Patch 8.2.0938
Problem: NFA regexp uses tolower() to compare ignore-case. (Thayne McCombs)
Solution: Use utf_fold() when possible. (ref. neovim #12456)
Files: src/macros.h, src/diff.c, src/regexp_nfa.c,
src/testdir/test_regexp_utf8.vim
*** ../vim-8.2.0937/src/macros.h 2020-06-01 17:28:31.511939716 +0200
--- src/macros.h 2020-06-09 19:04:44.000093479 +0200
***************
*** 93,98 ****
--- 93,99 ----
#define MB_ISUPPER(c) vim_isupper(c)
#define MB_TOLOWER(c) vim_tolower(c)
#define MB_TOUPPER(c) vim_toupper(c)
+ #define MB_CASEFOLD(c) (enc_utf8 ? utf_fold(c) : MB_TOLOWER(c))
// Use our own isdigit() replacement, because on MS-Windows isdigit() returns
// non-zero for superscript 1. Also avoids that isdigit() crashes for numbers
*** ../vim-8.2.0937/src/diff.c 2020-06-07 20:49:02.077891881 +0200
--- src/diff.c 2020-06-09 19:04:25.088157656 +0200
***************
*** 747,753 ****
// xdiff doesn't support ignoring case, fold-case the text.
c = PTR2CHAR(s);
! c = enc_utf8 ? utf_fold(c) : MB_TOLOWER(c);
orig_len = mb_ptr2len(s);
if (mb_char2bytes(c, cbuf) != orig_len)
// TODO: handle byte length difference
--- 747,753 ----
// xdiff doesn't support ignoring case, fold-case the text.
c = PTR2CHAR(s);
! c = MB_CASEFOLD(c);
orig_len = mb_ptr2len(s);
if (mb_char2bytes(c, cbuf) != orig_len)
// TODO: handle byte length difference
*** ../vim-8.2.0937/src/regexp_nfa.c 2020-06-03 18:55:35.961570633 +0200
--- src/regexp_nfa.c 2020-06-09 19:26:59.135466405 +0200
***************
*** 5459,5465 ****
{
c1 = PTR2CHAR(match_text + len1);
c2 = PTR2CHAR(rex.line + col + len2);
! if (c1 != c2 && (!rex.reg_ic || MB_TOLOWER(c1) != MB_TOLOWER(c2)))
{
match = FALSE;
break;
--- 5459,5465 ----
{
c1 = PTR2CHAR(match_text + len1);
c2 = PTR2CHAR(rex.line + col + len2);
! if (c1 != c2 && (!rex.reg_ic || MB_CASEFOLD(c1) != MB_CASEFOLD(c2)))
{
match = FALSE;
break;
***************
*** 6271,6281 ****
}
if (rex.reg_ic)
{
! int curc_low = MB_TOLOWER(curc);
int done = FALSE;
for ( ; c1 <= c2; ++c1)
! if (MB_TOLOWER(c1) == curc_low)
{
result = result_if_matched;
done = TRUE;
--- 6271,6281 ----
}
if (rex.reg_ic)
{
! int curc_low = MB_CASEFOLD(curc);
int done = FALSE;
for ( ; c1 <= c2; ++c1)
! if (MB_CASEFOLD(c1) == curc_low)
{
result = result_if_matched;
done = TRUE;
***************
*** 6287,6294 ****
}
else if (state->c < 0 ? check_char_class(state->c, curc)
: (curc == state->c
! || (rex.reg_ic && MB_TOLOWER(curc)
! == MB_TOLOWER(state->c))))
{
result = result_if_matched;
break;
--- 6287,6294 ----
}
else if (state->c < 0 ? check_char_class(state->c, curc)
: (curc == state->c
! || (rex.reg_ic && MB_CASEFOLD(curc)
! == MB_CASEFOLD(state->c))))
{
result = result_if_matched;
break;
***************
*** 6713,6719 ****
result = (c == curc);
if (!result && rex.reg_ic)
! result = MB_TOLOWER(c) == MB_TOLOWER(curc);
// If rex.reg_icombine is not set only skip over the character
// itself. When it is set skip over composing characters.
if (result && enc_utf8 && !rex.reg_icombine)
--- 6713,6719 ----
result = (c == curc);
if (!result && rex.reg_ic)
! result = MB_CASEFOLD(c) == MB_CASEFOLD(curc);
// If rex.reg_icombine is not set only skip over the character
// itself. When it is set skip over composing characters.
if (result && enc_utf8 && !rex.reg_icombine)
***************
*** 6882,6888 ****
// cheaper than adding a state that won't match.
c = PTR2CHAR(rex.input + clen);
if (c != prog->regstart && (!rex.reg_ic
! || MB_TOLOWER(c) != MB_TOLOWER(prog->regstart)))
{
#ifdef ENABLE_LOG
fprintf(log_fd, " Skipping start state, regstart
does not match\n");
--- 6882,6888 ----
// cheaper than adding a state that won't match.
c = PTR2CHAR(rex.input + clen);
if (c != prog->regstart && (!rex.reg_ic
! || MB_CASEFOLD(c) != MB_CASEFOLD(prog->regstart)))
{
#ifdef ENABLE_LOG
fprintf(log_fd, " Skipping start state, regstart
does not match\n");
*** ../vim-8.2.0937/src/testdir/test_regexp_utf8.vim 2019-12-16
22:43:22.328823387 +0100
--- src/testdir/test_regexp_utf8.vim 2020-06-09 19:28:03.727241430 +0200
***************
*** 355,358 ****
--- 355,377 ----
set regexpengine& ambiwidth&
endfunc
+ func Run_regexp_ignore_case()
+ call assert_equal('iIİ', substitute('iIİ', '\([iIİ]\)', '\1', 'g'))
+
+ call assert_equal('iIx', substitute('iIİ', '\c\([İ]\)', 'x', 'g'))
+ call assert_equal('xxİ', substitute('iIİ', '\(i\c\)', 'x', 'g'))
+ call assert_equal('iIx', substitute('iIİ', '\(İ\c\)', 'x', 'g'))
+ call assert_equal('iIx', substitute('iIİ', '\c\(\%u0130\)', 'x', 'g'))
+ call assert_equal('iIx', substitute('iIİ', '\c\([\u0130]\)', 'x', 'g'))
+ call assert_equal('iIx', substitute('iIİ', '\c\([\u012f-\u0131]\)', 'x',
'g'))
+ endfunc
+
+ func Test_regexp_ignore_case()
+ set regexpengine=1
+ call Run_regexp_ignore_case()
+ set regexpengine=2
+ call Run_regexp_ignore_case()
+ set regexpengine&
+ endfunc
+
" vim: shiftwidth=2 sts=2 expandtab
*** ../vim-8.2.0937/src/version.c 2020-06-09 17:30:00.515654723 +0200
--- src/version.c 2020-06-09 19:30:58.914635149 +0200
***************
*** 756,757 ****
--- 756,759 ----
{ /* Add new patch number below this line */
+ /**/
+ 938,
/**/
--
Not too long ago, a keyboard was something to make music with...
/// Bram Moolenaar -- [email protected] -- http://www.Moolenaar.net \\\
/// sponsor Vim, vote for features -- http://www.Vim.org/sponsor/ \\\
\\\ an exciting new programming language -- http://www.Zimbu.org ///
\\\ help me help AIDS victims -- http://ICCF-Holland.org ///
--
--
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].
To view this discussion on the web visit
https://groups.google.com/d/msgid/vim_dev/202006091735.059HZVIn629512%40masaka.moolenaar.net.
