Proper support for changing User-Agents from IEs

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 25b67db0665af756ee32dac797f18a647004ff1f..08be9e6373f6d4129327a6125b7d3b8c863273c0 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -3,10 +3,12 @@
 
 import gzip
 import io
+import json
 import locale
 import os
 import re
 import sys
+import traceback
 import zlib
 import email.utils
 import json
@@ -153,6 +155,7 @@ std_headers = {
     'Accept-Encoding': 'gzip, deflate',
     'Accept-Language': 'en-us,en;q=0.5',
 }
+
 def preferredencoding():
     """Get preferred encoding.
 
@@ -175,6 +178,17 @@ else:
         assert type(s) == type(u'')
         print(s)
 
+# In Python 2.x, json.dump expects a bytestream.
+# In Python 3.x, it writes to a character stream
+if sys.version_info < (3,0):
+    def write_json_file(obj, fn):
+        with open(fn, 'wb') as f:
+            json.dump(obj, f)
+else:
+    def write_json_file(obj, fn):
+        with open(fn, 'w', encoding='utf-8') as f:
+            json.dump(obj, f)
+
 def htmlentity_transform(matchobj):
     """Transforms an HTML entity to a character.
 
@@ -201,10 +215,11 @@ def htmlentity_transform(matchobj):
     return (u'&%s;' % entity)
 
 compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
-class IDParser(compat_html_parser.HTMLParser):
-    """Modified HTMLParser that isolates a tag with the specified id"""
-    def __init__(self, id):
-        self.id = id
+class AttrParser(compat_html_parser.HTMLParser):
+    """Modified HTMLParser that isolates a tag with the specified attribute"""
+    def __init__(self, attribute, value):
+        self.attribute = attribute
+        self.value = value
         self.result = None
         self.started = False
         self.depth = {}
@@ -229,7 +244,7 @@ class IDParser(compat_html_parser.HTMLParser):
         attrs = dict(attrs)
         if self.started:
             self.find_startpos(None)
-        if 'id' in attrs and attrs['id'] == self.id:
+        if self.attribute in attrs and attrs[self.attribute] == self.value:
             self.result = [tag]
             self.started = True
             self.watch_startpos = True
@@ -267,8 +282,12 @@ class IDParser(compat_html_parser.HTMLParser):
         return '\n'.join(lines).strip()
 
 def get_element_by_id(id, html):
-    """Return the content of the tag with the specified id in the passed HTML document"""
-    parser = IDParser(id)
+    """Return the content of the tag with the specified ID in the passed HTML document"""
+    return get_element_by_attribute("id", id, html)
+
+def get_element_by_attribute(attribute, value, html):
+    """Return the content of the tag with the specified attribute in the passed HTML document"""
+    parser = AttrParser(attribute, value)
     try:
         parser.loads(html)
     except compat_html_parser.HTMLParseError:
@@ -280,7 +299,8 @@ def clean_html(html):
     """Clean an HTML snippet into a readable string"""
     # Newline vs <br />
     html = html.replace('\n', ' ')
-    html = re.sub('\s*<\s*br\s*/?\s*>\s*', '\n', html)
+    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)
     # Strip html tags
     html = re.sub('<.*?>', '', html)
     # Replace html entities
@@ -391,6 +411,20 @@ def encodeFilename(s):
     else:
         return s.encode(sys.getfilesystemencoding(), 'ignore')
 
+
+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). """
+        super(ExtractorError, self).__init__(msg)
+        self.traceback = tb
+
+    def format_traceback(self):
+        if self.traceback is None:
+            return None
+        return u''.join(traceback.format_tb(self.traceback))
+
+
 class DownloadError(Exception):
     """Download Error exception.
 
@@ -416,7 +450,8 @@ class PostProcessingError(Exception):
     This exception may be raised by PostProcessor's .run() method to
     indicate an error in the postprocessing task.
     """
-    pass
+    def __init__(self, msg):
+        self.msg = msg
 
 class MaxDownloadsReached(Exception):
     """ --max-downloads limit has been reached. """
@@ -447,14 +482,6 @@ class ContentTooShortError(Exception):
         self.downloaded = downloaded
         self.expected = expected
 
-
-class Trouble(Exception):
-    """Trouble helper exception
-
-    This is an exception to be handled with
-    FileDownloader.trouble
-    """
-
 class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
     """Handler for HTTP requests and responses.
 
@@ -497,6 +524,11 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
             if 'Accept-encoding' in req.headers:
                 del req.headers['Accept-encoding']
             del req.headers['Youtubedl-no-compression']
+        if 'Youtubedl-user-agent' in req.headers:
+            if 'User-Agent' in req.headers:
+                del req.headers['User-Agent']
+            req.headers['User-Agent'] = req.headers['Youtubedl-user-agent']
+            del req.headers['Youtubedl-user-agent']
         return req
 
     def http_response(self, req, resp):