"""HTML Parsers
This file is a part of Bookmarks database and Internet robot.
+
"""
-__version__ = "$Revision$"[11:-2]
-__revision__ = "$Id$"[5:-2]
-__date__ = "$Date$"[7:-2]
__author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 1997-2011 PhiloSoft Design"
+__copyright__ = "Copyright (C) 1997-2017 PhiloSoft Design"
__license__ = "GNU GPL"
-__all__ = ['parse_html', 'universal_charset']
+__all__ = ['parse_html', 'parse_filename', 'universal_charset']
import codecs
parsers = []
-# Statistics by parser - successfully parsed HTML pages:
-# 4358 beautifulsoup
-# 4310 htmlparser
-# 4307 html5
-# 4250 lxml
-
try:
from . import bkmk_ph_beautifulsoup
except ImportError:
parsers.append(bkmk_ph_beautifulsoup.parse_html)
try:
- from . import bkmk_ph_htmlparser
+ from . import bkmk_ph_html5
except ImportError:
pass
else:
- parsers.append(bkmk_ph_htmlparser.parse_html)
+ parsers.append(bkmk_ph_html5.parse_html)
try:
- from . import bkmk_ph_html5
+ from . import bkmk_ph_lxml
except ImportError:
pass
else:
- parsers.append(bkmk_ph_html5.parse_html)
+ parsers.append(bkmk_ph_lxml.parse_html)
try:
- from . import bkmk_ph_lxml
+ from . import bkmk_ph_htmlparser
except ImportError:
pass
else:
- parsers.append(bkmk_ph_lxml.parse_html)
+ parsers.append(bkmk_ph_htmlparser.parse_html)
# ElementTidy often segfaults
#try:
return ''.join(output)
-def parse_html(filename, charset=None, log=None):
+import os
+BKMK_DEBUG_HTML_PARSERS = os.environ.get("BKMK_DEBUG_HTML_PARSERS")
+
+def parse_html(html_text, charset=None, log=None):
if not parsers:
return None
charsets.remove(charset)
charsets.insert(0, charset)
- #_parsers = []
+ if BKMK_DEBUG_HTML_PARSERS:
+ _parsers = []
for p in parsers:
parser = None
for c in charsets:
try:
- parser = p(filename, c, log)
+ parser = p(html_text, c, log)
except UnicodeError:
pass
else:
if parser:
- if log: log(" Parser %s: ok" % p.__module__)
- #_parsers.append(parser)
+ if BKMK_DEBUG_HTML_PARSERS:
+ if log: log(" Parser %s: ok" % p.__module__)
+ _parsers.append((p, parser))
break
else:
if log: log(" Parser %s: fail" % p.__module__)
- if parser:
+ if not BKMK_DEBUG_HTML_PARSERS and parser:
break
- #if not _parsers:
- if not parser:
- if log: log(" All parser has failed")
+ if BKMK_DEBUG_HTML_PARSERS:
+ if not _parsers:
+ if log: log(" All parsers have failed")
+ return None
+ elif not parser:
+ if log: log(" All parsers have failed")
return None
- #parser = _parsers[0]
- if log: log(" Using %s" % parser.__module__)
+ if BKMK_DEBUG_HTML_PARSERS:
+ p, parser = _parsers[0]
+ if log: log(" Using %s" % p.__module__)
converted_title = title = parser.title
if title and (not parser.charset):
parser.charset = parser.charset.lower().replace("windows-", "cp")
if title and parser.charset and (
- (parser.charset <> universal_charset) or
- ((not charset) or (charset <> parser.charset))):
+ (parser.charset != universal_charset) or
+ ((not charset) or (charset != parser.charset))):
try:
if parser.meta_charset:
if log: log(" META charset : %s" % parser.charset)
- elif (not charset) or (charset <> parser.charset):
+ elif (not charset) or (charset != parser.charset):
if log: log(" guessed charset: %s" % parser.charset)
#if log: log(" current charset: %s" % universal_charset)
if log: log(" title : %s" % title)
- if parser.charset <> universal_charset:
+ if parser.charset != universal_charset:
try:
converted_title = unicode(title, parser.charset).encode(universal_charset)
except UnicodeError:
if log: log(" incorrect conversion from %s, converting from %s" % (parser.charset, DEFAULT_CHARSET))
converted_title = unicode(title, DEFAULT_CHARSET, "replace").encode(universal_charset, "replace")
parser.charset = DEFAULT_CHARSET
- if log and (converted_title <> title): log(" converted title: %s" % converted_title)
+ if log and (converted_title != title): log(" converted title: %s" % converted_title)
except LookupError:
if log: log(" unknown charset: '%s'" % parser.charset)
else:
final_title = recode_entities(converted_title, universal_charset)
parts = [s.strip() for s in final_title.replace('\r', '').split('\n')]
final_title = ' '.join([s for s in parts if s])
- if log and (final_title <> converted_title): log(" final title : %s" % final_title)
+ if log and (final_title != converted_title): log(" final title : %s" % final_title)
parser.title = final_title
icon = parser.icon
if parser.charset:
parser.icon = icon.encode(parser.charset)
return parser
+
+def parse_filename(filename, charset=None, log=None):
+ fp = open(filename, 'r')
+ try:
+ parser = parse_html(fp.read(), charset=charset, log=log)
+ finally:
+ fp.close()
+ return parser