X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html.py;h=3e20a5516a775eb8e425bdf932831bbcb559641f;hb=efa27458378a9a253f0002fd060e2f8a2ae5a25f;hp=e95dcef24e0e2f7dc76c8e3f2908b466e17e83b9;hpb=fb5c3b2b91aeeb615d6d6d890491af3fdff69556;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index e95dcef..3e20a55 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -1,111 +1,75 @@
-#! /usr/local/bin/python -O
+#! /usr/bin/env python
 """
-   HTML Parser
+   HTML Parsers wrapper
 
-   Written by BroytMann, Jun 2002 - Aug 2002. Copyright (C) 1997-2002 PhiloSoft Design
+   Written by BroytMann. Copyright (C) 1997-2007 PhiloSoft Design
 """
 
+import codecs
 
-import sys
-current_charset = sys.getdefaultencoding()
-DEFAULT_CHARSET = "windows-1251"
+from m_lib.defenc import default_encoding
+current_charset = default_encoding.replace("windows-", "cp")
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
 
+try:
+   from parse_html_beautifulsoup import parse_html as _parse_html
+except ImportError:
+   from parse_html_htmlparser import parse_html as _parse_html
 
-from HTMLParser import HTMLParseError
-from m_lib.www.html import HTMLParser as _HTMLParser
 
+import re
+entity_re = re.compile("(&#[0-9]+;)")
 
-class HTMLHeadDone(Exception): pass
+def recode_entities(title, charset):
+   output = []
+   for part in entity_re.split(title):
+      if entity_re.match(part):
+         part = unichr(int(part[2:-1])).encode(charset, "replace")
+      output.append(part)
 
-
-class HTMLParser(_HTMLParser):
-   def __init__(self, charset=None):
-      _HTMLParser.__init__(self)
-      self.charset = charset
-      self.meta_charset = 0
-      self.title = ''
-      self.refresh = ''
-
-   def end_head(self):
-      raise HTMLHeadDone()
-
-
-   def do_meta(self, attrs):
-      http_equiv = ""
-      content = ""
-
-      for attrname, value in attrs:
-         if value:
-            value = value.strip()
-            if attrname == 'http-equiv':
-               http_equiv = value.lower()
-            elif attrname == 'content':
-               content = value
-
-      if (not self.charset) and (http_equiv == "content-type"):
-         try:
-            # extract charset from "text/html; foo; charset=UTF-8; bar;"
-            self.charset = content.lower().split('charset=')[1].split(';')[0]
-            self.meta_charset = 1
-         except IndexError:
-            pass
-
-      if http_equiv == "refresh":
-         self.refresh = content
-
-
-   def start_title(self, attrs):
-      self.accumulator = ''
-   def end_title(self):
-      if not self.title: # use only the first title
-         self.title = self.accumulator
+   return ''.join(output)
 
 
 def parse_html(filename, charset=None, log=None):
-   infile = open(filename, 'r')
-   parser = HTMLParser(charset)
-
-   for line in infile:
+   if charset:
       try:
-         parser.feed(line)
-      except (HTMLParseError, HTMLHeadDone):
-         break
-
-   infile.close()
+         codecs.lookup(charset) # In case of unknown charset...
+      except (ValueError, LookupError):
+         charset = None         # ...try charset from HTML
 
-   try:
-      parser.close()
-   except (HTMLParseError, HTMLHeadDone):
-      pass
+   parser = _parse_html(filename, charset)
+   title = parser.title
 
    if not parser.charset:
-      title = parser.title
-      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:
             if log: log("   META charset   : %s" % parser.charset)
          else:
             if log: log("   charset        : %s" % parser.charset)
-         if log: log("   title          : %s" % parser.title)
-         parser.title = unicode(parser.title, parser.charset, "replace").encode(current_charset, "replace")
+         if log: log("   title          : %s" % title)
+         title = unicode(title, parser.charset, "replace").encode(current_charset, "replace")
          if log: log("   current charset: %s" % current_charset)
-         if log: log("   converted title: %s" % parser.title)
+         if log: log("   converted title: %s" % title)
       except LookupError:
          if log: log("   unknown charset: `%s' or `%s'" % (parser.charset, current_charset))
 
+   parser.title = recode_entities(title, current_charset)
    return parser
 
 
 if __name__ == '__main__':
    import sys
-   parser = parse_html(sys.argv[1])
+   parser = parse_html(sys.argv[1], current_charset)
    print parser.charset
    print parser.title
    print parser.refresh
+   print parser.icon