X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;ds=sidebyside;f=Robots%2Fparse_html.py;h=5e8061c793e307b0fad3d1163bf3b122884d185c;hb=c26ca363d27b778a96f7d292dabd5258f307b693;hp=cc7655bfa950770d3f092ad7c32d387f2f65bd8b;hpb=349d6e0241d43e42a257d6019972402116236ee7;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index cc7655b..5e8061c 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -13,11 +13,11 @@ DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
 parsers = []
 
 try:
-   import parse_html_etreetidy
+   from parse_html_lxml import parse_html
 except ImportError:
    pass
 else:
-   parsers.append(parse_html_etreetidy.parse_html)
+    parsers.append(parse_html)
 
 try:
    import parse_html_beautifulsoup
@@ -27,8 +27,19 @@ except ImportError:
 else:
    parsers.append(parse_html_beautifulsoup.parse_html)
 
-from parse_html_htmlparser import parse_html
-parsers.append(parse_html)
+try:
+   from parse_html_htmlparser import parse_html
+except ImportError:
+   pass
+else:
+    parsers.append(parse_html)
+
+try:
+   import parse_html_html5
+except ImportError:
+   pass
+else:
+   parsers.append(parse_html_html5.parse_html)
 
 
 import re
@@ -61,6 +72,9 @@ def recode_entities(title, charset):
 
 
 def parse_html(filename, charset=None, log=None):
+   if not parsers:
+       return None
+
    if charset:
       try:
          codecs.lookup(charset) # In case of unknown charset...
@@ -70,8 +84,9 @@ def parse_html(filename, charset=None, log=None):
    charsets = [universal_charset, DEFAULT_CHARSET]
    if charset:
       charset = charset.lower().replace("windows-", "cp")
-      if charset not in charsets:
-         charsets.insert(0, charset)
+      if charset in charsets:
+         charsets.remove(charset)
+      charsets.insert(0, charset)
 
    for p in parsers:
       parser = None
@@ -86,6 +101,9 @@ def parse_html(filename, charset=None, log=None):
       else:
          if log: log("Parser %s.%s failed, trying next one." % (p.__module__, p.__name__))
 
+   if not parser:
+       return None
+
    converted_title = title = parser.title
    if title and (not parser.charset):
       try: