Update changelog.

[youtubedl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 616948e1080d18352189a48c4967644f4e761b20..5dd5b2923d2a773a526006d71769d10486fe8730 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -12,7 +12,7 @@ import sys
 import traceback
 import zlib
 import email.utils
-import json
+import socket
 import datetime
 
 try:
@@ -35,6 +35,11 @@ try:
 except ImportError: # Python 2
     from urlparse import urlparse as compat_urllib_parse_urlparse
 
+try:
+    import urllib.parse as compat_urlparse
+except ImportError: # Python 2
+    import urlparse as compat_urlparse
+
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError: # Python 2
@@ -150,6 +155,13 @@ try:
 except NameError:
     compat_chr = chr
 
+def compat_ord(c):
+    if type(c) is int: return c
+    else: return ord(c)
+
+# This is not clearly defined otherwise
+compiled_regex_type = type(re.compile(''))
+
 std_headers = {
     'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0',
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
@@ -191,6 +203,20 @@ else:
         with open(fn, 'w', encoding='utf-8') as f:
             json.dump(obj, f)
 
+if sys.version_info >= (2,7):
+    def find_xpath_attr(node, xpath, key, val):
+        """ Find the xpath xpath[@key=val] """
+        assert re.match(r'^[a-zA-Z]+$', key)
+        assert re.match(r'^[a-zA-Z@\s]*$', val)
+        expr = xpath + u"[@%s='%s']" % (key, val)
+        return node.find(expr)
+else:
+    def find_xpath_attr(node, xpath, key, val):
+        for f in node.findall(xpath):
+            if f.attrib.get(key) == val:
+                return f
+        return None
+
 def htmlentity_transform(matchobj):
     """Transforms an HTML entity to a character.
 
@@ -463,9 +489,17 @@ def make_HTTPS_handler(opts):
 
 class ExtractorError(Exception):
     """Error during info extraction."""
-    def __init__(self, msg, tb=None):
-        """ tb, if given, is the original traceback (so that it can be printed out). """
+    def __init__(self, msg, tb=None, expected=False):
+        """ tb, if given, is the original traceback (so that it can be printed out).
+        If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
+        """
+
+        if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
+            expected = True
+        if not expected:
+            msg = msg + u'; please report this issue on https://yt-dl.org/bug . Be sure to call youtube-dl with the --verbose flag and include its complete output. Make sure you are using the latest version; type  youtube-dl -U  to update.'
         super(ExtractorError, self).__init__(msg)
+
         self.traceback = tb
         self.exc_info = sys.exc_info()  # preserve original exception
 
@@ -608,7 +642,7 @@ def unified_strdate(date_str):
     date_str = date_str.replace(',',' ')
     # %z (UTC offset) is only supported in python>=3.2
     date_str = re.sub(r' (\+|-)[\d]*$', '', date_str)
-    format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y', '%Y-%m-%d', '%d/%m/%Y', '%Y/%m/%d %H:%M:%S']
+    format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y', '%Y-%m-%d', '%d/%m/%Y', '%Y/%m/%d %H:%M:%S', '%d.%m.%Y %H:%M']
     for expression in format_expressions:
         try:
             upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
@@ -616,6 +650,13 @@ def unified_strdate(date_str):
             pass
     return upload_date
 
+def determine_ext(url, default_ext=u'unknown_video'):
+    guess = url.partition(u'?')[0].rpartition(u'.')[2]
+    if re.match(r'^[A-Za-z0-9]+$', guess):
+        return guess
+    else:
+        return default_ext
+
 def date_from_str(date_str):
     """
     Return a datetime object from a string in the format YYYYMMDD or