[utils] Add support for zero years and months in parse_duration

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 16bf49408c5228202cd43d9f7265d77c1c5122dc..34866a54b6efc122f4d0edb22712c503fa448ec0 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -22,7 +22,6 @@ import locale
 import math
 import operator
 import os
-import pipes
 import platform
 import random
 import re
@@ -36,6 +35,7 @@ import xml.etree.ElementTree
 import zlib
 
 from .compat import (
+    compat_HTMLParseError,
     compat_HTMLParser,
     compat_basestring,
     compat_chr,
@@ -365,9 +365,9 @@ def get_elements_by_attribute(attribute, value, html, escape_value=True):
     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>
@@ -409,8 +409,12 @@ 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
 
 
@@ -592,7 +596,7 @@ 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():
@@ -1530,7 +1534,7 @@ 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)
 
 
@@ -1811,6 +1815,10 @@ 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()
 
@@ -1827,10 +1835,20 @@ def parse_duration(s):
         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*
                 )?
@@ -1925,7 +1943,7 @@ 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
@@ -2564,14 +2582,18 @@ 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 = (
-        ('http://www.w3.org/ns/ttml', [
-            'http://www.w3.org/2004/11/ttaf1',
-            'http://www.w3.org/2006/04/ttaf1',
-            'http://www.w3.org/2006/10/ttaf1',
+        (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',
         ]),
-        ('http://www.w3.org/ns/ttml#styling', [
-            'http://www.w3.org/ns/ttml#style',
+        (b'http://www.w3.org/ns/ttml#styling', [
+            b'http://www.w3.org/ns/ttml#style',
         ]),
     )
 
@@ -2666,7 +2688,7 @@ def dfxp2srt(dfxp_data):
         for ns in v:
             dfxp_data = dfxp_data.replace(ns, k)
 
-    dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8'))
+    dfxp = compat_etree_fromstring(dfxp_data)
     out = []
     paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')
 
@@ -2729,6 +2751,8 @@ 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)]