On 9/22/07, Tels <nospam-abuse bloodgate.com> wrote:
> Moin,
>
> On Saturday 22 September 2007 13:19:48 demerphq wrote:
> > On 9/22/07, Tels <nospam-abuse bloodgate.com> wrote:
> > > Moin,
> > >
> > > On Saturday 22 September 2007 12:25:52 Tels
wrote:
> > > > Moin,
> > >
> > > [snip]
> > >
> > > > make && make test still runs, I
will report in probably 25 mins how
> > > > it goes.
> > >
> > > Attached is the patch I used. Unfortunately,
it doesn't seem to work,
> > > as you can see from the output also attached
:(
> >
> > Dang. I guess itll have to wait until i have the
time and
> > circumstances to look into this further.
>
> Do you have any hints on what the "something's
wrong" warning means or where
> it comes from?
Its the normal warning produced by 'warn' when called with
no arguments.
d:sync-clone>perl -e"warn"
Warning: something's wrong at -e line 1.
> I might have a try with "DEBUG => ALL" and
see what I can glean from it. In
> any way, Remote Debugging Via Proxy[tm] isn't easy
Hmm. Actually i just realized that I was being dumb, folder
is only
true when !UTF and we are doing a case insensitive match.
Change the
patch as follows and I think it should work. (IOW instead of
replace
the if (folder) line, insert after it the new logic with the
right
test):
/* store the codepoint in the bitmap,
and if its ascii
also store its folded equivelent.
*/
TRIE_BITMAP_SET(trie,uvc);
if ( folder )
TRIE_BITMAP_SET(trie,folder[ uvc ]);
+
+ if ( !UTF ) {
+ /* store first byte of utf8 representation of
+ codepoints in the 127 < uvc < 256 range */
+ if (127 < uvc && uvc < 192) {
+ TRIE_BITMAP_SET(trie,194);
+ } else if (191 < uvc ) {
+ TRIE_BITMAP_SET(trie,195);
+ /* && uvc < 256 -- we know uvc is < 256
already */
+ }
+ }
set_bit = 0; /* We've done our bit */
}
} else {
--
perl -Mre=debug -e "/just|another|perl|hacker/"
|