Pass charset from the command line.

[bookmarks_db.git] / Robots / parse_html.py

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index 038cfaf7aa3b0e51be09518f001c9abbda240808..678f3e2dae49d9961c07e040aeca2ec7e8e832aa 100755 (executable)
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -57,14 +57,17 @@ def parse_html(filename, charset=None, log=None):
       except (ValueError, LookupError):
          charset = None         # ...try charset from HTML
 
-   for p in parsers:
-      charsets = [universal_charset, DEFAULT_CHARSET]
+   charsets = [universal_charset, DEFAULT_CHARSET]
+   if charset:
+      charset = charset.lower().replace("windows-", "cp")
       if charset not in charsets:
          charsets.insert(0, charset)
+
+   for p in parsers:
       parser = None
       for c in charsets:
          try:
-            parser = p(filename, c)
+            parser = p(filename, c, log)
             break
          except UnicodeEncodeError:
             pass
@@ -81,25 +84,28 @@ def parse_html(filename, charset=None, log=None):
          parser.charset = DEFAULT_CHARSET
 
    if parser.charset:
-      parser.charset = parser.charset.replace("windows-", "cp").lower()
+      parser.charset = parser.charset.lower().replace("windows-", "cp")
 
-   if parser.charset and (parser.charset <> universal_charset):
+   if 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)
-         else:
+         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)
-         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 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)
       except LookupError:
-         if log: log("   unknown charset: `%s' or `%s'" % (parser.charset, universal_charset))
+         if log: log("   unknown charset: '%s'" % parser.charset)
    else:
       if log: log("   title          : %s" % title)
 
@@ -113,10 +119,18 @@ def parse_html(filename, charset=None, log=None):
 
 if __name__ == '__main__':
    import sys
-   from m_lib.defenc import default_encoding
-   current_charset = default_encoding.replace("windows-", "cp")
 
-   parser = parse_html(sys.argv[1], universal_charset,
+   l = len(sys.argv)
+   if l == 3:
+      filename = sys.argv[1]
+      charset = sys.argv[2]
+   elif l == 2:
+      filename = sys.argv[1]
+      charset = universal_charset
+   else:
+      sys.exit("Usage: %s filename [charset]" % sys.argv[0])
+
+   parser = parse_html(filename, universal_charset,
       log=lambda s: sys.stdout.write(s + '\n'))
    print "   refresh:", parser.refresh
    print "   icon   :", parser.icon