"""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) 1997-2023 PhiloSoft Design"
__license__ = "GNU GPL"
-__all__ = ['parse_html', 'universal_charset']
+__all__ = ['parse_html', 'parse_filename', 'universal_charset']
import codecs
+import os
+import re
+try:
+ from html.entities import name2codepoint
+except ImportError:
+ from htmlentitydefs import name2codepoint
-universal_charset = "utf-8"
-DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+from compat import unicode, unichr
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
parsers = []
try:
- from . import bkmk_ph_beautifulsoup
+ from . import bkmk_ph_beautifulsoup
except ImportError:
- pass
+ pass
else:
- bkmk_ph_beautifulsoup.DEFAULT_CHARSET = DEFAULT_CHARSET
- parsers.append(bkmk_ph_beautifulsoup.parse_html)
+ bkmk_ph_beautifulsoup.DEFAULT_CHARSET = DEFAULT_CHARSET
+ parsers.append(bkmk_ph_beautifulsoup.parse_html)
try:
- from . import bkmk_ph_htmlparser
+ from . import bkmk_ph_beautifulsoup4
except ImportError:
- pass
+ pass
else:
- parsers.append(bkmk_ph_htmlparser.parse_html)
+ bkmk_ph_beautifulsoup4.DEFAULT_CHARSET = DEFAULT_CHARSET
+ parsers.append(bkmk_ph_beautifulsoup4.parse_html)
try:
- from . import bkmk_ph_html5
+ from . import bkmk_ph_htmlparser
except ImportError:
- pass
+ pass
else:
- parsers.append(bkmk_ph_html5.parse_html)
+ parsers.append(bkmk_ph_htmlparser.parse_html)
try:
- from . import bkmk_ph_lxml
+ from . import bkmk_ph_lxml
except ImportError:
- pass
+ pass
else:
parsers.append(bkmk_ph_lxml.parse_html)
-# ElementTidy often segfaults
-#try:
-# from . import bkmk_ph_etreetidy
-#except ImportError:
-# pass
-#else:
-# parsers.append(bkmk_ph_etreetidy.parse_html)
-
-import re
-from htmlentitydefs import name2codepoint
-
-entity_re = re.compile("(&\w+;)")
+universal_charset = "utf-8"
+entity_re = re.compile("(&\\w+;)")
num_entity_re = re.compile("(&#[0-9]+;)")
+
def recode_entities(title, charset):
- output = []
- for part in entity_re.split(title):
- if part not in ("&", "<", ">", """) and \
- entity_re.match(part):
- _part = name2codepoint.get(part[1:-1], None)
- if _part is not None:
- part = unichr(_part).encode(charset)
- output.append(part)
- title = ''.join(output)
-
- output = []
- for part in num_entity_re.split(title):
- if num_entity_re.match(part):
- try:
- part = unichr(int(part[2:-1])).encode(charset)
- except UnicodeEncodeError:
- pass # Leave the entity as is
- output.append(part)
-
- return ''.join(output)
+ output = []
+ for part in entity_re.split(title):
+ if part not in ("&", "<", ">", """) and \
+ entity_re.match(part):
+ _part = name2codepoint.get(part[1:-1], None)
+ if _part is not None:
+ part = unichr(_part)
+ output.append(part)
+ title = ''.join(output)
+
+ output = []
+ for part in num_entity_re.split(title):
+ if num_entity_re.match(part):
+ try:
+ part = unichr(int(part[2:-1]))
+ except UnicodeEncodeError:
+ pass # Leave the entity as is
+ output.append(part)
+
+ return ''.join(output)
-import os
BKMK_DEBUG_HTML_PARSERS = os.environ.get("BKMK_DEBUG_HTML_PARSERS")
-def parse_html(filename, charset=None, log=None):
- if not parsers:
- return None
-
- if charset:
- try:
- codecs.lookup(charset) # In case of unknown charset...
- except (ValueError, LookupError):
- charset = None # ...try charset from HTML
-
- charsets = [universal_charset, DEFAULT_CHARSET]
- if charset:
- charset = charset.lower().replace("windows-", "cp")
- if charset in charsets:
- charsets.remove(charset)
- charsets.insert(0, charset)
-
- if BKMK_DEBUG_HTML_PARSERS:
- _parsers = []
- for p in parsers:
- parser = None
- for c in charsets:
- try:
- parser = p(filename, c, log)
- except UnicodeError:
- pass
- else:
- if 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 not BKMK_DEBUG_HTML_PARSERS and parser:
- break
-
- 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
-
- 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):
- try:
- unicode(title, "ascii")
- except UnicodeDecodeError:
- parser.charset = DEFAULT_CHARSET
-
- if 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))):
- try:
- if parser.meta_charset:
- if log: log(" META charset : %s" % 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:
+
+def parse_html(html_text, charset=None, log=None):
+ if not html_text or not parsers:
+ return None
+
+ if charset:
+ try:
+ codecs.lookup(charset) # In case of unknown charset...
+ except (ValueError, LookupError):
+ charset = None # ...try charset from HTML
+
+ charsets = [universal_charset, DEFAULT_CHARSET]
+ if charset:
+ charset = charset.lower().replace("windows-", "cp")
+ if charset in charsets:
+ charsets.remove(charset)
+ charsets.insert(0, charset)
+
+ if BKMK_DEBUG_HTML_PARSERS:
+ _parsers = []
+ for p in parsers:
+ parser = None
+ for c in charsets:
try:
- converted_title = unicode(title, parser.charset).encode(universal_charset)
+ parser = p(html_text, c, log)
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)
- except LookupError:
- if log: log(" unknown charset: '%s'" % parser.charset)
- else:
- if log: log(" title : %s" % title)
-
- if title:
- 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)
- parser.title = final_title
-
- icon = parser.icon
- if isinstance(icon, unicode):
- try:
- parser.icon = icon.encode('ascii')
- except UnicodeEncodeError:
- if parser.charset:
- parser.icon = icon.encode(parser.charset)
- return parser
+ pass
+ else:
+ if 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 not BKMK_DEBUG_HTML_PARSERS and parser:
+ break
+
+ 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
+
+ if BKMK_DEBUG_HTML_PARSERS:
+ p, parser = _parsers[0]
+ if log: log(" Using %s" % p.__module__)
+
+ #title = parser.title
+ #if isinstance(title, unicode):
+ # if parser.charset:
+ # parser.title = title.encode(parser.charset)
+ # else:
+ # try:
+ # parser.title = title.encode('ascii')
+ # except UnicodeEncodeError:
+ # try:
+ # parser.title = title.encode(DEFAULT_CHARSET)
+ # except UnicodeEncodeError:
+ # parser.title = title.encode(universal_charset)
+ # parser.charset = universal_charset
+ # else:
+ # parser.charset = DEFAULT_CHARSET
+ # else:
+ # parser.charset = 'ascii'
+
+ converted_title = title = parser.title
+ if title and isinstance(title, bytes) and (not parser.charset):
+ try:
+ title.decode("ascii")
+ except UnicodeDecodeError:
+ parser.charset = DEFAULT_CHARSET
+
+ if 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))):
+ try:
+ if parser.meta_charset:
+ if log: log(" META charset : %s" % 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:
+ # try:
+ # converted_title = title.decode(parser.charset).\
+ # encode(universal_charset)
+ # except UnicodeError:
+ # if log:
+ # log(" incorrect conversion from %s,"
+ # "converting from %s"
+ # % (parser.charset, DEFAULT_CHARSET))
+ # converted_title = \
+ # title.decode(DEFAULT_CHARSET, "replace").\
+ # encode(universal_charset, "replace")
+ # parser.charset = DEFAULT_CHARSET
+ #if log and (converted_title != title):
+ # log(" converted title: %s" % converted_title)
+ except LookupError:
+ if log: log(" unknown charset: '%s'" % parser.charset)
+ else:
+ if log: log(" title : %s" % title)
+
+ if title:
+ 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)
+ parser.title = final_title
+
+ #icon = parser.icon
+ #if isinstance(icon, unicode):
+ # try:
+ # parser.icon = icon.encode('ascii')
+ # except UnicodeEncodeError:
+ # if parser.charset:
+ # parser.icon = icon.encode(parser.charset)
+ return parser
+
+
+def parse_filename(filename, charset=None, log=None):
+ fp = open(filename, 'rt')
+ try:
+ parser = parse_html(fp.read(), charset=charset, log=log)
+ finally:
+ fp.close()
+ return parser