-#! /usr/local/bin/python -O
+#! /usr/bin/env python
"""
HTML Parser
- Written by BroytMann, Jun 2002 - May 2003. Copyright (C) 1997-2003 PhiloSoft Design
+ Written by BroytMann. Copyright (C) 1997-2007 PhiloSoft Design
"""
import sys
current_charset = sys.getdefaultencoding()
-DEFAULT_CHARSET = "windows-1251" # Stupid default for Russian Cyrillic
+if current_charset == "ascii":
+ try:
+ import locale
+ except ImportError:
+ pass
+ else:
+ current_charset = locale.getpreferredencoding()
+current_charset = current_charset.replace("windows-", "cp").lower()
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
from HTMLParser import HTMLParseError
-from m_lib.www.html import HTMLParser as _HTMLParser
+from m_lib.net.www.html import HTMLParser as _HTMLParser
class HTMLHeadDone(Exception): pass
self.title = self.accumulator
+ def do_link(self, attrs):
+ has_icon = False
+ href = None
+
+ for attrname, value in attrs:
+ if value:
+ value = value.strip().lower()
+ if (attrname == 'rel') and (value in ('icon', 'shortcut icon')):
+ has_icon = True
+ elif attrname == 'href':
+ href = value
+
+ if has_icon:
+ self.icon = href
+ else:
+ self.icon = None
+
+
import re
entity_re = re.compile("(&#[0-9]+;)")
title = parser.title
if not parser.charset:
- ascii = 1
- for c in title:
- if not (32 <= ord(c) <= 127): # non-ASCII character
- ascii = 0
- break
- if not ascii:
+ try:
+ unicode(title, "ascii")
+ except UnicodeDecodeError:
parser.charset = DEFAULT_CHARSET
+ if parser.charset:
+ parser.charset = parser.charset.replace("windows-", "cp").lower()
+
if parser.charset and (parser.charset <> current_charset):
try:
if parser.meta_charset:
print parser.charset
print parser.title
print parser.refresh
+ print parser.icon