a big one:

 - old mangle code has gone, the new one based on tdb seem resonably ok
   probably the valid.dat table need to be updated to treat wild chars as
   invalid ones (work ok without it)
 - a LOT of new string manipulation function for unicode, they are somewhat
   tested but a review would not be bad
 - some new function I will need for the new unix_convert function I'm writing,
   this will be renamed filename_convert and use only unicode strings.
 - charconv, I attached a comment, if someone wnat to look if I'm right or
   just was hacking to late in the night to make a sane one :)

of course any bug is my responsibility an will be pleased to see patches if
you find any. :-)

Simo.
(This used to be commit ee19f7efb6ea9216fc91cf112ac1afa691983e9d)

1 files changed, 7 insertions, 1 deletions

diff --git a/source3/lib/charcnv.c b/source3/lib/charcnv.c
index e55de729c2a..ebf316d9dd3 100644
--- a/source3/lib/charcnv.c
+++ b/source3/lib/charcnv.c
@@ -123,7 +123,13 @@ size_t convert_string(charset_t from, charset_t to,
 		{ case EINVAL: reason="Incomplete multibyte sequence"; break;
 		  case E2BIG:  reason="No more room"; 
 		   	       DEBUG(0, ("Required %d, available %d\n",
-			       srclen, destlen));	
+			       srclen, destlen));
+			       /* we are not sure we need srclen bytes,
+			          may be more, may be less.
+				  We only know we need more than destlen
+				  bytes ---simo */
+				
+						
 		               break;
 		  case EILSEQ: reason="Illegal myltibyte sequence"; break;
 		}