On Tue, Mar 29, 2016 at 9:53 PM, Bram Moolenaar <[email protected]> wrote:
>
> Dominique wrote:
>
>> Bram Moolenaar wrote:
>>
>> > Patch 7.4.704
>> > Problem:    Searching for a character matches an illegal byte and causes
>> >             invalid memory access. (Dominique Pelle)
>> > Solution:   Do not match an invalid byte when search for a character in a
>> >             string.  Fix equivalence classes using negative numbers, which
>> >             result in illegal bytes.
>> > Files:      src/misc2.c, src/regexp.c, src/testdir/test44.in
>>
>> ...snip...
>>
>> > *** ../vim-7.4.703/src/regexp.c 2015-03-31 14:17:22.004608294 +0200
>> > --- src/regexp.c        2015-04-21 13:55:32.778050479 +0200
>> > ***************
>> > *** 845,857 ****
>> >   #else
>> >         switch (c)
>> >         {
>> > !           case 'A': case '\300': case '\301': case '\302':
>> >             CASEMBC(0x100) CASEMBC(0x102) CASEMBC(0x104) CASEMBC(0x1cd)
>> >             CASEMBC(0x1de) CASEMBC(0x1e0) CASEMBC(0x1ea2)
>> > !           case '\303': case '\304': case '\305':
>> > !                     regmbc('A'); regmbc('\300'); regmbc('\301');
>> > !                     regmbc('\302'); regmbc('\303'); regmbc('\304');
>> > !                     regmbc('\305');
>> >                       REGMBC(0x100) REGMBC(0x102) REGMBC(0x104)
>> >                       REGMBC(0x1cd) REGMBC(0x1de) REGMBC(0x1e0)
>> >                       REGMBC(0x1ea2)
>> > --- 845,858 ----
>> >   #else
>> >         switch (c)
>> >         {
>> > !           /* Do not use '\300' style, it results in a negative number. */
>> > !           case 'A': case 0xc0: case 0xc1: case 0xc2:
>> > !           case 0xc3: case 0xc4: case 0xc5:
>> >             CASEMBC(0x100) CASEMBC(0x102) CASEMBC(0x104) CASEMBC(0x1cd)
>> >             CASEMBC(0x1de) CASEMBC(0x1e0) CASEMBC(0x1ea2)
>> > !                     regmbc('A'); regmbc(0xc0); regmbc(0xc1);
>> > !                     regmbc(0xc2); regmbc(0xc3); regmbc(0xc4);
>> > !                     regmbc(0xc5);
>> >                       REGMBC(0x100) REGMBC(0x102) REGMBC(0x104)
>> >                       REGMBC(0x1cd) REGMBC(0x1de) REGMBC(0x1e0)
>> >                       REGMBC(0x1ea2)
>>
>>
>> The replacements of things like case '\0300' into case 0xc0 (etc)
>> were made in regexp.c in patch-7.4.704. Similar code exists in
>> regexp_nfa.c in nfa_emit_equi_class() where the replacements
>> were not made. I suppose that they should be made there too
>> for the same reason explained in the comment:
>> /* Do not use '\300' style, it results in a negative number. */
>>
>> I can write a patch when I find the time, and try to find an
>> example of where the change matters.
>
> A patch was just suggested for the same code, related to supporting
> EBCDIC.  Perhaps you two can get together?  And add the test that's
> missing.


I was wrong actually: patch 7.4.704 replaced things like
case '\300':  into case 0xc0. But regexp_nfa.c does not have
things like case '\300':  it has case 0300: which is fine.

As for EBCDIC, I see that a patch has already been proposed.
I don't have z/OS so I'd rather let someone with z/OS submit
the patch for EBCDIC.

Regards
Dominique

-- 
-- 
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php

--- 
You received this message because you are subscribed to the Google Groups 
"vim_dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Raspunde prin e-mail lui