Compare commits
17 Commits
2013.12.23
...
2013.12.23
Author | SHA1 | Date | |
---|---|---|---|
![]() |
8958b6916c | ||
![]() |
9fc3bef87a | ||
![]() |
d80044c235 | ||
![]() |
bc2103f3bf | ||
![]() |
f82b18efc1 | ||
![]() |
504c668d3b | ||
![]() |
466617f539 | ||
![]() |
196938835a | ||
![]() |
a94e129a65 | ||
![]() |
5d681e960d | ||
![]() |
62d68c43ed | ||
![]() |
bfaae0a768 | ||
![]() |
e56f22ae20 | ||
![]() |
dbd1988ed9 | ||
![]() |
4ea3be0a5c | ||
![]() |
3bc2ddccc8 | ||
![]() |
8ab470f1b2 |
@@ -39,7 +39,8 @@ which means you can modify it, redistribute it or use it however you like.
|
|||||||
/youtube-dl .
|
/youtube-dl .
|
||||||
--no-cache-dir Disable filesystem caching
|
--no-cache-dir Disable filesystem caching
|
||||||
--bidi-workaround Work around terminals that lack bidirectional
|
--bidi-workaround Work around terminals that lack bidirectional
|
||||||
text support. Requires fribidi executable in PATH
|
text support. Requires bidiv or fribidi
|
||||||
|
executable in PATH
|
||||||
|
|
||||||
## Video Selection:
|
## Video Selection:
|
||||||
--playlist-start NUMBER playlist video to start at (default is 1)
|
--playlist-start NUMBER playlist video to start at (default is 1)
|
||||||
|
2
setup.py
2
setup.py
@@ -71,7 +71,7 @@ setup(
|
|||||||
author_email='ytdl@yt-dl.org',
|
author_email='ytdl@yt-dl.org',
|
||||||
maintainer='Philipp Hagemeister',
|
maintainer='Philipp Hagemeister',
|
||||||
maintainer_email='phihag@phihag.de',
|
maintainer_email='phihag@phihag.de',
|
||||||
packages=['youtube_dl', 'youtube_dl.extractor'],
|
packages=['youtube_dl', 'youtube_dl.extractor', 'youtube_dl.downloader'],
|
||||||
|
|
||||||
# Provokes warning on most systems (why?!)
|
# Provokes warning on most systems (why?!)
|
||||||
# test_suite = 'nose.collector',
|
# test_suite = 'nose.collector',
|
||||||
|
@@ -90,7 +90,7 @@ def generator(test_case):
|
|||||||
def _hook(status):
|
def _hook(status):
|
||||||
if status['status'] == 'finished':
|
if status['status'] == 'finished':
|
||||||
finished_hook_called.add(status['filename'])
|
finished_hook_called.add(status['filename'])
|
||||||
ydl.fd.add_progress_hook(_hook)
|
ydl.add_downloader_progress_hook(_hook)
|
||||||
|
|
||||||
def get_tc_filename(tc):
|
def get_tc_filename(tc):
|
||||||
return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
|
return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
|
||||||
|
@@ -1,724 +1,12 @@
|
|||||||
import os
|
# Legacy file for backwards compatibility, use youtube_dl.downloader instead!
|
||||||
import re
|
from .downloader import FileDownloader as RealFileDownloader
|
||||||
import subprocess
|
from .downloader import get_suitable_downloader
|
||||||
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_status(self, msg, is_last_line=False):
|
|
||||||
fullmsg = u'[download] ' + msg
|
|
||||||
if self.params.get('progress_with_newline', False):
|
|
||||||
self.to_screen(fullmsg)
|
|
||||||
else:
|
|
||||||
if os.name == 'nt':
|
|
||||||
prev_len = getattr(self, '_report_progress_prev_line_length',
|
|
||||||
0)
|
|
||||||
if prev_len > len(fullmsg):
|
|
||||||
fullmsg += u' ' * (prev_len - len(fullmsg))
|
|
||||||
self._report_progress_prev_line_length = len(fullmsg)
|
|
||||||
clear_line = u'\r'
|
|
||||||
else:
|
|
||||||
clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
|
|
||||||
self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
|
|
||||||
self.to_console_title(u'youtube-dl ' + msg)
|
|
||||||
|
|
||||||
def report_progress(self, percent, data_len_str, speed, eta):
|
|
||||||
"""Report download progress."""
|
|
||||||
if self.params.get('noprogress', False):
|
|
||||||
return
|
|
||||||
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)
|
|
||||||
|
|
||||||
msg = (u'%s of %s at %s ETA %s' %
|
|
||||||
(percent_str, data_len_str, speed_str, eta_str))
|
|
||||||
self._report_progress_status(msg)
|
|
||||||
|
|
||||||
def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
|
|
||||||
if self.params.get('noprogress', False):
|
|
||||||
return
|
|
||||||
downloaded_str = format_bytes(downloaded_data_len)
|
|
||||||
speed_str = self.format_speed(speed)
|
|
||||||
elapsed_str = FileDownloader.format_seconds(elapsed)
|
|
||||||
msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
|
|
||||||
self._report_progress_status(msg)
|
|
||||||
|
|
||||||
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:
|
|
||||||
self._report_progress_status(
|
|
||||||
(u'100%% of %s in %s' %
|
|
||||||
(data_len_str, self.format_seconds(tot_time))),
|
|
||||||
is_last_line=True)
|
|
||||||
|
|
||||||
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 _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
|
|
||||||
|
|
||||||
|
|
||||||
|
# This class reproduces the old behaviour of FileDownloader
|
||||||
|
class FileDownloader(RealFileDownloader):
|
||||||
def _do_download(self, filename, info_dict):
|
def _do_download(self, filename, info_dict):
|
||||||
url = info_dict['url']
|
real_fd = get_suitable_downloader(info_dict)(self.ydl, self.params)
|
||||||
|
|
||||||
# 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:
|
for ph in self._progress_hooks:
|
||||||
ph(status)
|
real_fd.add_progress_hook(ph)
|
||||||
|
return real_fd.download(filename, info_dict)
|
||||||
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)
|
|
||||||
|
@@ -53,7 +53,7 @@ from .utils import (
|
|||||||
YoutubeDLHandler,
|
YoutubeDLHandler,
|
||||||
)
|
)
|
||||||
from .extractor import get_info_extractor, gen_extractors
|
from .extractor import get_info_extractor, gen_extractors
|
||||||
from .FileDownloader import FileDownloader
|
from .downloader import get_suitable_downloader
|
||||||
from .version import __version__
|
from .version import __version__
|
||||||
|
|
||||||
|
|
||||||
@@ -167,7 +167,7 @@ class YoutubeDL(object):
|
|||||||
self._ies = []
|
self._ies = []
|
||||||
self._ies_instances = {}
|
self._ies_instances = {}
|
||||||
self._pps = []
|
self._pps = []
|
||||||
self._progress_hooks = []
|
self._fd_progress_hooks = []
|
||||||
self._download_retcode = 0
|
self._download_retcode = 0
|
||||||
self._num_downloads = 0
|
self._num_downloads = 0
|
||||||
self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
|
self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
|
||||||
@@ -183,12 +183,18 @@ class YoutubeDL(object):
|
|||||||
width_args = []
|
width_args = []
|
||||||
else:
|
else:
|
||||||
width_args = ['-w', str(width)]
|
width_args = ['-w', str(width)]
|
||||||
self._fribidi = subprocess.Popen(
|
sp_kwargs = dict(
|
||||||
['fribidi', '-c', 'UTF-8'] + width_args,
|
|
||||||
stdin=subprocess.PIPE,
|
stdin=subprocess.PIPE,
|
||||||
stdout=slave,
|
stdout=slave,
|
||||||
stderr=self._err_file)
|
stderr=self._err_file)
|
||||||
self._fribidi_channel = os.fdopen(master, 'rb')
|
try:
|
||||||
|
self._output_process = subprocess.Popen(
|
||||||
|
['bidiv'] + width_args, **sp_kwargs
|
||||||
|
)
|
||||||
|
except OSError:
|
||||||
|
self._output_process = subprocess.Popen(
|
||||||
|
['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
|
||||||
|
self._output_channel = os.fdopen(master, 'rb')
|
||||||
except OSError as ose:
|
except OSError as ose:
|
||||||
if ose.errno == 2:
|
if ose.errno == 2:
|
||||||
self.report_warning(u'Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
|
self.report_warning(u'Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
|
||||||
@@ -205,8 +211,6 @@ class YoutubeDL(object):
|
|||||||
u'Set the LC_ALL environment variable to fix this.')
|
u'Set the LC_ALL environment variable to fix this.')
|
||||||
self.params['restrictfilenames'] = True
|
self.params['restrictfilenames'] = True
|
||||||
|
|
||||||
self.fd = FileDownloader(self, self.params)
|
|
||||||
|
|
||||||
if '%(stitle)s' in self.params.get('outtmpl', ''):
|
if '%(stitle)s' in self.params.get('outtmpl', ''):
|
||||||
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.')
|
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.')
|
||||||
|
|
||||||
@@ -242,15 +246,20 @@ class YoutubeDL(object):
|
|||||||
self._pps.append(pp)
|
self._pps.append(pp)
|
||||||
pp.set_downloader(self)
|
pp.set_downloader(self)
|
||||||
|
|
||||||
|
def add_downloader_progress_hook(self, ph):
|
||||||
|
"""Add the progress hook to the file downloader"""
|
||||||
|
self._fd_progress_hooks.append(ph)
|
||||||
|
|
||||||
def _bidi_workaround(self, message):
|
def _bidi_workaround(self, message):
|
||||||
if not hasattr(self, '_fribidi_channel'):
|
if not hasattr(self, '_output_channel'):
|
||||||
return message
|
return message
|
||||||
|
|
||||||
|
assert hasattr(self, '_output_process')
|
||||||
assert type(message) == type(u'')
|
assert type(message) == type(u'')
|
||||||
line_count = message.count(u'\n') + 1
|
line_count = message.count(u'\n') + 1
|
||||||
self._fribidi.stdin.write((message + u'\n').encode('utf-8'))
|
self._output_process.stdin.write((message + u'\n').encode('utf-8'))
|
||||||
self._fribidi.stdin.flush()
|
self._output_process.stdin.flush()
|
||||||
res = u''.join(self._fribidi_channel.readline().decode('utf-8')
|
res = u''.join(self._output_channel.readline().decode('utf-8')
|
||||||
for _ in range(line_count))
|
for _ in range(line_count))
|
||||||
return res[:-len(u'\n')]
|
return res[:-len(u'\n')]
|
||||||
|
|
||||||
@@ -636,7 +645,7 @@ class YoutubeDL(object):
|
|||||||
info_dict['playlist_index'] = None
|
info_dict['playlist_index'] = None
|
||||||
|
|
||||||
# This extractors handle format selection themselves
|
# This extractors handle format selection themselves
|
||||||
if info_dict['extractor'] in [u'youtube', u'Youku']:
|
if info_dict['extractor'] in [u'Youku']:
|
||||||
if download:
|
if download:
|
||||||
self.process_info(info_dict)
|
self.process_info(info_dict)
|
||||||
return info_dict
|
return info_dict
|
||||||
@@ -662,10 +671,6 @@ class YoutubeDL(object):
|
|||||||
if 'ext' not in format:
|
if 'ext' not in format:
|
||||||
format['ext'] = determine_ext(format['url'])
|
format['ext'] = determine_ext(format['url'])
|
||||||
|
|
||||||
if self.params.get('listformats', None):
|
|
||||||
self.list_formats(info_dict)
|
|
||||||
return
|
|
||||||
|
|
||||||
format_limit = self.params.get('format_limit', None)
|
format_limit = self.params.get('format_limit', None)
|
||||||
if format_limit:
|
if format_limit:
|
||||||
formats = list(takewhile_inclusive(
|
formats = list(takewhile_inclusive(
|
||||||
@@ -678,9 +683,16 @@ class YoutubeDL(object):
|
|||||||
except ValueError:
|
except ValueError:
|
||||||
ext_ord = -1
|
ext_ord = -1
|
||||||
# We only compare the extension if they have the same height and width
|
# We only compare the extension if they have the same height and width
|
||||||
return (f.get('height'), f.get('width'), ext_ord)
|
return (f.get('height') if f.get('height') is not None else -1,
|
||||||
|
f.get('width') if f.get('width') is not None else -1,
|
||||||
|
ext_ord)
|
||||||
formats = sorted(formats, key=_free_formats_key)
|
formats = sorted(formats, key=_free_formats_key)
|
||||||
|
|
||||||
|
info_dict['formats'] = formats
|
||||||
|
if self.params.get('listformats', None):
|
||||||
|
self.list_formats(info_dict)
|
||||||
|
return
|
||||||
|
|
||||||
req_format = self.params.get('format', 'best')
|
req_format = self.params.get('format', 'best')
|
||||||
if req_format is None:
|
if req_format is None:
|
||||||
req_format = 'best'
|
req_format = 'best'
|
||||||
@@ -870,7 +882,10 @@ class YoutubeDL(object):
|
|||||||
success = True
|
success = True
|
||||||
else:
|
else:
|
||||||
try:
|
try:
|
||||||
success = self.fd._do_download(filename, info_dict)
|
fd = get_suitable_downloader(info_dict)(self, self.params)
|
||||||
|
for ph in self._fd_progress_hooks:
|
||||||
|
fd.add_progress_hook(ph)
|
||||||
|
success = fd.download(filename, info_dict)
|
||||||
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
|
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
|
||||||
self.report_error(u'unable to download video data: %s' % str(err))
|
self.report_error(u'unable to download video data: %s' % str(err))
|
||||||
return
|
return
|
||||||
|
@@ -194,7 +194,7 @@ def parseOpts(overrideArguments=None):
|
|||||||
type=float, default=None, help=optparse.SUPPRESS_HELP)
|
type=float, default=None, help=optparse.SUPPRESS_HELP)
|
||||||
general.add_option(
|
general.add_option(
|
||||||
'--bidi-workaround', dest='bidi_workaround', action='store_true',
|
'--bidi-workaround', dest='bidi_workaround', action='store_true',
|
||||||
help=u'Work around terminals that lack bidirectional text support. Requires fribidi executable in PATH')
|
help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
|
||||||
|
|
||||||
|
|
||||||
selection.add_option(
|
selection.add_option(
|
||||||
|
23
youtube_dl/downloader/__init__.py
Normal file
23
youtube_dl/downloader/__init__.py
Normal file
@@ -0,0 +1,23 @@
|
|||||||
|
from .common import FileDownloader
|
||||||
|
from .hls import HlsFD
|
||||||
|
from .http import HttpFD
|
||||||
|
from .mplayer import MplayerFD
|
||||||
|
from .rtmp import RtmpFD
|
||||||
|
|
||||||
|
from ..utils import (
|
||||||
|
determine_ext,
|
||||||
|
)
|
||||||
|
|
||||||
|
def get_suitable_downloader(info_dict):
|
||||||
|
"""Get the downloader class that can handle the info dict."""
|
||||||
|
url = info_dict['url']
|
||||||
|
|
||||||
|
if url.startswith('rtmp'):
|
||||||
|
return RtmpFD
|
||||||
|
if determine_ext(url) == u'm3u8':
|
||||||
|
return HlsFD
|
||||||
|
if url.startswith('mms') or url.startswith('rtsp'):
|
||||||
|
return MplayerFD
|
||||||
|
else:
|
||||||
|
return HttpFD
|
||||||
|
|
321
youtube_dl/downloader/common.py
Normal file
321
youtube_dl/downloader/common.py
Normal file
@@ -0,0 +1,321 @@
|
|||||||
|
import math
|
||||||
|
import os
|
||||||
|
import re
|
||||||
|
import subprocess
|
||||||
|
import sys
|
||||||
|
import time
|
||||||
|
|
||||||
|
from ..utils import (
|
||||||
|
encodeFilename,
|
||||||
|
timeconvert,
|
||||||
|
format_bytes,
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
|
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
|
||||||
|
|
||||||
|
Subclasses of this one must re-define the real_download method.
|
||||||
|
"""
|
||||||
|
|
||||||
|
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) as err:
|
||||||
|
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_status(self, msg, is_last_line=False):
|
||||||
|
fullmsg = u'[download] ' + msg
|
||||||
|
if self.params.get('progress_with_newline', False):
|
||||||
|
self.to_screen(fullmsg)
|
||||||
|
else:
|
||||||
|
if os.name == 'nt':
|
||||||
|
prev_len = getattr(self, '_report_progress_prev_line_length',
|
||||||
|
0)
|
||||||
|
if prev_len > len(fullmsg):
|
||||||
|
fullmsg += u' ' * (prev_len - len(fullmsg))
|
||||||
|
self._report_progress_prev_line_length = len(fullmsg)
|
||||||
|
clear_line = u'\r'
|
||||||
|
else:
|
||||||
|
clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
|
||||||
|
self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
|
||||||
|
self.to_console_title(u'youtube-dl ' + msg)
|
||||||
|
|
||||||
|
def report_progress(self, percent, data_len_str, speed, eta):
|
||||||
|
"""Report download progress."""
|
||||||
|
if self.params.get('noprogress', False):
|
||||||
|
return
|
||||||
|
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)
|
||||||
|
|
||||||
|
msg = (u'%s of %s at %s ETA %s' %
|
||||||
|
(percent_str, data_len_str, speed_str, eta_str))
|
||||||
|
self._report_progress_status(msg)
|
||||||
|
|
||||||
|
def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
|
||||||
|
if self.params.get('noprogress', False):
|
||||||
|
return
|
||||||
|
downloaded_str = format_bytes(downloaded_data_len)
|
||||||
|
speed_str = self.format_speed(speed)
|
||||||
|
elapsed_str = FileDownloader.format_seconds(elapsed)
|
||||||
|
msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
|
||||||
|
self._report_progress_status(msg)
|
||||||
|
|
||||||
|
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:
|
||||||
|
self._report_progress_status(
|
||||||
|
(u'100%% of %s in %s' %
|
||||||
|
(data_len_str, self.format_seconds(tot_time))),
|
||||||
|
is_last_line=True)
|
||||||
|
|
||||||
|
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 download(self, filename, info_dict):
|
||||||
|
"""Download to a filename using the info from info_dict
|
||||||
|
Return True on success and False otherwise
|
||||||
|
"""
|
||||||
|
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
|
||||||
|
else:
|
||||||
|
return self.real_download(filename, info_dict)
|
||||||
|
|
||||||
|
def real_download(self, filename, info_dict):
|
||||||
|
"""Real download process. Redefine in subclasses."""
|
||||||
|
raise NotImplementedError(u'This method must be implemented by sublcasses')
|
||||||
|
|
||||||
|
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)
|
||||||
|
|
44
youtube_dl/downloader/hls.py
Normal file
44
youtube_dl/downloader/hls.py
Normal file
@@ -0,0 +1,44 @@
|
|||||||
|
import os
|
||||||
|
import subprocess
|
||||||
|
|
||||||
|
from .common import FileDownloader
|
||||||
|
from ..utils import (
|
||||||
|
encodeFilename,
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
|
class HlsFD(FileDownloader):
|
||||||
|
def real_download(self, filename, info_dict):
|
||||||
|
url = info_dict['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
|
191
youtube_dl/downloader/http.py
Normal file
191
youtube_dl/downloader/http.py
Normal file
@@ -0,0 +1,191 @@
|
|||||||
|
import os
|
||||||
|
import re
|
||||||
|
import subprocess
|
||||||
|
import sys
|
||||||
|
import time
|
||||||
|
|
||||||
|
from .common import FileDownloader
|
||||||
|
from ..utils import (
|
||||||
|
compat_urllib_request,
|
||||||
|
compat_urllib_error,
|
||||||
|
ContentTooShortError,
|
||||||
|
|
||||||
|
encodeFilename,
|
||||||
|
sanitize_open,
|
||||||
|
format_bytes,
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
|
class HttpFD(FileDownloader):
|
||||||
|
def real_download(self, filename, info_dict):
|
||||||
|
url = info_dict['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):
|
||||||
|
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
|
39
youtube_dl/downloader/mplayer.py
Normal file
39
youtube_dl/downloader/mplayer.py
Normal file
@@ -0,0 +1,39 @@
|
|||||||
|
import os
|
||||||
|
import subprocess
|
||||||
|
|
||||||
|
from .common import FileDownloader
|
||||||
|
from ..utils import (
|
||||||
|
encodeFilename,
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
|
class MplayerFD(FileDownloader):
|
||||||
|
def real_download(self, filename, info_dict):
|
||||||
|
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
|
178
youtube_dl/downloader/rtmp.py
Normal file
178
youtube_dl/downloader/rtmp.py
Normal file
@@ -0,0 +1,178 @@
|
|||||||
|
import os
|
||||||
|
import re
|
||||||
|
import subprocess
|
||||||
|
import sys
|
||||||
|
import time
|
||||||
|
|
||||||
|
from .common import FileDownloader
|
||||||
|
from ..utils import (
|
||||||
|
encodeFilename,
|
||||||
|
format_bytes,
|
||||||
|
)
|
||||||
|
|
||||||
|
|
||||||
|
class RtmpFD(FileDownloader):
|
||||||
|
def real_download(self, filename, info_dict):
|
||||||
|
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
|
||||||
|
|
||||||
|
url = info_dict['url']
|
||||||
|
player_url = info_dict.get('player_url', None)
|
||||||
|
page_url = info_dict.get('page_url', None)
|
||||||
|
play_path = info_dict.get('play_path', None)
|
||||||
|
tc_url = info_dict.get('tc_url', None)
|
||||||
|
live = info_dict.get('rtmp_live', False)
|
||||||
|
conn = info_dict.get('rtmp_conn', None)
|
||||||
|
|
||||||
|
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
|
@@ -70,13 +70,14 @@ class BlipTVIE(InfoExtractor):
|
|||||||
info = None
|
info = None
|
||||||
urlh = self._request_webpage(request, None, False,
|
urlh = self._request_webpage(request, None, False,
|
||||||
u'unable to download video info webpage')
|
u'unable to download video info webpage')
|
||||||
|
|
||||||
if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download
|
if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download
|
||||||
basename = url.split('/')[-1]
|
basename = url.split('/')[-1]
|
||||||
title,ext = os.path.splitext(basename)
|
title,ext = os.path.splitext(basename)
|
||||||
title = title.decode('UTF-8')
|
title = title.decode('UTF-8')
|
||||||
ext = ext.replace('.', '')
|
ext = ext.replace('.', '')
|
||||||
self.report_direct_download(title)
|
self.report_direct_download(title)
|
||||||
info = {
|
return {
|
||||||
'id': title,
|
'id': title,
|
||||||
'url': url,
|
'url': url,
|
||||||
'uploader': None,
|
'uploader': None,
|
||||||
@@ -85,7 +86,7 @@ class BlipTVIE(InfoExtractor):
|
|||||||
'ext': ext,
|
'ext': ext,
|
||||||
'urlhandle': urlh
|
'urlhandle': urlh
|
||||||
}
|
}
|
||||||
if info is None: # Regular URL
|
|
||||||
try:
|
try:
|
||||||
json_code_bytes = urlh.read()
|
json_code_bytes = urlh.read()
|
||||||
json_code = json_code_bytes.decode('utf-8')
|
json_code = json_code_bytes.decode('utf-8')
|
||||||
@@ -111,7 +112,7 @@ class BlipTVIE(InfoExtractor):
|
|||||||
raise ValueError('Can not determine filename extension')
|
raise ValueError('Can not determine filename extension')
|
||||||
ext = umobj.group(1)
|
ext = umobj.group(1)
|
||||||
|
|
||||||
info = {
|
return {
|
||||||
'id': compat_str(data['item_id']),
|
'id': compat_str(data['item_id']),
|
||||||
'url': video_url,
|
'url': video_url,
|
||||||
'uploader': data['display_name'],
|
'uploader': data['display_name'],
|
||||||
@@ -124,11 +125,9 @@ class BlipTVIE(InfoExtractor):
|
|||||||
'player_url': data['embedUrl'],
|
'player_url': data['embedUrl'],
|
||||||
'user_agent': 'iTunes/10.6.1',
|
'user_agent': 'iTunes/10.6.1',
|
||||||
}
|
}
|
||||||
except (ValueError,KeyError) as err:
|
except (ValueError, KeyError) as err:
|
||||||
raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
|
raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
|
||||||
|
|
||||||
return [info]
|
|
||||||
|
|
||||||
|
|
||||||
class BlipTVUserIE(InfoExtractor):
|
class BlipTVUserIE(InfoExtractor):
|
||||||
"""Information Extractor for blip.tv users."""
|
"""Information Extractor for blip.tv users."""
|
||||||
|
@@ -162,23 +162,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
|
|||||||
# Dash audio
|
# Dash audio
|
||||||
'141', '172', '140', '171', '139',
|
'141', '172', '140', '171', '139',
|
||||||
]
|
]
|
||||||
_available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13',
|
|
||||||
# Apple HTTP Live Streaming
|
|
||||||
'96', '95', '94', '93', '92', '132', '151',
|
|
||||||
# 3D
|
|
||||||
'85', '102', '84', '101', '83', '100', '82',
|
|
||||||
# Dash video
|
|
||||||
'138', '248', '137', '247', '136', '246', '245',
|
|
||||||
'244', '135', '243', '134', '242', '133', '160',
|
|
||||||
# Dash audio
|
|
||||||
'172', '141', '171', '140', '139',
|
|
||||||
]
|
|
||||||
_video_formats_map = {
|
|
||||||
'flv': ['35', '34', '6', '5'],
|
|
||||||
'3gp': ['36', '17', '13'],
|
|
||||||
'mp4': ['38', '37', '22', '18'],
|
|
||||||
'webm': ['46', '45', '44', '43'],
|
|
||||||
}
|
|
||||||
_video_extensions = {
|
_video_extensions = {
|
||||||
'13': '3gp',
|
'13': '3gp',
|
||||||
'17': '3gp',
|
'17': '3gp',
|
||||||
@@ -236,54 +219,54 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
|
|||||||
'248': 'webm',
|
'248': 'webm',
|
||||||
}
|
}
|
||||||
_video_dimensions = {
|
_video_dimensions = {
|
||||||
'5': '400x240',
|
'5': {'width': 400, 'height': 240},
|
||||||
'6': '???',
|
'6': {},
|
||||||
'13': '???',
|
'13': {},
|
||||||
'17': '176x144',
|
'17': {'width': 176, 'height': 144},
|
||||||
'18': '640x360',
|
'18': {'width': 640, 'height': 360},
|
||||||
'22': '1280x720',
|
'22': {'width': 1280, 'height': 720},
|
||||||
'34': '640x360',
|
'34': {'width': 640, 'height': 360},
|
||||||
'35': '854x480',
|
'35': {'width': 854, 'height': 480},
|
||||||
'36': '320x240',
|
'36': {'width': 320, 'height': 240},
|
||||||
'37': '1920x1080',
|
'37': {'width': 1920, 'height': 1080},
|
||||||
'38': '4096x3072',
|
'38': {'width': 4096, 'height': 3072},
|
||||||
'43': '640x360',
|
'43': {'width': 640, 'height': 360},
|
||||||
'44': '854x480',
|
'44': {'width': 854, 'height': 480},
|
||||||
'45': '1280x720',
|
'45': {'width': 1280, 'height': 720},
|
||||||
'46': '1920x1080',
|
'46': {'width': 1920, 'height': 1080},
|
||||||
'82': '360p',
|
'82': {'height': 360, 'display': '360p'},
|
||||||
'83': '480p',
|
'83': {'height': 480, 'display': '480p'},
|
||||||
'84': '720p',
|
'84': {'height': 720, 'display': '720p'},
|
||||||
'85': '1080p',
|
'85': {'height': 1080, 'display': '1080p'},
|
||||||
'92': '240p',
|
'92': {'height': 240, 'display': '240p'},
|
||||||
'93': '360p',
|
'93': {'height': 360, 'display': '360p'},
|
||||||
'94': '480p',
|
'94': {'height': 480, 'display': '480p'},
|
||||||
'95': '720p',
|
'95': {'height': 720, 'display': '720p'},
|
||||||
'96': '1080p',
|
'96': {'height': 1080, 'display': '1080p'},
|
||||||
'100': '360p',
|
'100': {'height': 360, 'display': '360p'},
|
||||||
'101': '480p',
|
'101': {'height': 480, 'display': '480p'},
|
||||||
'102': '720p',
|
'102': {'height': 720, 'display': '720p'},
|
||||||
'132': '240p',
|
'132': {'height': 240, 'display': '240p'},
|
||||||
'151': '72p',
|
'151': {'height': 72, 'display': '72p'},
|
||||||
'133': '240p',
|
'133': {'height': 240, 'display': '240p'},
|
||||||
'134': '360p',
|
'134': {'height': 360, 'display': '360p'},
|
||||||
'135': '480p',
|
'135': {'height': 480, 'display': '480p'},
|
||||||
'136': '720p',
|
'136': {'height': 720, 'display': '720p'},
|
||||||
'137': '1080p',
|
'137': {'height': 1080, 'display': '1080p'},
|
||||||
'138': '>1080p',
|
'138': {'height': 1081, 'display': '>1080p'},
|
||||||
'139': '48k',
|
'139': {'display': '48k'},
|
||||||
'140': '128k',
|
'140': {'display': '128k'},
|
||||||
'141': '256k',
|
'141': {'display': '256k'},
|
||||||
'160': '192p',
|
'160': {'height': 192, 'display': '192p'},
|
||||||
'171': '128k',
|
'171': {'display': '128k'},
|
||||||
'172': '256k',
|
'172': {'display': '256k'},
|
||||||
'242': '240p',
|
'242': {'height': 240, 'display': '240p'},
|
||||||
'243': '360p',
|
'243': {'height': 360, 'display': '360p'},
|
||||||
'244': '480p',
|
'244': {'height': 480, 'display': '480p'},
|
||||||
'245': '480p',
|
'245': {'height': 480, 'display': '480p'},
|
||||||
'246': '480p',
|
'246': {'height': 480, 'display': '480p'},
|
||||||
'247': '720p',
|
'247': {'height': 720, 'display': '720p'},
|
||||||
'248': '1080p',
|
'248': {'height': 1080, 'display': '1080p'},
|
||||||
}
|
}
|
||||||
_special_itags = {
|
_special_itags = {
|
||||||
'82': '3D',
|
'82': '3D',
|
||||||
@@ -1153,13 +1136,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
|
|||||||
self._downloader.report_warning(err_msg)
|
self._downloader.report_warning(err_msg)
|
||||||
return {}
|
return {}
|
||||||
|
|
||||||
def _print_formats(self, formats):
|
|
||||||
print('Available formats:')
|
|
||||||
for x in formats:
|
|
||||||
print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
|
|
||||||
self._video_dimensions.get(x, '???'),
|
|
||||||
' ('+self._special_itags[x]+')' if x in self._special_itags else ''))
|
|
||||||
|
|
||||||
def _extract_id(self, url):
|
def _extract_id(self, url):
|
||||||
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
|
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
|
||||||
if mobj is None:
|
if mobj is None:
|
||||||
@@ -1172,48 +1148,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
|
|||||||
Transform a dictionary in the format {itag:url} to a list of (itag, url)
|
Transform a dictionary in the format {itag:url} to a list of (itag, url)
|
||||||
with the requested formats.
|
with the requested formats.
|
||||||
"""
|
"""
|
||||||
req_format = self._downloader.params.get('format', None)
|
existing_formats = [x for x in self._available_formats if x in url_map]
|
||||||
format_limit = self._downloader.params.get('format_limit', None)
|
|
||||||
available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
|
|
||||||
if format_limit is not None and format_limit in available_formats:
|
|
||||||
format_list = available_formats[available_formats.index(format_limit):]
|
|
||||||
else:
|
|
||||||
format_list = available_formats
|
|
||||||
existing_formats = [x for x in format_list if x in url_map]
|
|
||||||
if len(existing_formats) == 0:
|
if len(existing_formats) == 0:
|
||||||
raise ExtractorError(u'no known formats available for video')
|
raise ExtractorError(u'no known formats available for video')
|
||||||
if self._downloader.params.get('listformats', None):
|
|
||||||
self._print_formats(existing_formats)
|
|
||||||
return
|
|
||||||
if req_format is None or req_format == 'best':
|
|
||||||
video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
|
|
||||||
elif req_format == 'worst':
|
|
||||||
video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
|
|
||||||
elif req_format in ('-1', 'all'):
|
|
||||||
video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
|
video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
|
||||||
else:
|
video_url_list.reverse() # order worst to best
|
||||||
# Specific formats. We pick the first in a slash-delimeted sequence.
|
|
||||||
# Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality
|
|
||||||
# available in the specified format. For example,
|
|
||||||
# if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
|
|
||||||
# if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'.
|
|
||||||
# if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'.
|
|
||||||
req_formats = req_format.split('/')
|
|
||||||
video_url_list = None
|
|
||||||
for rf in req_formats:
|
|
||||||
if rf in url_map:
|
|
||||||
video_url_list = [(rf, url_map[rf])]
|
|
||||||
break
|
|
||||||
if rf in self._video_formats_map:
|
|
||||||
for srf in self._video_formats_map[rf]:
|
|
||||||
if srf in url_map:
|
|
||||||
video_url_list = [(srf, url_map[srf])]
|
|
||||||
break
|
|
||||||
else:
|
|
||||||
continue
|
|
||||||
break
|
|
||||||
if video_url_list is None:
|
|
||||||
raise ExtractorError(u'requested format not available')
|
|
||||||
return video_url_list
|
return video_url_list
|
||||||
|
|
||||||
def _extract_from_m3u8(self, manifest_url, video_id):
|
def _extract_from_m3u8(self, manifest_url, video_id):
|
||||||
@@ -1462,40 +1401,57 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
|
|||||||
url += '&ratebypass=yes'
|
url += '&ratebypass=yes'
|
||||||
url_map[url_data['itag'][0]] = url
|
url_map[url_data['itag'][0]] = url
|
||||||
video_url_list = self._get_video_url_list(url_map)
|
video_url_list = self._get_video_url_list(url_map)
|
||||||
if not video_url_list:
|
|
||||||
return
|
|
||||||
elif video_info.get('hlsvp'):
|
elif video_info.get('hlsvp'):
|
||||||
manifest_url = video_info['hlsvp'][0]
|
manifest_url = video_info['hlsvp'][0]
|
||||||
url_map = self._extract_from_m3u8(manifest_url, video_id)
|
url_map = self._extract_from_m3u8(manifest_url, video_id)
|
||||||
video_url_list = self._get_video_url_list(url_map)
|
video_url_list = self._get_video_url_list(url_map)
|
||||||
if not video_url_list:
|
|
||||||
return
|
|
||||||
|
|
||||||
else:
|
else:
|
||||||
raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
|
raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
|
||||||
|
|
||||||
results = []
|
formats = []
|
||||||
for itag, video_real_url in video_url_list:
|
for itag, video_real_url in video_url_list:
|
||||||
# Extension
|
# Extension
|
||||||
video_extension = self._video_extensions.get(itag, 'flv')
|
video_extension = self._video_extensions.get(itag, 'flv')
|
||||||
|
resolution = self._video_dimensions.get(itag, {}).get('display')
|
||||||
|
width = self._video_dimensions.get(itag, {}).get('width')
|
||||||
|
height = self._video_dimensions.get(itag, {}).get('height')
|
||||||
|
note = self._special_itags.get(itag)
|
||||||
|
|
||||||
video_format = '{0} - {1}{2}'.format(itag if itag else video_extension,
|
video_format = '{0} - {1}{2}'.format(itag if itag else video_extension,
|
||||||
self._video_dimensions.get(itag, '???'),
|
'%dx%d' % (width, height) if width is not None and height is not None else (resolution if resolution is not None else '???'),
|
||||||
' ('+self._special_itags[itag]+')' if itag in self._special_itags else '')
|
' ('+self._special_itags[itag]+')' if itag in self._special_itags else '')
|
||||||
|
|
||||||
results.append({
|
formats.append({
|
||||||
'id': video_id,
|
|
||||||
'url': video_real_url,
|
'url': video_real_url,
|
||||||
|
'ext': video_extension,
|
||||||
|
'format': video_format,
|
||||||
|
'format_id': itag,
|
||||||
|
'player_url': player_url,
|
||||||
|
'_resolution': resolution,
|
||||||
|
'width': width,
|
||||||
|
'height': height,
|
||||||
|
'format_note': note,
|
||||||
|
})
|
||||||
|
|
||||||
|
def _formats_key(f):
|
||||||
|
note = f.get('format_note')
|
||||||
|
if note is None:
|
||||||
|
note = u''
|
||||||
|
is_dash = u'DASH' in note
|
||||||
|
return (
|
||||||
|
0 if is_dash else 1,
|
||||||
|
f.get('height') if f.get('height') is not None else -1,
|
||||||
|
f.get('width') if f.get('width') is not None else -1)
|
||||||
|
formats.sort(key=_formats_key)
|
||||||
|
|
||||||
|
return {
|
||||||
|
'id': video_id,
|
||||||
'uploader': video_uploader,
|
'uploader': video_uploader,
|
||||||
'uploader_id': video_uploader_id,
|
'uploader_id': video_uploader_id,
|
||||||
'upload_date': upload_date,
|
'upload_date': upload_date,
|
||||||
'title': video_title,
|
'title': video_title,
|
||||||
'ext': video_extension,
|
|
||||||
'format': video_format,
|
|
||||||
'format_id': itag,
|
|
||||||
'thumbnail': video_thumbnail,
|
'thumbnail': video_thumbnail,
|
||||||
'description': video_description,
|
'description': video_description,
|
||||||
'player_url': player_url,
|
|
||||||
'subtitles': video_subtitles,
|
'subtitles': video_subtitles,
|
||||||
'duration': video_duration,
|
'duration': video_duration,
|
||||||
'age_limit': 18 if age_gate else 0,
|
'age_limit': 18 if age_gate else 0,
|
||||||
@@ -1504,8 +1460,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
|
|||||||
'view_count': view_count,
|
'view_count': view_count,
|
||||||
'like_count': like_count,
|
'like_count': like_count,
|
||||||
'dislike_count': dislike_count,
|
'dislike_count': dislike_count,
|
||||||
})
|
'formats': formats,
|
||||||
return results
|
}
|
||||||
|
|
||||||
class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
|
class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
|
||||||
IE_DESC = u'YouTube.com playlists'
|
IE_DESC = u'YouTube.com playlists'
|
||||||
|
@@ -1,2 +1,2 @@
|
|||||||
|
|
||||||
__version__ = '2013.12.23'
|
__version__ = '2013.12.23.4'
|
||||||
|
Reference in New Issue
Block a user