trouble(u
'ERROR: giving up after %s retries' % retries
)
759 data_len
= data
.info().get('Content-length', None)
760 if data_len
is not None:
761 data_len
= int(data_len
) + resume_len
762 min_data_len
= self
.params
.get("min_filesize", None)
763 max_data_len
= self
.params
.get("max_filesize", None)
764 if min_data_len
is not None and data_len
< min_data_len
:
765 self
.to_screen(u
'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len
, min_data_len
))
767 if max_data_len
is not None and data_len
> max_data_len
:
768 self
.to_screen(u
'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len
, max_data_len
))
771 data_len_str
= self
.format_bytes(data_len
)
772 byte_counter
= 0 + resume_len
773 block_size
= self
.params
.get('buffersize', 1024)
778 data_block
= data
.read(block_size
)
780 if len(data_block
) == 0:
782 byte_counter
+= len(data_block
)
784 # Open file just in time
787 (stream
, tmpfilename
) = sanitize_open(tmpfilename
, open_mode
)
788 assert stream
is not None
789 filename
= self
.undo_temp_name(tmpfilename
)
790 self
.report_destination(filename
)
791 except (OSError, IOError) as err
:
792 self
.trouble(u
'ERROR: unable to open for writing: %s' % str(err
))
795 stream
.write(data_block
)
796 except (IOError, OSError) as err
:
797 self
.trouble(u
'\nERROR: unable to write data: %s' % str(err
))
799 if not self
.params
.get('noresizebuffer', False):
800 block_size
= self
.best_block_size(after
- before
, len(data_block
))
803 speed_str
= self
.calc_speed(start
, time
.time(), byte_counter
- resume_len
)
805 self
.report_progress('Unknown %', data_len_str
, speed_str
, 'Unknown ETA')
807 percent_str
= self
.calc_percent(byte_counter
, data_len
)
808 eta_str
= self
.calc_eta(start
, time
.time(), data_len
- resume_len
, byte_counter
- resume_len
)
809 self
.report_progress(percent_str
, data_len_str
, speed_str
, eta_str
)
811 self
._hook
_progress
({
812 'downloaded_bytes': byte_counter
,
813 'total_bytes': data_len
,
814 'tmpfilename': tmpfilename
,
815 'filename': filename
,
816 'status': 'downloading',
820 self
.slow_down(start
, byte_counter
- resume_len
)
823 self
.trouble(u
'\nERROR: Did not get any data blocks')
827 if data_len
is not None and byte_counter
!= data_len
:
828 raise ContentTooShortError(byte_counter
, int(data_len
))
829 self
.try_rename(tmpfilename
, filename
)
831 # Update file modification time
832 if self
.params
.get('updatetime', True):
833 info_dict
['filetime'] = self
.try_utime(filename
, data
.info().get('last-modified', None))
835 self
._hook
_progress
({
836 'downloaded_bytes': byte_counter
,
837 'total_bytes': byte_counter
,
838 'filename': filename
,
839 'status': 'finished',
844 def _hook_progress(self
, status
):
845 for ph
in self
._progress
_hooks
:
848 def add_progress_hook(self
, ph
):
849 """ ph gets called on download progress, with a dictionary with the entries
850 * filename: The final filename
851 * status: One of "downloading" and "finished"
853 It can also have some of the following entries:
855 * downloaded_bytes: Bytes on disks
856 * total_bytes: Total bytes, None if unknown
857 * tmpfilename: The filename we're currently writing to
859 Hooks are guaranteed to be called at least once (with status "finished")
860 if the download is successful.
862 self
._progress
_hooks
.append(ph
)