aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/FileDownloader.py
blob: de1dc66bb4cdd8d43cfb3feecaf1d3e04f4b313f (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
import os
import re
import subprocess
import sys
import time

from .utils import (
    compat_urllib_error,
    compat_urllib_request,
    ContentTooShortError,
    determine_ext,
    encodeFilename,
    format_bytes,
    sanitize_open,
    timeconvert,
)


class FileDownloader(object):
    """File Downloader class.

    File downloader objects are the ones responsible of downloading the
    actual video file and writing it to disk.

    File downloaders accept a lot of parameters. In order not to saturate
    the object constructor with arguments, it receives a dictionary of
    options instead.

    Available options:

    verbose:           Print additional info to stdout.
    quiet:             Do not print messages to stdout.
    ratelimit:         Download speed limit, in bytes/sec.
    retries:           Number of times to retry for HTTP error 5xx
    buffersize:        Size of download buffer in bytes.
    noresizebuffer:    Do not automatically resize the download buffer.
    continuedl:        Try to continue downloads if possible.
    noprogress:        Do not print the progress bar.
    logtostderr:       Log messages to stderr instead of stdout.
    consoletitle:      Display progress in console window's titlebar.
    nopart:            Do not use temporary .part files.
    updatetime:        Use the Last-modified header to set output file timestamps.
    test:              Download only first bytes to test the downloader.
    min_filesize:      Skip files smaller than this size
    max_filesize:      Skip files larger than this size
    """

    params = None

    def __init__(self, ydl, params):
        """Create a FileDownloader object with the given options."""
        self.ydl = ydl
        self._progress_hooks = []
        self.params = params

    @staticmethod
    def format_seconds(seconds):
        (mins, secs) = divmod(seconds, 60)
        (hours, mins) = divmod(mins, 60)
        if hours > 99:
            return '--:--:--'
        if hours == 0:
            return '%02d:%02d' % (mins, secs)
        else:
            return '%02d:%02d:%02d' % (hours, mins, secs)

    @staticmethod
    def calc_percent(byte_counter, data_len):
        if data_len is None:
            return None
        return float(byte_counter) / float(data_len) * 100.0

    @staticmethod
    def format_percent(percent):
        if percent is None:
            return '---.-%'
        return '%6s' % ('%3.1f%%' % percent)

    @staticmethod
    def calc_eta(start, now, total, current):
        if total is None:
            return None
        dif = now - start
        if current == 0 or dif < 0.001: # One millisecond
            return None
        rate = float(current) / dif
        return int((float(total) - float(current)) / rate)

    @staticmethod
    def format_eta(eta):
        if eta is None:
            return '--:--'
        return FileDownloader.format_seconds(eta)

    @staticmethod
    def calc_speed(start, now, bytes):
        dif = now - start
        if bytes == 0 or dif < 0.001: # One millisecond
            return None
        return float(bytes) / dif

    @staticmethod
    def format_speed(speed):
        if speed is None:
            return '%10s' % '---b/s'
        return '%10s' % ('%s/s' % format_bytes(speed))

    @staticmethod
    def best_block_size(elapsed_time, bytes):
        new_min = max(bytes / 2.0, 1.0)
        new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
        if elapsed_time < 0.001:
            return int(new_max)
        rate = bytes / elapsed_time
        if rate > new_max:
            return int(new_max)
        if rate < new_min:
            return int(new_min)
        return int(rate)

    @staticmethod
    def parse_bytes(bytestr):
        """Parse a string indicating a byte quantity into an integer."""
        matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
        if matchobj is None:
            return None
        number = float(matchobj.group(1))
        multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
        return int(round(number * multiplier))

    def to_screen(self, *args, **kargs):
        self.ydl.to_screen(*args, **kargs)

    def to_stderr(self, message):
        self.ydl.to_screen(message)

    def to_console_title(self, message):
        self.ydl.to_console_title(message)

    def trouble(self, *args, **kargs):
        self.ydl.trouble(*args, **kargs)

    def report_warning(self, *args, **kargs):
        self.ydl.report_warning(*args, **kargs)

    def report_error(self, *args, **kargs):
        self.ydl.report_error(*args, **kargs)

    def slow_down(self, start_time, byte_counter):
        """Sleep if the download speed is over the rate limit."""
        rate_limit = self.params.get('ratelimit', None)
        if rate_limit is None or byte_counter == 0:
            return
        now = time.time()
        elapsed = now - start_time
        if elapsed <= 0.0:
            return
        speed = float(byte_counter) / elapsed
        if speed > rate_limit:
            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)

    def temp_name(self, filename):
        """Returns a temporary filename for the given filename."""
        if self.params.get('nopart', False) or filename == u'-' or \
                (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
            return filename
        return filename + u'.part'

    def undo_temp_name(self, filename):
        if filename.endswith(u'.part'):
            return filename[:-len(u'.part')]
        return filename

    def try_rename(self, old_filename, new_filename):
        try:
            if old_filename == new_filename:
                return
            os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
        except (IOError, OSError):
            self.report_error(u'unable to rename file')

    def try_utime(self, filename, last_modified_hdr):
        """Try to set the last-modified time of the given file."""
        if last_modified_hdr is None:
            return
        if not os.path.isfile(encodeFilename(filename)):
            return
        timestr = last_modified_hdr
        if timestr is None:
            return
        filetime = timeconvert(timestr)
        if filetime is None:
            return filetime
        # Ignore obviously invalid dates
        if filetime == 0:
            return
        try:
            os.utime(filename, (time.time(), filetime))
        except:
            pass
        return filetime

    def report_destination(self, filename):
        """Report destination filename."""
        self.to_screen(u'[download] Destination: ' + filename)

    def report_progress(self, percent, data_len_str, speed, eta):
        """Report download progress."""
        if self.params.get('noprogress', False):
            return
        clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
        if eta is not None:
            eta_str = self.format_eta(eta)
        else:
            eta_str = 'Unknown ETA'
        if percent is not None:
            percent_str = self.format_percent(percent)
        else:
            percent_str = 'Unknown %'
        speed_str = self.format_speed(speed)
        if self.params.get('progress_with_newline', False):
            self.to_screen(u'[download] %s of %s at %s ETA %s' %
                (percent_str, data_len_str, speed_str, eta_str))
        else:
            self.to_screen(u'\r%s[download] %s of %s at %s ETA %s' %
                (clear_line, percent_str, data_len_str, speed_str, eta_str), skip_eol=True)
        self.to_console_title(u'youtube-dl - %s of %s at %s ETA %s' %
                (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip()))
        
    def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
        if self.params.get('noprogress', False):
            return
        clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
        downloaded_str = format_bytes(downloaded_data_len)
        speed_str = self.format_speed(speed)
        elapsed_str = FileDownloader.format_seconds(elapsed)
        if self.params.get('progress_with_newline', False):
            self.to_screen(u'[download] %s at %s' %
                (downloaded_str, speed_str))
        else:
            self.to_screen(u'\r%s[download] %s at %s ET %s' %
                (clear_line, downloaded_str, speed_str, elapsed_str), skip_eol=True)
        self.to_console_title(u'youtube-dl - %s at %s ET %s' %
                (downloaded_str.strip(), speed_str.strip(), elapsed_str.strip())) 

    def report_resuming_byte(self, resume_len):
        """Report attempt to resume at given byte."""
        self.to_screen(u'[download] Resuming download at byte %s' % resume_len)

    def report_retry(self, count, retries):
        """Report retry in case of HTTP error 5xx"""
        self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))

    def report_file_already_downloaded(self, file_name):
        """Report file has already been fully downloaded."""
        try:
            self.to_screen(u'[download] %s has already been downloaded' % file_name)
        except UnicodeEncodeError:
            self.to_screen(u'[download] The file has already been downloaded')

    def report_unable_to_resume(self):
        """Report it was impossible to resume download."""
        self.to_screen(u'[download] Unable to resume')

    def report_finish(self, data_len_str, tot_time):
        """Report download finished."""
        if self.params.get('noprogress', False):
            self.to_screen(u'[download] Download completed')
        else:
            clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
            self.to_screen(u'\r%s[download] 100%% of %s in %s' %
                (clear_line, data_len_str, self.format_seconds(tot_time)))

    def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url, live, conn):
        def run_rtmpdump(args):
            start = time.time()
            resume_percent = None
            resume_downloaded_data_len = None
            proc = subprocess.Popen(args, stderr=subprocess.PIPE)
            cursor_in_new_line = True
            proc_stderr_closed = False
            while not proc_stderr_closed:
                # read line from stderr
                line = u''
                while True:
                    char = proc.stderr.read(1)
                    if not char:
                        proc_stderr_closed = True
                        break
                    if char in [b'\r', b'\n']:
                        break
                    line += char.decode('ascii', 'replace')
                if not line:
                    # proc_stderr_closed is True
                    continue
                mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line)
                if mobj:
                    downloaded_data_len = int(float(mobj.group(1))*1024)
                    percent = float(mobj.group(2))
                    if not resume_percent:
                        resume_percent = percent
                        resume_downloaded_data_len = downloaded_data_len
                    eta = self.calc_eta(start, time.time(), 100-resume_percent, percent-resume_percent)
                    speed = self.calc_speed(start, time.time(), downloaded_data_len-resume_downloaded_data_len)
                    data_len = None
                    if percent > 0:
                        data_len = int(downloaded_data_len * 100 / percent)
                    data_len_str = u'~' + format_bytes(data_len)
                    self.report_progress(percent, data_len_str, speed, eta)
                    cursor_in_new_line = False
                    self._hook_progress({
                        'downloaded_bytes': downloaded_data_len,
                        'total_bytes': data_len,
                        'tmpfilename': tmpfilename,
                        'filename': filename,
                        'status': 'downloading',
                        'eta': eta,
                        'speed': speed,
                    })
                else:
                    # no percent for live streams
                    mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line)
                    if mobj:
                        downloaded_data_len = int(float(mobj.group(1))*1024)
                        time_now = time.time()
                        speed = self.calc_speed(start, time_now, downloaded_data_len)
                        self.report_progress_live_stream(downloaded_data_len, speed, time_now - start)
                        cursor_in_new_line = False
                        self._hook_progress({
                            'downloaded_bytes': downloaded_data_len,
                            'tmpfilename': tmpfilename,
                            'filename': filename,
                            'status': 'downloading',
                            'speed': speed,
                        })
                    elif self.params.get('verbose', False):
                        if not cursor_in_new_line:
                            self.to_screen(u'')
                        cursor_in_new_line = True
                        self.to_screen(u'[rtmpdump] '+line)
            proc.wait()
            if not cursor_in_new_line:
                self.to_screen(u'')
            return proc.returncode

        self.report_destination(filename)
        tmpfilename = self.temp_name(filename)
        test = self.params.get('test', False)

        # Check for rtmpdump first
        try:
            subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
        except (OSError, IOError):
            self.report_error(u'RTMP download detected but "rtmpdump" could not be run')
            return False

        # Download using rtmpdump. rtmpdump returns exit code 2 when
        # the connection was interrumpted and resuming appears to be
        # possible. This is part of rtmpdump's normal usage, AFAIK.
        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
        if player_url is not None:
            basic_args += ['--swfVfy', player_url]
        if page_url is not None:
            basic_args += ['--pageUrl', page_url]
        if play_path is not None:
            basic_args += ['--playpath', play_path]
        if tc_url is not None:
            basic_args += ['--tcUrl', url]
        if test:
            basic_args += ['--stop', '1']
        if live:
            basic_args += ['--live']
        if conn:
            basic_args += ['--conn', conn]
        args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]

        if sys.platform == 'win32' and sys.version_info < (3, 0):
            # Windows subprocess module does not actually support Unicode
            # on Python 2.x
            # See http://stackoverflow.com/a/9951851/35070
            subprocess_encoding = sys.getfilesystemencoding()
            args = [a.encode(subprocess_encoding, 'ignore') for a in args]
        else:
            subprocess_encoding = None

        if self.params.get('verbose', False):
            if subprocess_encoding:
                str_args = [
                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
                    for a in args]
            else:
                str_args = args
            try:
                import pipes
                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
            except ImportError:
                shell_quote = repr
            self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(str_args))

        retval = run_rtmpdump(args)

        while (retval == 2 or retval == 1) and not test:
            prevsize = os.path.getsize(encodeFilename(tmpfilename))
            self.to_screen(u'[rtmpdump] %s bytes' % prevsize)
            time.sleep(5.0) # This seems to be needed
            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
            cursize = os.path.getsize(encodeFilename(tmpfilename))
            if prevsize == cursize and retval == 1:
                break
             # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
            if prevsize == cursize and retval == 2 and cursize > 1024:
                self.to_screen(u'[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
                retval = 0
                break
        if retval == 0 or (test and retval == 2):
            fsize = os.path.getsize(encodeFilename(tmpfilename))
            self.to_screen(u'[rtmpdump] %s bytes' % fsize)
            self.try_rename(tmpfilename, filename)
            self._hook_progress({
                'downloaded_bytes': fsize,
                'total_bytes': fsize,
                'filename': filename,
                'status': 'finished',
            })
            return True
        else:
            self.to_stderr(u"\n")
            self.report_error(u'rtmpdump exited with code %d' % retval)
            return False

    def _download_with_mplayer(self, filename, url):
        self.report_destination(filename)
        tmpfilename = self.temp_name(filename)

        args = ['mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy', '-dumpstream', '-dumpfile', tmpfilename, url]
        # Check for mplayer first
        try:
            subprocess.call(['mplayer', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
        except (OSError, IOError):
            self.report_error(u'MMS or RTSP download detected but "%s" could not be run' % args[0] )
            return False

        # Download using mplayer. 
        retval = subprocess.call(args)
        if retval == 0:
            fsize = os.path.getsize(encodeFilename(tmpfilename))
            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
            self.try_rename(tmpfilename, filename)
            self._hook_progress({
                'downloaded_bytes': fsize,
                'total_bytes': fsize,
                'filename': filename,
                'status': 'finished',
            })
            return True
        else:
            self.to_stderr(u"\n")
            self.report_error(u'mplayer exited with code %d' % retval)
            return False

    def _download_m3u8_with_ffmpeg(self, filename, url):
        self.report_destination(filename)
        tmpfilename = self.temp_name(filename)

        args = ['-y', '-i', url, '-f', 'mp4', '-c', 'copy',
            '-bsf:a', 'aac_adtstoasc', tmpfilename]

        for program in ['avconv', 'ffmpeg']:
            try:
                subprocess.call([program, '-version'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
                break
            except (OSError, IOError):
                pass
        else:
            self.report_error(u'm3u8 download detected but ffmpeg or avconv could not be found')
        cmd = [program] + args

        retval = subprocess.call(cmd)
        if retval == 0:
            fsize = os.path.getsize(encodeFilename(tmpfilename))
            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
            self.try_rename(tmpfilename, filename)
            self._hook_progress({
                'downloaded_bytes': fsize,
                'total_bytes': fsize,
                'filename': filename,
                'status': 'finished',
            })
            return True
        else:
            self.to_stderr(u"\n")
            self.report_error(u'ffmpeg exited with code %d' % retval)
            return False


    def _do_download(self, filename, info_dict):
        url = info_dict['url']

        # Check file already present
        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._hook_progress({
                'filename': filename,
                'status': 'finished',
                'total_bytes': os.path.getsize(encodeFilename(filename)),
            })
            return True

        # Attempt to download using rtmpdump
        if url.startswith('rtmp'):
            return self._download_with_rtmpdump(filename, url,
                                                info_dict.get('player_url', None),
                                                info_dict.get('page_url', None),
                                                info_dict.get('play_path', None),
                                                info_dict.get('tc_url', None),
                                                info_dict.get('rtmp_live', False),
                                                info_dict.get('rtmp_conn', None))

        # Attempt to download using mplayer
        if url.startswith('mms') or url.startswith('rtsp'):
            return self._download_with_mplayer(filename, url)

        # m3u8 manifest are downloaded with ffmpeg
        if determine_ext(url) == u'm3u8':
            return self._download_m3u8_with_ffmpeg(filename, url)

        tmpfilename = self.temp_name(filename)
        stream = None

        # Do not include the Accept-Encoding header
        headers = {'Youtubedl-no-compression': 'True'}
        if 'user_agent' in info_dict:
            headers['Youtubedl-user-agent'] = info_dict['user_agent']
        basic_request = compat_urllib_request.Request(url, None, headers)
        request = compat_urllib_request.Request(url, None, headers)

        if self.params.get('test', False):
            request.add_header('Range','bytes=0-10240')

        # Establish possible resume length
        if os.path.isfile(encodeFilename(tmpfilename)):
            resume_len = os.path.getsize(encodeFilename(tmpfilename))
        else:
            resume_len = 0

        open_mode = 'wb'
        if resume_len != 0:
            if self.params.get('continuedl', False):
                self.report_resuming_byte(resume_len)
                request.add_header('Range','bytes=%d-' % resume_len)
                open_mode = 'ab'
            else:
                resume_len = 0

        count = 0
        retries = self.params.get('retries', 0)
        while count <= retries:
            # Establish connection
            try:
                if count == 0 and 'urlhandle' in info_dict:
                    data = info_dict['urlhandle']
                data = compat_urllib_request.urlopen(request)
                break
            except (compat_urllib_error.HTTPError, ) as err:
                if (err.code < 500 or err.code >= 600) and err.code != 416:
                    # Unexpected HTTP error
                    raise
                elif err.code == 416:
                    # Unable to resume (requested range not satisfiable)
                    try:
                        # Open the connection again without the range header
                        data = compat_urllib_request.urlopen(basic_request)
                        content_length = data.info()['Content-Length']
                    except (compat_urllib_error.HTTPError, ) as err:
                        if err.code < 500 or err.code >= 600:
                            raise
                    else:
                        # Examine the reported length
                        if (content_length is not None and
                                (resume_len - 100 < int(content_length) < resume_len + 100)):
                            # The file had already been fully downloaded.
                            # Explanation to the above condition: in issue #175 it was revealed that
                            # YouTube sometimes adds or removes a few bytes from the end of the file,
                            # changing the file size slightly and causing problems for some users. So
                            # I decided to implement a suggested change and consider the file
                            # completely downloaded if the file size differs less than 100 bytes from
                            # the one in the hard drive.
                            self.report_file_already_downloaded(filename)
                            self.try_rename(tmpfilename, filename)
                            self._hook_progress({
                                'filename': filename,
                                'status': 'finished',
                            })
                            return True
                        else:
                            # The length does not match, we start the download over
                            self.report_unable_to_resume()
                            open_mode = 'wb'
                            break
            # Retry
            count += 1
            if count <= retries:
                self.report_retry(count, retries)

        if count > retries:
            self.report_error(u'giving up after %s retries' % retries)
            return False

        data_len = data.info().get('Content-length', None)
        if data_len is not None:
            data_len = int(data_len) + resume_len
            min_data_len = self.params.get("min_filesize", None)
            max_data_len =  self.params.get("max_filesize", None)
            if min_data_len is not None and data_len < min_data_len:
                self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
                return False
            if max_data_len is not None and data_len > max_data_len:
                self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
                return False

        data_len_str = format_bytes(data_len)
        byte_counter = 0 + resume_len
        block_size = self.params.get('buffersize', 1024)
        start = time.time()
        while True:
            # Download and write
            before = time.time()
            data_block = data.read(block_size)
            after = time.time()
            if len(data_block) == 0:
                break
            byte_counter += len(data_block)

            # Open file just in time
            if stream is None:
                try:
                    (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
                    assert stream is not None
                    filename = self.undo_temp_name(tmpfilename)
                    self.report_destination(filename)
                except (OSError, IOError) as err:
                    self.report_error(u'unable to open for writing: %s' % str(err))
                    return False
            try:
                stream.write(data_block)
            except (IOError, OSError) as err:
                self.to_stderr(u"\n")
                self.report_error(u'unable to write data: %s' % str(err))
                return False
            if not self.params.get('noresizebuffer', False):
                block_size = self.best_block_size(after - before, len(data_block))

            # Progress message
            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
            if data_len is None:
                eta = percent = None
            else:
                percent = self.calc_percent(byte_counter, data_len)
                eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
            self.report_progress(percent, data_len_str, speed, eta)

            self._hook_progress({
                'downloaded_bytes': byte_counter,
                'total_bytes': data_len,
                'tmpfilename': tmpfilename,
                'filename': filename,
                'status': 'downloading',
                'eta': eta,
                'speed': speed,
            })

            # Apply rate limit
            self.slow_down(start, byte_counter - resume_len)

        if stream is None:
            self.to_stderr(u"\n")
            self.report_error(u'Did not get any data blocks')
            return False
        stream.close()
        self.report_finish(data_len_str, (time.time() - start))
        if data_len is not None and byte_counter != data_len:
            raise ContentTooShortError(byte_counter, int(data_len))
        self.try_rename(tmpfilename, filename)

        # Update file modification time
        if self.params.get('updatetime', True):
            info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))

        self._hook_progress({
            'downloaded_bytes': byte_counter,
            'total_bytes': byte_counter,
            'filename': filename,
            'status': 'finished',
        })

        return True

    def _hook_progress(self, status):
        for ph in self._progress_hooks:
            ph(status)

    def add_progress_hook(self, ph):
        """ ph gets called on download progress, with a dictionary with the entries
        * filename: The final filename
        * status: One of "downloading" and "finished"

        It can also have some of the following entries:

        * downloaded_bytes: Bytes on disks
        * total_bytes: Total bytes, None if unknown
        * tmpfilename: The filename we're currently writing to
        * eta: The estimated time in seconds, None if unknown
        * speed: The download speed in bytes/second, None if unknown

        Hooks are guaranteed to be called at least once (with status "finished")
        if the download is successful.
        """
        self._progress_hooks.append(ph)