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 .srt file
82 subtitleslang: Language of the subtitles to download
83 test: Download only first bytes to test the downloader.
84 keepvideo: Keep the video file after post-processing
85 min_filesize: Skip files smaller than this size
86 max_filesize: Skip files larger than this size
92 _download_retcode = None
96 def __init__(self, params):
97 """Create a FileDownloader object with the given options."""
100 self._progress_hooks = []
101 self._download_retcode = 0
102 self._num_downloads = 0
103 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
106 if '%(stitle)s' in self.params['outtmpl']:
107 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.')
110 def format_bytes(bytes):
113 if type(bytes) is str:
118 exponent = int(math.log(bytes, 1024.0))
119 suffix = 'bkMGTPEZY'[exponent]
120 converted = float(bytes) / float(1024 ** exponent)
121 return '%.2f%s' % (converted, suffix)
124 def calc_percent(byte_counter, data_len):
127 return '%6s' % ('%3.1f%%' % (float(byte_counter) / float(data_len) * 100.0))
130 def calc_eta(start, now, total, current):
134 if current == 0 or dif < 0.001: # One millisecond
136 rate = float(current) / dif
137 eta = int((float(total) - float(current)) / rate)
138 (eta_mins, eta_secs) = divmod(eta, 60)
141 return '%02d:%02d' % (eta_mins, eta_secs)
144 def calc_speed(start, now, bytes):
146 if bytes == 0 or dif < 0.001: # One millisecond
147 return '%10s' % '---b/s'
148 return '%10s' % ('%s/s' % FileDownloader.format_bytes(float(bytes) / dif))
151 def best_block_size(elapsed_time, bytes):
152 new_min = max(bytes / 2.0, 1.0)
153 new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
154 if elapsed_time < 0.001:
156 rate = bytes / elapsed_time
164 def parse_bytes(bytestr):
165 """Parse a string indicating a byte quantity into an integer."""
166 matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
169 number = float(matchobj.group(1))
170 multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
171 return int(round(number * multiplier))
173 def add_info_extractor(self, ie):
174 """Add an InfoExtractor object to the end of the list."""
176 ie.set_downloader(self)
178 def add_post_processor(self, pp):
179 """Add a PostProcessor object to the end of the chain."""
181 pp.set_downloader(self)
183 def to_screen(self, message, skip_eol=False):
184 """Print message to stdout if not in quiet mode."""
185 assert type(message) == type(u'')
186 if not self.params.get('quiet', False):
187 terminator = [u'\n', u''][skip_eol]
188 output = message + terminator
189 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
190 output = output.encode(preferredencoding(), 'ignore')
191 self._screen_file.write(output)
192 self._screen_file.flush()
194 def to_stderr(self, message):
195 """Print message to stderr."""
196 assert type(message) == type(u'')
197 output = message + u'\n'
198 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
199 output = output.encode(preferredencoding())
200 sys.stderr.write(output)
202 def to_cons_title(self, message):
203 """Set console/terminal window title to message."""
204 if not self.params.get('consoletitle', False):
206 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
207 # c_wchar_p() might not be necessary if `message` is
208 # already of type unicode()
209 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
210 elif 'TERM' in os.environ:
211 self.to_screen('\033]0;%s\007' % message, skip_eol=True)
213 def fixed_template(self):
214 """Checks if the output template is fixed."""
215 return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None)
217 def trouble(self, message=None, tb=None):
218 """Determine action to take when a download problem appears.
220 Depending on if the downloader has been configured to ignore
221 download errors or not, this method may throw an exception or
222 not when errors are found, after printing the message.
224 tb, if given, is additional traceback information.
226 if message is not None:
227 self.to_stderr(message)
228 if self.params.get('verbose'):
230 if sys.exc_info()[0]: # if .trouble has been called from an except block
232 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
233 tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
234 tb += compat_str(traceback.format_exc())
236 tb_data = traceback.format_list(traceback.extract_stack())
237 tb = u''.join(tb_data)
239 if not self.params.get('ignoreerrors', False):
240 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
241 exc_info = sys.exc_info()[1].exc_info
243 exc_info = sys.exc_info()
244 raise DownloadError(message, exc_info)
245 self._download_retcode = 1
247 def report_warning(self, message):
249 Print the message to stderr, it will be prefixed with 'WARNING:'
250 If stderr is a tty file the 'WARNING:' will be colored
252 if sys.stderr.isatty():
253 _msg_header=u'\033[0;33mWARNING:\033[0m'
255 _msg_header=u'WARNING:'
256 warning_message=u'%s %s' % (_msg_header,message)
257 self.to_stderr(warning_message)
259 def slow_down(self, start_time, byte_counter):
260 """Sleep if the download speed is over the rate limit."""
261 rate_limit = self.params.get('ratelimit', None)
262 if rate_limit is None or byte_counter == 0:
265 elapsed = now - start_time
268 speed = float(byte_counter) / elapsed
269 if speed > rate_limit:
270 time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
272 def temp_name(self, filename):
273 """Returns a temporary filename for the given filename."""
274 if self.params.get('nopart', False) or filename == u'-' or \
275 (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
277 return filename + u'.part'
279 def undo_temp_name(self, filename):
280 if filename.endswith(u'.part'):
281 return filename[:-len(u'.part')]
284 def try_rename(self, old_filename, new_filename):
286 if old_filename == new_filename:
288 os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
289 except (IOError, OSError) as err:
290 self.trouble(u'ERROR: unable to rename file')
292 def try_utime(self, filename, last_modified_hdr):
293 """Try to set the last-modified time of the given file."""
294 if last_modified_hdr is None:
296 if not os.path.isfile(encodeFilename(filename)):
298 timestr = last_modified_hdr
301 filetime = timeconvert(timestr)
305 os.utime(filename, (time.time(), filetime))
310 def report_writedescription(self, descfn):
311 """ Report that the description file is being written """
312 self.to_screen(u'[info] Writing video description to: ' + descfn)
314 def report_writesubtitles(self, srtfn):
315 """ Report that the subtitles file is being written """
316 self.to_screen(u'[info] Writing video subtitles to: ' + srtfn)
318 def report_writeinfojson(self, infofn):
319 """ Report that the metadata file has been written """
320 self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn)
322 def report_destination(self, filename):
323 """Report destination filename."""
324 self.to_screen(u'[download] Destination: ' + filename)
326 def report_progress(self, percent_str, data_len_str, speed_str, eta_str):
327 """Report download progress."""
328 if self.params.get('noprogress', False):
330 if self.params.get('progress_with_newline', False):
331 self.to_screen(u'[download] %s of %s at %s ETA %s' %
332 (percent_str, data_len_str, speed_str, eta_str))
334 self.to_screen(u'\r[download] %s of %s at %s ETA %s' %
335 (percent_str, data_len_str, speed_str, eta_str), skip_eol=True)
336 self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' %
337 (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip()))
339 def report_resuming_byte(self, resume_len):
340 """Report attempt to resume at given byte."""
341 self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
343 def report_retry(self, count, retries):
344 """Report retry in case of HTTP error 5xx"""
345 self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
347 def report_file_already_downloaded(self, file_name):
348 """Report file has already been fully downloaded."""
350 self.to_screen(u'[download] %s has already been downloaded' % file_name)
351 except (UnicodeEncodeError) as err:
352 self.to_screen(u'[download] The file has already been downloaded')
354 def report_unable_to_resume(self):
355 """Report it was impossible to resume download."""
356 self.to_screen(u'[download] Unable to resume')
358 def report_finish(self):
359 """Report download finished."""
360 if self.params.get('noprogress', False):
361 self.to_screen(u'[download] Download completed')
365 def increment_downloads(self):
366 """Increment the ordinal that assigns a number to each file."""
367 self._num_downloads += 1
369 def prepare_filename(self, info_dict):
370 """Generate the output filename."""
372 template_dict = dict(info_dict)
374 template_dict['epoch'] = int(time.time())
375 template_dict['autonumber'] = u'%05d' % self._num_downloads
377 sanitize = lambda k,v: sanitize_filename(
378 u'NA' if v is None else compat_str(v),
379 restricted=self.params.get('restrictfilenames'),
381 template_dict = dict((k, sanitize(k, v)) for k,v in template_dict.items())
383 filename = self.params['outtmpl'] % template_dict
385 except KeyError as err:
386 self.trouble(u'ERROR: Erroneous output template')
388 except ValueError as err:
389 self.trouble(u'ERROR: Insufficient system charset ' + repr(preferredencoding()))
392 def _match_entry(self, info_dict):
393 """ Returns None iff the file should be downloaded """
395 title = info_dict['title']
396 matchtitle = self.params.get('matchtitle', False)
398 if not re.search(matchtitle, title, re.IGNORECASE):
399 return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"'
400 rejecttitle = self.params.get('rejecttitle', False)
402 if re.search(rejecttitle, title, re.IGNORECASE):
403 return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
406 def process_info(self, info_dict):
407 """Process a single dictionary returned by an InfoExtractor."""
409 # Keep for backwards compatibility
410 info_dict['stitle'] = info_dict['title']
412 if not 'format' in info_dict:
413 info_dict['format'] = info_dict['ext']
415 reason = self._match_entry(info_dict)
416 if reason is not None:
417 self.to_screen(u'[download] ' + reason)
420 max_downloads = self.params.get('max_downloads')
421 if max_downloads is not None:
422 if self._num_downloads > int(max_downloads):
423 raise MaxDownloadsReached()
425 filename = self.prepare_filename(info_dict)
428 if self.params.get('forcetitle', False):
429 compat_print(info_dict['title'])
430 if self.params.get('forceurl', False):
431 compat_print(info_dict['url'])
432 if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
433 compat_print(info_dict['thumbnail'])
434 if self.params.get('forcedescription', False) and 'description' in info_dict:
435 compat_print(info_dict['description'])
436 if self.params.get('forcefilename', False) and filename is not None:
437 compat_print(filename)
438 if self.params.get('forceformat', False):
439 compat_print(info_dict['format'])
441 # Do nothing else if in simulate mode
442 if self.params.get('simulate', False):
449 dn = os.path.dirname(encodeFilename(filename))
450 if dn != '' and not os.path.exists(dn): # dn is already encoded
452 except (OSError, IOError) as err:
453 self.trouble(u'ERROR: unable to create directory ' + compat_str(err))
456 if self.params.get('writedescription', False):
458 descfn = filename + u'.description'
459 self.report_writedescription(descfn)
460 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
461 descfile.write(info_dict['description'])
462 except (OSError, IOError):
463 self.trouble(u'ERROR: Cannot write description file ' + descfn)
466 if self.params.get('writesubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']:
467 # subtitles download errors are already managed as troubles in relevant IE
468 # that way it will silently go on when used with unsupporting IE
470 srtfn = filename.rsplit('.', 1)[0] + u'.srt'
471 self.report_writesubtitles(srtfn)
472 with io.open(encodeFilename(srtfn), 'w', encoding='utf-8') as srtfile:
473 srtfile.write(info_dict['subtitles'])
474 except (OSError, IOError):
475 self.trouble(u'ERROR: Cannot write subtitles file ' + descfn)
478 if self.params.get('writeinfojson', False):
479 infofn = filename + u'.info.json'
480 self.report_writeinfojson(infofn)
482 json_info_dict = dict((k, v) for k,v in info_dict.items() if not k in ['urlhandle'])
483 write_json_file(json_info_dict, encodeFilename(infofn))
484 except (OSError, IOError):
485 self.trouble(u'ERROR: Cannot write metadata to JSON file ' + infofn)
488 if not self.params.get('skip_download', False):
489 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)):
493 success = self._do_download(filename, info_dict)
494 except (OSError, IOError) as err:
495 raise UnavailableVideoError()
496 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
497 self.trouble(u'ERROR: unable to download video data: %s' % str(err))
499 except (ContentTooShortError, ) as err:
500 self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
505 self.post_process(filename, info_dict)
506 except (PostProcessingError) as err:
507 self.trouble(u'ERROR: postprocessing: %s' % str(err))
510 def download(self, url_list):
511 """Download a given list of URLs."""
512 if len(url_list) > 1 and self.fixed_template():
513 raise SameFileError(self.params['outtmpl'])
516 suitable_found = False
518 # Go to next InfoExtractor if not suitable
519 if not ie.suitable(url):
522 # Warn if the _WORKING attribute is False
524 self.report_warning(u'the program functionality for this site has been marked as broken, '
525 u'and will probably not work. If you want to go on, use the -i option.')
527 # Suitable InfoExtractor found
528 suitable_found = True
530 # Extract information from URL and process it
532 videos = ie.extract(url)
533 except ExtractorError as de: # An error we somewhat expected
534 self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback())
536 except Exception as e:
537 if self.params.get('ignoreerrors', False):
538 self.trouble(u'ERROR: ' + compat_str(e), tb=compat_str(traceback.format_exc()))
543 if len(videos or []) > 1 and self.fixed_template():
544 raise SameFileError(self.params['outtmpl'])
546 for video in videos or []:
547 video['extractor'] = ie.IE_NAME
549 self.increment_downloads()
550 self.process_info(video)
551 except UnavailableVideoError:
552 self.trouble(u'\nERROR: unable to download video')
554 # Suitable InfoExtractor had been found; go to next URL
557 if not suitable_found:
558 self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
560 return self._download_retcode
562 def post_process(self, filename, ie_info):
563 """Run all the postprocessors on the given file."""
565 info['filepath'] = filename
569 keep_video_wish,new_info = pp.run(info)
570 if keep_video_wish is not None:
572 keep_video = keep_video_wish
573 elif keep_video is None:
574 # No clear decision yet, let IE decide
575 keep_video = keep_video_wish
576 except PostProcessingError as e:
577 self.to_stderr(u'ERROR: ' + e.msg)
578 if keep_video is False and not self.params.get('keepvideo', False):
580 self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
581 os.remove(encodeFilename(filename))
582 except (IOError, OSError):
583 self.report_warning(u'Unable to remove downloaded video file')
585 def _download_with_rtmpdump(self, filename, url, player_url, page_url):
586 self.report_destination(filename)
587 tmpfilename = self.temp_name(filename)
589 # Check for rtmpdump first
591 subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
592 except (OSError, IOError):
593 self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run')
596 # Download using rtmpdump. rtmpdump returns exit code 2 when
597 # the connection was interrumpted and resuming appears to be
598 # possible. This is part of rtmpdump's normal usage, AFAIK.
599 basic_args = ['rtmpdump', '-q', '-r', url, '-o', tmpfilename]
600 if player_url is not None:
601 basic_args += ['-W', player_url]
602 if page_url is not None:
603 basic_args += ['--pageUrl', page_url]
604 args = basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)]
605 if self.params.get('verbose', False):
608 shell_quote = lambda args: ' '.join(map(pipes.quote, args))
611 self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(args))
612 retval = subprocess.call(args)
613 while retval == 2 or retval == 1:
614 prevsize = os.path.getsize(encodeFilename(tmpfilename))
615 self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
616 time.sleep(5.0) # This seems to be needed
617 retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
618 cursize = os.path.getsize(encodeFilename(tmpfilename))
619 if prevsize == cursize and retval == 1:
621 # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
622 if prevsize == cursize and retval == 2 and cursize > 1024:
623 self.to_screen(u'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
627 fsize = os.path.getsize(encodeFilename(tmpfilename))
628 self.to_screen(u'\r[rtmpdump] %s bytes' % fsize)
629 self.try_rename(tmpfilename, filename)
630 self._hook_progress({
631 'downloaded_bytes': fsize,
632 'total_bytes': fsize,
633 'filename': filename,
634 'status': 'finished',
638 self.trouble(u'\nERROR: rtmpdump exited with code %d' % retval)
641 def _do_download(self, filename, info_dict):
642 url = info_dict['url']
644 # Check file already present
645 if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
646 self.report_file_already_downloaded(filename)
647 self._hook_progress({
648 'filename': filename,
649 'status': 'finished',
653 # Attempt to download using rtmpdump
654 if url.startswith('rtmp'):
655 return self._download_with_rtmpdump(filename, url,
656 info_dict.get('player_url', None),
657 info_dict.get('page_url', None))
659 tmpfilename = self.temp_name(filename)
662 # Do not include the Accept-Encoding header
663 headers = {'Youtubedl-no-compression': 'True'}
664 if 'user_agent' in info_dict:
665 headers['Youtubedl-user-agent'] = info_dict['user_agent']
666 basic_request = compat_urllib_request.Request(url, None, headers)
667 request = compat_urllib_request.Request(url, None, headers)
669 if self.params.get('test', False):
670 request.add_header('Range','bytes=0-10240')
672 # Establish possible resume length
673 if os.path.isfile(encodeFilename(tmpfilename)):
674 resume_len = os.path.getsize(encodeFilename(tmpfilename))
680 if self.params.get('continuedl', False):
681 self.report_resuming_byte(resume_len)
682 request.add_header('Range','bytes=%d-' % resume_len)
688 retries = self.params.get('retries', 0)
689 while count <= retries:
690 # Establish connection
692 if count == 0 and 'urlhandle' in info_dict:
693 data = info_dict['urlhandle']
694 data = compat_urllib_request.urlopen(request)
696 except (compat_urllib_error.HTTPError, ) as err:
697 if (err.code < 500 or err.code >= 600) and err.code != 416:
698 # Unexpected HTTP error
700 elif err.code == 416:
701 # Unable to resume (requested range not satisfiable)
703 # Open the connection again without the range header
704 data = compat_urllib_request.urlopen(basic_request)
705 content_length = data.info()['Content-Length']
706 except (compat_urllib_error.HTTPError, ) as err:
707 if err.code < 500 or err.code >= 600:
710 # Examine the reported length
711 if (content_length is not None and
712 (resume_len - 100 < int(content_length) < resume_len + 100)):
713 # The file had already been fully downloaded.
714 # Explanation to the above condition: in issue #175 it was revealed that
715 # YouTube sometimes adds or removes a few bytes from the end of the file,
716 # changing the file size slightly and causing problems for some users. So
717 # I decided to implement a suggested change and consider the file
718 # completely downloaded if the file size differs less than 100 bytes from
719 # the one in the hard drive.
720 self.report_file_already_downloaded(filename)
721 self.try_rename(tmpfilename, filename)
722 self._hook_progress({
723 'filename': filename,
724 'status': 'finished',
728 # The length does not match, we start the download over
729 self.report_unable_to_resume()
735 self.report_retry(count, retries)
738 self.trouble(u'ERROR: giving up after %s retries' % retries)
741 data_len = data.info().get('Content-length', None)
742 if data_len is not None:
743 data_len = int(data_len) + resume_len
744 min_data_len = self.params.get("min_filesize", None)
745 max_data_len = self.params.get("max_filesize", None)
746 if min_data_len is not None and data_len < min_data_len:
747 self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
749 if max_data_len is not None and data_len > max_data_len:
750 self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
753 data_len_str = self.format_bytes(data_len)
754 byte_counter = 0 + resume_len
755 block_size = self.params.get('buffersize', 1024)
760 data_block = data.read(block_size)
762 if len(data_block) == 0:
764 byte_counter += len(data_block)
766 # Open file just in time
769 (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
770 assert stream is not None
771 filename = self.undo_temp_name(tmpfilename)
772 self.report_destination(filename)
773 except (OSError, IOError) as err:
774 self.trouble(u'ERROR: unable to open for writing: %s' % str(err))
777 stream.write(data_block)
778 except (IOError, OSError) as err:
779 self.trouble(u'\nERROR: unable to write data: %s' % str(err))
781 if not self.params.get('noresizebuffer', False):
782 block_size = self.best_block_size(after - before, len(data_block))
785 speed_str = self.calc_speed(start, time.time(), byte_counter - resume_len)
787 self.report_progress('Unknown %', data_len_str, speed_str, 'Unknown ETA')
789 percent_str = self.calc_percent(byte_counter, data_len)
790 eta_str = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
791 self.report_progress(percent_str, data_len_str, speed_str, eta_str)
793 self._hook_progress({
794 'downloaded_bytes': byte_counter,
795 'total_bytes': data_len,
796 'tmpfilename': tmpfilename,
797 'filename': filename,
798 'status': 'downloading',
802 self.slow_down(start, byte_counter - resume_len)
805 self.trouble(u'\nERROR: Did not get any data blocks')
809 if data_len is not None and byte_counter != data_len:
810 raise ContentTooShortError(byte_counter, int(data_len))
811 self.try_rename(tmpfilename, filename)
813 # Update file modification time
814 if self.params.get('updatetime', True):
815 info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))
817 self._hook_progress({
818 'downloaded_bytes': byte_counter,
819 'total_bytes': byte_counter,
820 'filename': filename,
821 'status': 'finished',
826 def _hook_progress(self, status):
827 for ph in self._progress_hooks:
830 def add_progress_hook(self, ph):
831 """ ph gets called on download progress, with a dictionary with the entries
832 * filename: The final filename
833 * status: One of "downloading" and "finished"
835 It can also have some of the following entries:
837 * downloaded_bytes: Bytes on disks
838 * total_bytes: Total bytes, None if unknown
839 * tmpfilename: The filename we're currently writing to
841 Hooks are guaranteed to be called at least once (with status "finished")
842 if the download is successful.
844 self._progress_hooks.append(ph)