X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=1279a9042f343eee1dbd8eaf713e3eed06d60d3e;hb=fcca0d53a8fa47614a39a433a3da7d1ab1d88ed9;hp=1a5ce8688e715bf84264d24f34acf8aeb71fdf2b;hpb=e5a088dc4be4fdcc96927a9f1b7284d4cd49c415;p=youtube-dl

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 1a5ce8688..1279a9042 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -86,6 +86,11 @@ std_headers = {
 }
 
 
+USER_AGENTS = {
+    'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27',
+}
+
+
 NO_DEFAULT = object()
 
 ENGLISH_MONTH_NAMES = [
@@ -123,7 +128,13 @@ 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',
@@ -132,6 +143,7 @@ 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',
@@ -325,17 +337,30 @@ 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+%s=['"]?%s['"]?
@@ -343,16 +368,15 @@ def get_element_by_attribute(attribute, value, html, escape_value=True):
         \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):
@@ -496,7 +520,7 @@ def sanitize_path(s):
     if drive_or_unc:
         norm_path.pop(0)
     sanitized_path = [
-        path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|[\s.]$)', '#', path_part)
+        path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part)
         for path_part in norm_path]
     if drive_or_unc:
         sanitized_path.insert(0, drive_or_unc + os.path.sep)
@@ -1178,7 +1202,7 @@ def date_from_str(date_str):
         return today
     if date_str == 'yesterday':
         return today - datetime.timedelta(days=1)
-    match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
+    match = re.match(r'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
     if match is not None:
         sign = match.group('sign')
         time = int(match.group('time'))
@@ -1660,6 +1684,11 @@ 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
@@ -1691,6 +1720,20 @@ def url_basename(url):
     return path.strip('/').split('/')[-1]
 
 
+def base_url(url):
+    return re.match(r'https?://[^?#&]+/', url).group()
+
+
+def urljoin(base, path):
+    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):
+        return None
+    return compat_urlparse.urljoin(base, path)
+
+
 class HEADRequest(compat_urllib_request.Request):
     def get_method(self):
         return 'HEAD'
@@ -1747,7 +1790,7 @@ 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:
@@ -1764,11 +1807,11 @@ 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:
@@ -2077,11 +2120,18 @@ def strip_jsonp(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 ("'", '"'):
@@ -2092,11 +2142,6 @@ 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:
@@ -2108,11 +2153,11 @@ 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):