"""
HTML Parser using BeautifulSoup
- Written by BroytMann. Copyright (C) 2007 PhiloSoft Design
+ Written by BroytMann. Copyright (C) 2007, 2008 PhiloSoft Design
"""
+import re
+from sgmllib import SGMLParser, SGMLParseError
from HTMLParser import HTMLParser
-from BeautifulSoup import BeautifulSoup
+from BeautifulSoup import BeautifulSoup, CData
class BSoupParser(HTMLParser):
self.icon = icon
-def parse_html(filename, charset=None):
+# http://groups.google.com/group/beautifulsoup/browse_thread/thread/69093cb0d3a3cf63
+class BadDeclParser(BeautifulSoup):
+ def parse_declaration(self, i):
+ """Treat a bogus SGML declaration as raw data. Treat a CDATA
+ declaration as a CData object."""
+ j = None
+ if self.rawdata[i:i+9] == '<![CDATA[':
+ k = self.rawdata.find(']]>', i)
+ if k == -1:
+ k = len(self.rawdata)
+ data = self.rawdata[i+9:k]
+ j = k+3
+ self._toStringSubclass(data, CData)
+ else:
+ try:
+ j = SGMLParser.parse_declaration(self, i)
+ except SGMLParseError:
+ # Could not parse the DOCTYPE declaration
+ # Try to just skip the actual declaration
+ match = re.search(r'<!DOCTYPE([^>]*?)>', self.rawdata[i:], re.MULTILINE)
+ if match:
+ toHandle = self.rawdata[i:match.end()]
+ else:
+ toHandle = self.rawdata[i:]
+ self.handle_data(toHandle)
+ j = i + len(toHandle)
+ return j
+
+
+def parse_html(filename, charset=None, log=None):
infile = open(filename, 'r')
- root = BeautifulSoup(infile, fromEncoding=charset)
- infile.close()
+ try:
+ root = BadDeclParser(infile, fromEncoding=charset)
+ except TypeError:
+ if log: log("TypeError")
+ return None
+ finally:
+ infile.close()
- _charset = root.originalEncoding
try:
- title = root.html.head.title.string.encode(_charset)
+ head = root.html.head
except AttributeError:
- title = ''
+ if log: log("No HTML in root or no HEAD in HTML")
+ return None
+
+ if head is None:
+ head = root.html # Some sites put TITLE in HTML without HEAD
+
+ _charset = root.originalEncoding
+ if _charset == "windows-1252": # Replace default
+ _charset = DEFAULT_CHARSET
try:
- meta = root.html.head.find(_find_refresh, recursive=False)
+ title = head.title.string.encode(_charset)
except AttributeError:
- refresh = None
- else:
- if meta:
- refresh = meta.get("content")
+ title = '' # HEAD but no TITLE
+
+ if (not title) and (head is not root.html):
+ # Some sites put TITLE in HTML outside of HEAD
+
+ try:
+ title = root.html.title.string.encode(_charset)
+ except AttributeError:
+ title = '' # no TITLE in HTML too
+
+ meta = head.find(_find_contenttype, recursive=False)
+ if meta:
+ try:
+ __charset = meta.get("content").lower().split('charset=')[1].split(';')[0]
+ except IndexError: # No charset in the META Content-Type
+ meta_charset = False
else:
- refresh = None
+ meta_charset = _charset == __charset
+ else:
+ meta_charset = False
- try:
- meta = root.html.head.find(_find_icon, recursive=False)
- except AttributeError:
- icon = None
+ meta = head.find(_find_refresh, recursive=False)
+ if meta:
+ refresh = meta.get("content")
else:
- if meta:
- icon = meta.get("href")
- else:
- icon = None
+ refresh = None
- return BSoupParser(_charset, _charset == charset, title, refresh, icon)
+ meta = head.find(_find_icon, recursive=False)
+ if meta:
+ icon = meta.get("href")
+ else:
+ icon = None
+
+ return BSoupParser(_charset, meta_charset, title, refresh, icon)
+
+def _find_contenttype(Tag):
+ return (Tag.name == "meta") and \
+ (Tag.get("http-equiv", '').lower() == "content-type")
def _find_refresh(Tag):
return (Tag.name == "meta") and \