X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html.py;h=c5c9eb6831e0db333cf95cd25314dbc453a5e28a;hb=4c8bc9dc1f441cf9256ce3a933f51213a0c3c3ff;hp=3250a0de5980c43a00c6cce8ff5e91a9e5b5471c;hpb=daa40d9fc61a15fe51955d97b5711af5973a64d3;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index 3250a0d..c5c9eb6 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -11,10 +11,16 @@ from m_lib.defenc import default_encoding
 current_charset = default_encoding.replace("windows-", "cp")
 DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
 
+parsers = []
 try:
-   from parse_html_beautifulsoup import parse_html as _parse_html
+   from parse_html_beautifulsoup import parse_html
 except ImportError:
-   from parse_html_htmlparser import parse_html as _parse_html
+   pass
+else:
+   parsers.append(parse_html)
+
+from parse_html_htmlparser import parse_html
+parsers.append(parse_html)
 
 
 import re
@@ -37,9 +43,14 @@ def parse_html(filename, charset=None, log=None):
       except (ValueError, LookupError):
          charset = None         # ...try charset from HTML
 
-   parser = _parse_html(filename, charset)
-   title = parser.title
+   for p in parsers:
+      parser = p(filename, charset)
+      if parser:
+         break
+      else:
+         if log: log("Parser %s.%s failed, trying next one." % (p.__module__, p.__name__))
 
+   title = parser.title
    if not parser.charset:
       try:
          unicode(title, "ascii")
@@ -72,7 +83,10 @@ def parse_html(filename, charset=None, log=None):
       except LookupError:
          if log: log("   unknown charset: `%s' or `%s'" % (parser.charset, current_charset))
 
-   parser.title = recode_entities(title, current_charset)
+   title = recode_entities(title, current_charset)
+   title = title.replace('\r', '').replace('\n', ' ').strip()
+   if log: log("   final title    : %s" % title)
+   parser.title = title
    return parser