aboutsummaryrefslogtreecommitdiffhomepage
path: root/youtube_dl/utils.py
diff options
context:
space:
mode:
authorFilippo Valsorda <[email protected]>2018-01-07 15:03:28 +0100
committerGitHub <[email protected]>2018-01-07 15:03:28 +0100
commit97bc05116e15df3cf1b7b7216c60fad7ee0acc9f (patch)
tree453059413184d22a44241d251a82196c5d2461eb /youtube_dl/utils.py
parent7608a91ee7b40c74a35c30b44731606382702592 (diff)
parent0a5b1295b7c1aa6395b65ee137087c540b37b32b (diff)
downloadyoutube-dl-97bc05116e15df3cf1b7b7216c60fad7ee0acc9f.tar.gz
youtube-dl-97bc05116e15df3cf1b7b7216c60fad7ee0acc9f.zip
Merge branch 'master' into totalwebcastingtotalwebcasting
Diffstat (limited to 'youtube_dl/utils.py')
-rw-r--r--youtube_dl/utils.py757
1 files changed, 668 insertions, 89 deletions
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 39dd6c49f..386897a85 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -11,6 +11,7 @@ import contextlib
import ctypes
import datetime
import email.utils
+import email.header
import errno
import functools
import gzip
@@ -21,8 +22,8 @@ import locale
import math
import operator
import os
-import pipes
import platform
+import random
import re
import socket
import ssl
@@ -34,10 +35,13 @@ import xml.etree.ElementTree
import zlib
from .compat import (
+ compat_HTMLParseError,
compat_HTMLParser,
compat_basestring,
compat_chr,
+ compat_ctypes_WINFUNCTYPE,
compat_etree_fromstring,
+ compat_expanduser,
compat_html_entities,
compat_html_entities_html5,
compat_http_client,
@@ -128,7 +132,13 @@ DATE_FORMATS = (
'%d %B %Y',
'%d %b %Y',
'%B %d %Y',
+ '%B %dst %Y',
+ '%B %dnd %Y',
+ '%B %dth %Y',
'%b %d %Y',
+ '%b %dst %Y',
+ '%b %dnd %Y',
+ '%b %dth %Y',
'%b %dst %Y %I:%M',
'%b %dnd %Y %I:%M',
'%b %dth %Y %I:%M',
@@ -137,6 +147,7 @@ DATE_FORMATS = (
'%Y/%m/%d',
'%Y/%m/%d %H:%M',
'%Y/%m/%d %H:%M:%S',
+ '%Y-%m-%d %H:%M',
'%Y-%m-%d %H:%M:%S',
'%Y-%m-%d %H:%M:%S.%f',
'%d.%m.%Y %H:%M',
@@ -149,6 +160,8 @@ DATE_FORMATS = (
'%Y-%m-%dT%H:%M',
'%b %d %Y at %H:%M',
'%b %d %Y at %H:%M:%S',
+ '%B %d %Y at %H:%M',
+ '%B %d %Y at %H:%M:%S',
)
DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS)
@@ -330,34 +343,46 @@ def get_element_by_id(id, html):
def get_element_by_class(class_name, html):
- return get_element_by_attribute(
+ """Return the content of the first tag with the specified class in the passed HTML document"""
+ retval = get_elements_by_class(class_name, html)
+ return retval[0] if retval else None
+
+
+def get_element_by_attribute(attribute, value, html, escape_value=True):
+ retval = get_elements_by_attribute(attribute, value, html, escape_value)
+ return retval[0] if retval else None
+
+
+def get_elements_by_class(class_name, html):
+ """Return the content of all tags with the specified class in the passed HTML document as a list"""
+ return get_elements_by_attribute(
'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
html, escape_value=False)
-def get_element_by_attribute(attribute, value, html, escape_value=True):
+def get_elements_by_attribute(attribute, value, html, escape_value=True):
"""Return the content of the tag with the specified attribute in the passed HTML document"""
value = re.escape(value) if escape_value else value
- m = re.search(r'''(?xs)
+ retlist = []
+ for m in re.finditer(r'''(?xs)
<([a-zA-Z0-9:._-]+)
- (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+ (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s+%s=['"]?%s['"]?
- (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+ (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s*>
(?P<content>.*?)
</\1>
- ''' % (re.escape(attribute), value), html)
+ ''' % (re.escape(attribute), value), html):
+ res = m.group('content')
- if not m:
- return None
- res = m.group('content')
+ if res.startswith('"') or res.startswith("'"):
+ res = res[1:-1]
- if res.startswith('"') or res.startswith("'"):
- res = res[1:-1]
+ retlist.append(unescapeHTML(res))
- return unescapeHTML(res)
+ return retlist
class HTMLAttributeParser(compat_HTMLParser):
@@ -387,8 +412,12 @@ def extract_attributes(html_element):
but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5.
"""
parser = HTMLAttributeParser()
- parser.feed(html_element)
- parser.close()
+ try:
+ parser.feed(html_element)
+ parser.close()
+ # Older Python may throw HTMLParseError in case of malformed HTML
+ except compat_HTMLParseError:
+ pass
return parser.attrs
@@ -400,8 +429,8 @@ def clean_html(html):
# Newline vs <br />
html = html.replace('\n', ' ')
- html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html)
- html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
+ html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html)
+ html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
# Strip html tags
html = re.sub('<.*?>', '', html)
# Replace html entities
@@ -453,7 +482,8 @@ def timeconvert(timestr):
def sanitize_filename(s, restricted=False, is_id=False):
"""Sanitizes a string so it could be used as part of a filename.
If restricted is set, use a stricter subset of allowed characters.
- Set is_id if this is not an arbitrary string, but an ID that should be kept if possible
+ Set is_id if this is not an arbitrary string, but an ID that should be kept
+ if possible.
"""
def replace_insane(char):
if restricted and char in ACCENT_CHARS:
@@ -518,6 +548,11 @@ def sanitized_Request(url, *args, **kwargs):
return compat_urllib_request.Request(sanitize_url(url), *args, **kwargs)
+def expand_path(s):
+ """Expand shell variables and ~"""
+ return os.path.expandvars(compat_expanduser(s))
+
+
def orderedSet(iterable):
""" Remove all duplicates from the input iterable """
res = []
@@ -564,7 +599,7 @@ def unescapeHTML(s):
assert type(s) == compat_str
return re.sub(
- r'&([^;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
+ r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s)
def get_subprocess_encoding():
@@ -682,7 +717,12 @@ def bug_reports_message():
return msg
-class ExtractorError(Exception):
+class YoutubeDLError(Exception):
+ """Base exception for YoutubeDL errors."""
+ pass
+
+
+class ExtractorError(YoutubeDLError):
"""Error during info extraction."""
def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None):
@@ -723,7 +763,19 @@ class RegexNotFoundError(ExtractorError):
pass
-class DownloadError(Exception):
+class GeoRestrictedError(ExtractorError):
+ """Geographic restriction Error exception.
+
+ This exception may be thrown when a video is not available from your
+ geographic location due to geographic restrictions imposed by a website.
+ """
+ def __init__(self, msg, countries=None):
+ super(GeoRestrictedError, self).__init__(msg, expected=True)
+ self.msg = msg
+ self.countries = countries
+
+
+class DownloadError(YoutubeDLError):
"""Download Error exception.
This exception may be thrown by FileDownloader objects if they are not
@@ -737,7 +789,7 @@ class DownloadError(Exception):
self.exc_info = exc_info
-class SameFileError(Exception):
+class SameFileError(YoutubeDLError):
"""Same File exception.
This exception will be thrown by FileDownloader objects if they detect
@@ -746,7 +798,7 @@ class SameFileError(Exception):
pass
-class PostProcessingError(Exception):
+class PostProcessingError(YoutubeDLError):
"""Post Processing exception.
This exception may be raised by PostProcessor's .run() method to
@@ -754,15 +806,16 @@ class PostProcessingError(Exception):
"""
def __init__(self, msg):
+ super(PostProcessingError, self).__init__(msg)
self.msg = msg
-class MaxDownloadsReached(Exception):
+class MaxDownloadsReached(YoutubeDLError):
""" --max-downloads limit has been reached. """
pass
-class UnavailableVideoError(Exception):
+class UnavailableVideoError(YoutubeDLError):
"""Unavailable Format exception.
This exception will be thrown when a video is requested
@@ -771,7 +824,7 @@ class UnavailableVideoError(Exception):
pass
-class ContentTooShortError(Exception):
+class ContentTooShortError(YoutubeDLError):
"""Content Too Short exception.
This exception may be raised by FileDownloader objects when a file they
@@ -780,12 +833,15 @@ class ContentTooShortError(Exception):
"""
def __init__(self, downloaded, expected):
+ super(ContentTooShortError, self).__init__(
+ 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected)
+ )
# Both in bytes
self.downloaded = downloaded
self.expected = expected
-class XAttrMetadataError(Exception):
+class XAttrMetadataError(YoutubeDLError):
def __init__(self, code=None, msg='Unknown error'):
super(XAttrMetadataError, self).__init__(msg)
self.code = code
@@ -801,7 +857,7 @@ class XAttrMetadataError(Exception):
self.reason = 'NOT_SUPPORTED'
-class XAttrUnavailableError(Exception):
+class XAttrUnavailableError(YoutubeDLError):
pass
@@ -883,14 +939,6 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
except zlib.error:
return zlib.decompress(data)
- @staticmethod
- def addinfourl_wrapper(stream, headers, url, code):
- if hasattr(compat_urllib_request.addinfourl, 'getcode'):
- return compat_urllib_request.addinfourl(stream, headers, url, code)
- ret = compat_urllib_request.addinfourl(stream, headers, url)
- ret.code = code
- return ret
-
def http_request(self, req):
# According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
# always respected by websites, some tend to give out URLs with non percent-encoded
@@ -942,13 +990,13 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
break
else:
raise original_ioerror
- resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
+ resp = compat_urllib_request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
del resp.headers['Content-encoding']
# deflate
if resp.headers.get('Content-encoding', '') == 'deflate':
gz = io.BytesIO(self.deflate(resp.read()))
- resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
+ resp = compat_urllib_request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
del resp.headers['Content-encoding']
# Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
@@ -1138,7 +1186,7 @@ def unified_timestamp(date_str, day_first=True):
if date_str is None:
return None
- date_str = date_str.replace(',', ' ')
+ date_str = re.sub(r'[,|]', '', date_str)
pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0
timezone, date_str = extract_timezone(date_str)
@@ -1146,6 +1194,11 @@ def unified_timestamp(date_str, day_first=True):
# Remove AM/PM + timezone
date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
+ # Remove unrecognized timezones from ISO 8601 alike timestamps
+ m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str)
+ if m:
+ date_str = date_str[:-len(m.group('tz'))]
+
for expression in date_formats(day_first):
try:
dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
@@ -1278,24 +1331,24 @@ def _windows_write_string(s, out):
if fileno not in WIN_OUTPUT_IDS:
return False
- GetStdHandle = ctypes.WINFUNCTYPE(
+ GetStdHandle = compat_ctypes_WINFUNCTYPE(
ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)(
- (b'GetStdHandle', ctypes.windll.kernel32))
+ ('GetStdHandle', ctypes.windll.kernel32))
h = GetStdHandle(WIN_OUTPUT_IDS[fileno])
- WriteConsoleW = ctypes.WINFUNCTYPE(
+ WriteConsoleW = compat_ctypes_WINFUNCTYPE(
ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR,
ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD),
- ctypes.wintypes.LPVOID)((b'WriteConsoleW', ctypes.windll.kernel32))
+ ctypes.wintypes.LPVOID)(('WriteConsoleW', ctypes.windll.kernel32))
written = ctypes.wintypes.DWORD(0)
- GetFileType = ctypes.WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)((b'GetFileType', ctypes.windll.kernel32))
+ GetFileType = compat_ctypes_WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)(('GetFileType', ctypes.windll.kernel32))
FILE_TYPE_CHAR = 0x0002
FILE_TYPE_REMOTE = 0x8000
- GetConsoleMode = ctypes.WINFUNCTYPE(
+ GetConsoleMode = compat_ctypes_WINFUNCTYPE(
ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE,
ctypes.POINTER(ctypes.wintypes.DWORD))(
- (b'GetConsoleMode', ctypes.windll.kernel32))
+ ('GetConsoleMode', ctypes.windll.kernel32))
INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value
def not_a_console(handle):
@@ -1484,7 +1537,7 @@ def shell_quote(args):
if isinstance(a, bytes):
# We may get a filename encoded with 'encodeFilename'
a = a.decode(encoding)
- quoted_args.append(pipes.quote(a))
+ quoted_args.append(compat_shlex_quote(a))
return ' '.join(quoted_args)
@@ -1665,6 +1718,11 @@ def setproctitle(title):
libc = ctypes.cdll.LoadLibrary('libc.so.6')
except OSError:
return
+ except TypeError:
+ # LoadLibrary in Windows Python 2.7.13 only expects
+ # a bytestring, but since unicode_literals turns
+ # every string into a unicode string, it fails.
+ return
title_bytes = title.encode('utf-8')
buf = ctypes.create_string_buffer(len(title_bytes))
buf.value = title_bytes
@@ -1701,11 +1759,16 @@ def base_url(url):
def urljoin(base, path):
+ if isinstance(path, bytes):
+ path = path.decode('utf-8')
if not isinstance(path, compat_str) or not path:
return None
if re.match(r'^(?:https?:)?//', path):
return path
- if not isinstance(base, compat_str) or not re.match(r'^(?:https?:)?//', base):
+ if isinstance(base, bytes):
+ base = base.decode('utf-8')
+ if not isinstance(base, compat_str) or not re.match(
+ r'^(?:https?:)?//', base):
return None
return compat_urlparse.urljoin(base, path)
@@ -1755,6 +1818,10 @@ def float_or_none(v, scale=1, invscale=1, default=None):
return default
+def bool_or_none(v, default=None):
+ return v if isinstance(v, bool) else default
+
+
def strip_or_none(v):
return None if v is None else v.strip()
@@ -1766,15 +1833,25 @@ def parse_duration(s):
s = s.strip()
days, hours, mins, secs, ms = [None] * 5
- m = re.match(r'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?$', s)
+ m = re.match(r'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?Z?$', s)
if m:
days, hours, mins, secs, ms = m.groups()
else:
m = re.match(
- r'''(?ix)(?:P?T)?
+ r'''(?ix)(?:P?
+ (?:
+ [0-9]+\s*y(?:ears?)?\s*
+ )?
+ (?:
+ [0-9]+\s*m(?:onths?)?\s*
+ )?
+ (?:
+ [0-9]+\s*w(?:eeks?)?\s*
+ )?
(?:
(?P<days>[0-9]+)\s*d(?:ays?)?\s*
)?
+ T)?
(?:
(?P<hours>[0-9]+)\s*h(?:ours?)?\s*
)?
@@ -1783,11 +1860,11 @@ def parse_duration(s):
)?
(?:
(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s*
- )?$''', s)
+ )?Z?$''', s)
if m:
days, hours, mins, secs, ms = m.groups()
else:
- m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)$', s)
+ m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s)
if m:
hours, mins = m.groups()
else:
@@ -1869,7 +1946,7 @@ class PagedList(object):
class OnDemandPagedList(PagedList):
- def __init__(self, pagefunc, pagesize, use_cache=False):
+ def __init__(self, pagefunc, pagesize, use_cache=True):
self._pagefunc = pagefunc
self._pagesize = pagesize
self._use_cache = use_cache
@@ -2034,6 +2111,58 @@ def update_Request(req, url=None, data=None, headers={}, query={}):
return new_req
+def _multipart_encode_impl(data, boundary):
+ content_type = 'multipart/form-data; boundary=%s' % boundary
+
+ out = b''
+ for k, v in data.items():
+ out += b'--' + boundary.encode('ascii') + b'\r\n'
+ if isinstance(k, compat_str):
+ k = k.encode('utf-8')
+ if isinstance(v, compat_str):
+ v = v.encode('utf-8')
+ # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578
+ # suggests sending UTF-8 directly. Firefox sends UTF-8, too
+ content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n'
+ if boundary.encode('ascii') in content:
+ raise ValueError('Boundary overlaps with data')
+ out += content
+
+ out += b'--' + boundary.encode('ascii') + b'--\r\n'
+
+ return out, content_type
+
+
+def multipart_encode(data, boundary=None):
+ '''
+ Encode a dict to RFC 7578-compliant form-data
+
+ data:
+ A dict where keys and values can be either Unicode or bytes-like
+ objects.
+ boundary:
+ If specified a Unicode object, it's used as the boundary. Otherwise
+ a random boundary is generated.
+
+ Reference: https://tools.ietf.org/html/rfc7578
+ '''
+ has_specified_boundary = boundary is not None
+
+ while True:
+ if boundary is None:
+ boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff))
+
+ try:
+ out, content_type = _multipart_encode_impl(data, boundary)
+ break
+ except ValueError:
+ if has_specified_boundary:
+ raise
+ boundary = None
+
+ return out, content_type
+
+
def dict_get(d, key_or_keys, default=None, skip_false_values=True):
if isinstance(key_or_keys, (list, tuple)):
for key in key_or_keys:
@@ -2045,13 +2174,16 @@ def dict_get(d, key_or_keys, default=None, skip_false_values=True):
def try_get(src, getter, expected_type=None):
- try:
- v = getter(src)
- except (AttributeError, KeyError, TypeError, IndexError):
- pass
- else:
- if expected_type is None or isinstance(v, expected_type):
- return v
+ if not isinstance(getter, (list, tuple)):
+ getter = [getter]
+ for get in getter:
+ try:
+ v = get(src)
+ except (AttributeError, KeyError, TypeError, IndexError):
+ pass
+ else:
+ if expected_type is None or isinstance(v, expected_type):
+ return v
def encode_compat_str(string, encoding=preferredencoding(), errors='strict'):
@@ -2092,15 +2224,27 @@ def parse_age_limit(s):
def strip_jsonp(code):
return re.sub(
- r'(?s)^[a-zA-Z0-9_.$]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
+ r'''(?sx)^
+ (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]+)
+ (?:\s*&&\s*(?P=func_name))?
+ \s*\(\s*(?P<callback_data>.*)\);?
+ \s*?(?://[^\n]*)*$''',
+ r'\g<callback_data>', code)
def js_to_json(code):
+ COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
+ SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE)
+ INTEGER_TABLE = (
+ (r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16),
+ (r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
+ )
+
def fix_kv(m):
v = m.group(0)
if v in ('true', 'false', 'null'):
return v
- elif v.startswith('/*') or v == ',':
+ elif v.startswith('/*') or v.startswith('//') or v == ',':
return ""
if v[0] in ("'", '"'):
@@ -2111,11 +2255,6 @@ def js_to_json(code):
'\\x': '\\u00',
}.get(m.group(0), m.group(0)), v[1:-1])
- INTEGER_TABLE = (
- (r'^(0[xX][0-9a-fA-F]+)\s*:?$', 16),
- (r'^(0+[0-7]+)\s*:?$', 8),
- )
-
for regex, base in INTEGER_TABLE:
im = re.match(regex, v)
if im:
@@ -2127,11 +2266,11 @@ def js_to_json(code):
return re.sub(r'''(?sx)
"(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
'(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
- /\*.*?\*/|,(?=\s*[\]}])|
+ {comment}|,(?={skip}[\]}}])|
[a-zA-Z_][.a-zA-Z_0-9]*|
- \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:\s*:)?|
- [0-9]+(?=\s*:)
- ''', fix_kv, code)
+ \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
+ [0-9]+(?={skip}:)
+ '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code)
def qualities(quality_ids):
@@ -2210,22 +2349,21 @@ def mimetype2ext(mt):
return {
'3gpp': '3gp',
'smptett+xml': 'tt',
- 'srt': 'srt',
'ttaf+xml': 'dfxp',
'ttml+xml': 'ttml',
- 'vtt': 'vtt',
'x-flv': 'flv',
'x-mp4-fragmented': 'mp4',
+ 'x-ms-sami': 'sami',
'x-ms-wmv': 'wmv',
'mpegurl': 'm3u8',
'x-mpegurl': 'm3u8',
'vnd.apple.mpegurl': 'm3u8',
'dash+xml': 'mpd',
- 'f4m': 'f4m',
'f4m+xml': 'f4m',
'hds+xml': 'f4m',
'vnd.ms-sstr+xml': 'ism',
'quicktime': 'mov',
+ 'mp2t': 'ts',
}.get(res, res)
@@ -2238,14 +2376,14 @@ def parse_codecs(codecs_str):
vcodec, acodec = None, None
for full_codec in splited_codecs:
codec = full_codec.split('.')[0]
- if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v'):
+ if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1'):
if not vcodec:
vcodec = full_codec
- elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3'):
+ elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
if not acodec:
acodec = full_codec
else:
- write_string('WARNING: Unknown codec %s' % full_codec, sys.stderr)
+ write_string('WARNING: Unknown codec %s\n' % full_codec, sys.stderr)
if not vcodec and not acodec:
if len(splited_codecs) == 2:
return {
@@ -2357,6 +2495,7 @@ def _match_one(filter_part, dct):
\s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
(?:
(?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
+ (?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)|
(?P<strval>(?![0-9.])[a-z0-9A-Z]*)
)
\s*$
@@ -2365,7 +2504,8 @@ def _match_one(filter_part, dct):
if m:
op = COMPARISON_OPERATORS[m.group('op')]
actual_value = dct.get(m.group('key'))
- if (m.group('strval') is not None or
+ if (m.group('quotedstrval') is not None or
+ m.group('strval') is not None or
# If the original field is a string and matching comparisonvalue is
# a number we should respect the origin of the original field
# and process comparison value as a string (see
@@ -2375,7 +2515,10 @@ def _match_one(filter_part, dct):
if m.group('op') not in ('=', '!='):
raise ValueError(
'Operator %s does not support string values!' % m.group('op'))
- comparison_value = m.group('strval') or m.group('intval')
+ comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval')
+ quote = m.group('quote')
+ if quote is not None:
+ comparison_value = comparison_value.replace(r'\%s' % quote, quote)
else:
try:
comparison_value = int(m.group('intval'))
@@ -2443,27 +2586,101 @@ def srt_subtitles_timecode(seconds):
def dfxp2srt(dfxp_data):
+ '''
+ @param dfxp_data A bytes-like object containing DFXP data
+ @returns A unicode object containing converted SRT data
+ '''
+ LEGACY_NAMESPACES = (
+ (b'http://www.w3.org/ns/ttml', [
+ b'http://www.w3.org/2004/11/ttaf1',
+ b'http://www.w3.org/2006/04/ttaf1',
+ b'http://www.w3.org/2006/10/ttaf1',
+ ]),
+ (b'http://www.w3.org/ns/ttml#styling', [
+ b'http://www.w3.org/ns/ttml#style',
+ ]),
+ )
+
+ SUPPORTED_STYLING = [
+ 'color',
+ 'fontFamily',
+ 'fontSize',
+ 'fontStyle',
+ 'fontWeight',
+ 'textDecoration'
+ ]
+
_x = functools.partial(xpath_with_ns, ns_map={
'ttml': 'http://www.w3.org/ns/ttml',
- 'ttaf1': 'http://www.w3.org/2006/10/ttaf1',
- 'ttaf1_0604': 'http://www.w3.org/2006/04/ttaf1',
+ 'tts': 'http://www.w3.org/ns/ttml#styling',
})
+ styles = {}
+ default_style = {}
+
class TTMLPElementParser(object):
- out = ''
+ _out = ''
+ _unclosed_elements = []
+ _applied_styles = []
def start(self, tag, attrib):
- if tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'):
- self.out += '\n'
+ if tag in (_x('ttml:br'), 'br'):
+ self._out += '\n'
+ else:
+ unclosed_elements = []
+ style = {}
+ element_style_id = attrib.get('style')
+ if default_style:
+ style.update(default_style)
+ if element_style_id:
+ style.update(styles.get(element_style_id, {}))
+ for prop in SUPPORTED_STYLING:
+ prop_val = attrib.get(_x('tts:' + prop))
+ if prop_val:
+ style[prop] = prop_val
+ if style:
+ font = ''
+ for k, v in sorted(style.items()):
+ if self._applied_styles and self._applied_styles[-1].get(k) == v:
+ continue
+ if k == 'color':
+ font += ' color="%s"' % v
+ elif k == 'fontSize':
+ font += ' size="%s"' % v
+ elif k == 'fontFamily':
+ font += ' face="%s"' % v
+ elif k == 'fontWeight' and v == 'bold':
+ self._out += '<b>'
+ unclosed_elements.append('b')
+ elif k == 'fontStyle' and v == 'italic':
+ self._out += '<i>'
+ unclosed_elements.append('i')
+ elif k == 'textDecoration' and v == 'underline':
+ self._out += '<u>'
+ unclosed_elements.append('u')
+ if font:
+ self._out += '<font' + font + '>'
+ unclosed_elements.append('font')
+ applied_style = {}
+ if self._applied_styles:
+ applied_style.update(self._applied_styles[-1])
+ applied_style.update(style)
+ self._applied_styles.append(applied_style)
+ self._unclosed_elements.append(unclosed_elements)
def end(self, tag):
- pass
+ if tag not in (_x('ttml:br'), 'br'):
+ unclosed_elements = self._unclosed_elements.pop()
+ for element in reversed(unclosed_elements):
+ self._out += '</%s>' % element
+ if unclosed_elements and self._applied_styles:
+ self._applied_styles.pop()
def data(self, data):
- self.out += data
+ self._out += data
def close(self):
- return self.out.strip()
+ return self._out.strip()
def parse_node(node):
target = TTMLPElementParser()
@@ -2471,13 +2688,45 @@ def dfxp2srt(dfxp_data):
parser.feed(xml.etree.ElementTree.tostring(node))
return parser.close()
- dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8'))
+ for k, v in LEGACY_NAMESPACES:
+ for ns in v:
+ dfxp_data = dfxp_data.replace(ns, k)
+
+ dfxp = compat_etree_fromstring(dfxp_data)
out = []
- paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall(_x('.//ttaf1:p')) or dfxp.findall(_x('.//ttaf1_0604:p')) or dfxp.findall('.//p')
+ paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
if not paras:
raise ValueError('Invalid dfxp/TTML subtitle')
+ repeat = False
+ while True:
+ for style in dfxp.findall(_x('.//ttml:style')):
+ style_id = style.get('id')
+ parent_style_id = style.get('style')
+ if parent_style_id:
+ if parent_style_id not in styles:
+ repeat = True
+ continue
+ styles[style_id] = styles[parent_style_id].copy()
+ for prop in SUPPORTED_STYLING:
+ prop_val = style.get(_x('tts:' + prop))
+ if prop_val:
+ styles.setdefault(style_id, {})[prop] = prop_val
+ if repeat:
+ repeat = False
+ else:
+ break
+
+ for p in ('body', 'div'):
+ ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p])
+ if ele is None:
+ continue
+ style = styles.get(ele.get('style'))
+ if not style:
+ continue
+ default_style.update(style)
+
for para, index in zip(paras, itertools.count(1)):
begin_time = parse_dfxp_time_expr(para.attrib.get('begin'))
end_time = parse_dfxp_time_expr(para.attrib.get('end'))
@@ -2506,6 +2755,8 @@ def cli_option(params, command_option, param):
def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None):
param = params.get(param)
+ if param is None:
+ return []
assert isinstance(param, bool)
if separator:
return [command_option + separator + (true_value if param else false_value)]
@@ -2987,6 +3238,260 @@ class ISO3166Utils(object):
return cls._country_map.get(code.upper())
+class GeoUtils(object):
+ # Major IPv4 address blocks per country
+ _country_ip_map = {
+ 'AD': '85.94.160.0/19',
+ 'AE': '94.200.0.0/13',
+ 'AF': '149.54.0.0/17',
+ 'AG': '209.59.64.0/18',
+ 'AI': '204.14.248.0/21',
+ 'AL': '46.99.0.0/16',
+ 'AM': '46.70.0.0/15',
+ 'AO': '105.168.0.0/13',
+ 'AP': '159.117.192.0/21',
+ 'AR': '181.0.0.0/12',
+ 'AS': '202.70.112.0/20',
+ 'AT': '84.112.0.0/13',
+ 'AU': '1.128.0.0/11',
+ 'AW': '181.41.0.0/18',
+ 'AZ': '5.191.0.0/16',
+ 'BA': '31.176.128.0/17',
+ 'BB': '65.48.128.0/17',
+ 'BD': '114.130.0.0/16',
+ 'BE': '57.0.0.0/8',
+ 'BF': '129.45.128.0/17',
+ 'BG': '95.42.0.0/15',
+ 'BH': '37.131.0.0/17',
+ 'BI': '154.117.192.0/18',
+ 'BJ': '137.255.0.0/16',
+ 'BL': '192.131.134.0/24',
+ 'BM': '196.12.64.0/18',
+ 'BN': '156.31.0.0/16',
+ 'BO': '161.56.0.0/16',
+ 'BQ': '161.0.80.0/20',
+ 'BR': '152.240.0.0/12',
+ 'BS': '24.51.64.0/18',
+ 'BT': '119.2.96.0/19',
+ 'BW': '168.167.0.0/16',
+ 'BY': '178.120.0.0/13',
+ 'BZ': '179.42.192.0/18',
+ 'CA': '99.224.0.0/11',
+ 'CD': '41.243.0.0/16',
+ 'CF': '196.32.200.0/21',
+ 'CG': '197.214.128.0/17',
+ 'CH': '85.0.0.0/13',
+ 'CI': '154.232.0.0/14',
+ 'CK': '202.65.32.0/19',
+ 'CL': '152.172.0.0/14',
+ 'CM': '165.210.0.0/15',
+ 'CN': '36.128.0.0/10',
+ 'CO': '181.240.0.0/12',
+ 'CR': '201.192.0.0/12',
+ 'CU': '152.206.0.0/15',
+ 'CV': '165.90.96.0/19',
+ 'CW': '190.88.128.0/17',
+ 'CY': '46.198.0.0/15',
+ 'CZ': '88.100.0.0/14',
+ 'DE': '53.0.0.0/8',
+ 'DJ': '197.241.0.0/17',
+ 'DK': '87.48.0.0/12',
+ 'DM': '192.243.48.0/20',
+ 'DO': '152.166.0.0/15',
+ 'DZ': '41.96.0.0/12',
+ 'EC': '186.68.0.0/15',
+ 'EE': '90.190.0.0/15',
+ 'EG': '156.160.0.0/11',
+ 'ER': '196.200.96.0/20',
+ 'ES': '88.0.0.0/11',
+ 'ET': '196.188.0.0/14',
+ 'EU': '2.16.0.0/13',
+ 'FI': '91.152.0.0/13',
+ 'FJ': '144.120.0.0/16',
+ 'FM': '119.252.112.0/20',
+ 'FO': '88.85.32.0/19',
+ 'FR': '90.0.0.0/9',
+ 'GA': '41.158.0.0/15',
+ 'GB': '25.0.0.0/8',
+ 'GD': '74.122.88.0/21',
+ 'GE': '31.146.0.0/16',
+ 'GF': '161.22.64.0/18',
+ 'GG': '62.68.160.0/19',
+ 'GH': '45.208.0.0/14',
+ 'GI': '85.115.128.0/19',
+ 'GL': '88.83.0.0/19',
+ 'GM': '160.182.0.0/15',
+ 'GN': '197.149.192.0/18',
+ 'GP': '104.250.0.0/19',
+ 'GQ': '105.235.224.0/20',
+ 'GR': '94.64.0.0/13',
+ 'GT': '168.234.0.0/16',
+ 'GU': '168.123.0.0/16',
+ 'GW': '197.214.80.0/20',
+ 'GY': '181.41.64.0/18',
+ 'HK': '113.252.0.0/14',
+ 'HN': '181.210.0.0/16',
+ 'HR': '93.136.0.0/13',
+ 'HT': '148.102.128.0/17',
+ 'HU': '84.0.0.0/14',
+ 'ID': '39.192.0.0/10',
+ 'IE': '87.32.0.0/12',
+ 'IL': '79.176.0.0/13',
+ 'IM': '5.62.80.0/20',
+ 'IN': '117.192.0.0/10',
+ 'IO': '203.83.48.0/21',
+ 'IQ': '37.236.0.0/14',
+ 'IR': '2.176.0.0/12',
+ 'IS': '82.221.0.0/16',
+ 'IT': '79.0.0.0/10',
+ 'JE': '87.244.64.0/18',
+ 'JM': '72.27.0.0/17',
+ 'JO': '176.29.0.0/16',
+ 'JP': '126.0.0.0/8',
+ 'KE': '105.48.0.0/12',
+ 'KG': '158.181.128.0/17',
+ 'KH': '36.37.128.0/17',
+ 'KI': '103.25.140.0/22',
+ 'KM': '197.255.224.0/20',
+ 'KN': '198.32.32.0/19',
+ 'KP': '175.45.176.0/22',
+ 'KR': '175.192.0.0/10',
+ 'KW': '37.36.0.0/14',
+ 'KY': '64.96.0.0/15',
+ 'KZ': '2.72.0.0/13',
+ 'LA': '115.84.64.0/18',
+ 'LB': '178.135.0.0/16',
+ 'LC': '192.147.231.0/24',
+ 'LI': '82.117.0.0/19',
+ 'LK': '112.134.0.0/15',
+ 'LR': '41.86.0.0/19',
+ 'LS': '129.232.0.0/17',
+ 'LT': '78.56.0.0/13',
+ 'LU': '188.42.0.0/16',
+ 'LV': '46.109.0.0/16',
+ 'LY': '41.252.0.0/14',
+ 'MA': '105.128.0.0/11',
+ 'MC': '88.209.64.0/18',
+ 'MD': '37.246.0.0/16',
+ 'ME': '178.175.0.0/17',
+ 'MF': '74.112.232.0/21',
+ 'MG': '154.126.0.0/17',
+ 'MH': '117.103.88.0/21',
+ 'MK': '77.28.0.0/15',
+ 'ML': '154.118.128.0/18',
+ 'MM': '37.111.0.0/17',
+ 'MN': '49.0.128.0/17',
+ 'MO': '60.246.0.0/16',
+ 'MP': '202.88.64.0/20',
+ 'MQ': '109.203.224.0/19',
+ 'MR': '41.188.64.0/18',
+ 'MS': '208.90.112.0/22',
+ 'MT': '46.11.0.0/16',
+ 'MU': '105.16.0.0/12',
+ 'MV': '27.114.128.0/18',
+ 'MW': '105.234.0.0/16',
+ 'MX': '187.192.0.0/11',
+ 'MY': '175.136.0.0/13',
+ 'MZ': '197.218.0.0/15',
+ 'NA': '41.182.0.0/16',
+ 'NC': '101.101.0.0/18',
+ 'NE': '197.214.0.0/18',
+ 'NF': '203.17.240.0/22',
+ 'NG': '105.112.0.0/12',
+ 'NI': '186.76.0.0/15',
+ 'NL': '145.96.0.0/11',
+ 'NO': '84.208.0.0/13',
+ 'NP': '36.252.0.0/15',
+ 'NR': '203.98.224.0/19',
+ 'NU': '49.156.48.0/22',
+ 'NZ': '49.224.0.0/14',
+ 'OM': '5.36.0.0/15',
+ 'PA': '186.72.0.0/15',
+ 'PE': '186.160.0.0/14',
+ 'PF': '123.50.64.0/18',
+ 'PG': '124.240.192.0/19',
+ 'PH': '49.144.0.0/13',
+ 'PK': '39.32.0.0/11',
+ 'PL': '83.0.0.0/11',
+ 'PM': '70.36.0.0/20',
+ 'PR': '66.50.0.0/16',
+ 'PS': '188.161.0.0/16',
+ 'PT': '85.240.0.0/13',
+ 'PW': '202.124.224.0/20',
+ 'PY': '181.120.0.0/14',
+ 'QA': '37.210.0.0/15',
+ 'RE': '139.26.0.0/16',
+ 'RO': '79.112.0.0/13',
+ 'RS': '178.220.0.0/14',
+ 'RU': '5.136.0.0/13',
+ 'RW': '105.178.0.0/15',
+ 'SA': '188.48.0.0/13',
+ 'SB': '202.1.160.0/19',
+ 'SC': '154.192.0.0/11',
+ 'SD': '154.96.0.0/13',
+ 'SE': '78.64.0.0/12',
+ 'SG': '152.56.0.0/14',
+ 'SI': '188.196.0.0/14',
+ 'SK': '78.98.0.0/15',
+ 'SL': '197.215.0.0/17',
+ 'SM': '89.186.32.0/19',
+ 'SN': '41.82.0.0/15',
+ 'SO': '197.220.64.0/19',
+ 'SR': '186.179.128.0/17',
+ 'SS': '105.235.208.0/21',
+ 'ST': '197.159.160.0/19',
+ 'SV': '168.243.0.0/16',
+ 'SX': '190.102.0.0/20',
+ 'SY': '5.0.0.0/16',
+ 'SZ': '41.84.224.0/19',
+ 'TC': '65.255.48.0/20',
+ 'TD': '154.68.128.0/19',
+ 'TG': '196.168.0.0/14',
+ 'TH': '171.96.0.0/13',
+ 'TJ': '85.9.128.0/18',
+ 'TK': '27.96.24.0/21',
+ 'TL': '180.189.160.0/20',
+ 'TM': '95.85.96.0/19',
+ 'TN': '197.0.0.0/11',
+ 'TO': '175.176.144.0/21',
+ 'TR': '78.160.0.0/11',
+ 'TT': '186.44.0.0/15',
+ 'TV': '202.2.96.0/19',
+ 'TW': '120.96.0.0/11',
+ 'TZ': '156.156.0.0/14',
+ 'UA': '93.72.0.0/13',
+ 'UG': '154.224.0.0/13',
+ 'US': '3.0.0.0/8',
+ 'UY': '167.56.0.0/13',
+ 'UZ': '82.215.64.0/18',
+ 'VA': '212.77.0.0/19',
+ 'VC': '24.92.144.0/20',
+ 'VE': '186.88.0.0/13',
+ 'VG': '172.103.64.0/18',
+ 'VI': '146.226.0.0/16',
+ 'VN': '14.160.0.0/11',
+ 'VU': '202.80.32.0/20',
+ 'WF': '117.20.32.0/21',
+ 'WS': '202.4.32.0/19',
+ 'YE': '134.35.0.0/16',
+ 'YT': '41.242.116.0/22',
+ 'ZA': '41.0.0.0/11',
+ 'ZM': '165.56.0.0/13',
+ 'ZW': '41.85.192.0/19',
+ }
+
+ @classmethod
+ def random_ipv4(cls, code):
+ block = cls._country_ip_map.get(code.upper())
+ if not block:
+ return None
+ addr, preflen = block.split('/')
+ addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
+ addr_max = addr_min | (0xffffffff >> int(preflen))
+ return compat_str(socket.inet_ntoa(
+ compat_struct_pack('!L', random.randint(addr_min, addr_max))))
+
+
class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
def __init__(self, proxies=None):
# Set default handlers
@@ -3012,6 +3517,57 @@ class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
self, req, proxy, type)
+# Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is
+# released into Public Domain
+# https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387
+
+def long_to_bytes(n, blocksize=0):
+ """long_to_bytes(n:long, blocksize:int) : string
+ Convert a long integer to a byte string.
+
+ If optional blocksize is given and greater than zero, pad the front of the
+ byte string with binary zeros so that the length is a multiple of
+ blocksize.
+ """
+ # after much testing, this algorithm was deemed to be the fastest
+ s = b''
+ n = int(n)
+ while n > 0:
+ s = compat_struct_pack('>I', n & 0xffffffff) + s
+ n = n >> 32
+ # strip off leading zeros
+ for i in range(len(s)):
+ if s[i] != b'\000'[0]:
+ break
+ else:
+ # only happens when n == 0
+ s = b'\000'
+ i = 0
+ s = s[i:]
+ # add back some pad bytes. this could be done more efficiently w.r.t. the
+ # de-padding being done above, but sigh...
+ if blocksize > 0 and len(s) % blocksize:
+ s = (blocksize - len(s) % blocksize) * b'\000' + s
+ return s
+
+
+def bytes_to_long(s):
+ """bytes_to_long(string) : long
+ Convert a byte string to a long integer.
+
+ This is (essentially) the inverse of long_to_bytes().
+ """
+ acc = 0
+ length = len(s)
+ if length % 4:
+ extra = (4 - length % 4)
+ s = b'\000' * extra + s
+ length = length + extra
+ for i in range(0, length, 4):
+ acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0]
+ return acc
+
+
def ohdave_rsa_encrypt(data, exponent, modulus):
'''
Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/
@@ -3029,6 +3585,21 @@ def ohdave_rsa_encrypt(data, exponent, modulus):
return '%x' % encrypted
+def pkcs1pad(data, length):
+ """
+ Padding input data with PKCS#1 scheme
+
+ @param {int[]} data input data
+ @param {int} length target length
+ @returns {int[]} padded data
+ """
+ if len(data) > length - 11:
+ raise ValueError('Input data too long for PKCS#1 padding')
+
+ pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)]
+ return [0, 2] + pseudo_random + [0] + data
+
+
def encode_base_n(num, n, table=None):
FULL_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
if not table:
@@ -3267,3 +3838,11 @@ def write_xattr(path, key, value):
"Couldn't find a tool to set the xattrs. "
"Install either the python 'xattr' module, "
"or the 'xattr' binary.")
+
+
+def random_birthday(year_field, month_field, day_field):
+ return {
+ year_field: str(random.randint(1950, 1995)),
+ month_field: str(random.randint(1, 12)),
+ day_field: str(random.randint(1, 31)),
+ }