Patch 7.3.1015
Problem: New regexp engine: Matching composing characters is wrong.
Solution: Fix matching composing characters.
Files: src/regexp_nfa.c, src/testdir/test95.in, src/testdir/test95.ok
*** ../vim-7.3.1014/src/regexp_nfa.c 2013-05-25 12:28:08.000000000 +0200
--- src/regexp_nfa.c 2013-05-25 14:25:42.000000000 +0200
***************
*** 716,721 ****
--- 716,722 ----
* the composing char is matched here. */
if (enc_utf8 && c == Magic('.') && utf_iscomposing(peekchr()))
{
+ old_regparse = regparse;
c = getchr();
goto nfa_do_multibyte;
}
***************
*** 1217,1225 ****
nfa_do_multibyte:
/* Length of current char with composing chars. */
! if (enc_utf8 && clen != (plen = (*mb_ptr2len)(old_regparse)))
{
! /* A base character plus composing characters.
* This requires creating a separate atom as if enclosing
* the characters in (), where NFA_COMPOSING is the ( and
* NFA_END_COMPOSING is the ). Note that right now we are
--- 1218,1228 ----
nfa_do_multibyte:
/* Length of current char with composing chars. */
! if (enc_utf8 && (clen != (plen = (*mb_ptr2len)(old_regparse))
! || utf_iscomposing(c)))
{
! /* A base character plus composing characters, or just one
! * or more composing characters.
* This requires creating a separate atom as if enclosing
* the characters in (), where NFA_COMPOSING is the ( and
* NFA_END_COMPOSING is the ). Note that right now we are
***************
*** 1400,1406 ****
/* Save pos after the repeated atom and the \{} */
new_regparse = regparse;
- new_regparse = regparse;
quest = (greedy == TRUE? NFA_QUEST : NFA_QUEST_NONGREEDY);
for (i = 0; i < maxval; i++)
{
--- 1403,1408 ----
***************
*** 3218,3228 ****
result = OK;
sta = t->state->out;
len = 0;
if (ireg_icombine)
{
! /* If \Z was present, then ignore composing characters. */
/* TODO: How about negated? */
! if (sta->c != c)
result = FAIL;
len = n;
while (sta->c != NFA_END_COMPOSING)
--- 3220,3238 ----
result = OK;
sta = t->state->out;
len = 0;
+ if (utf_iscomposing(sta->c))
+ {
+ /* Only match composing character(s), ignore base
+ * character. Used for ".{composing}" and "{composing}"
+ * (no preceding character). */
+ len += mb_char2len(c);
+ }
if (ireg_icombine)
{
! /* If \Z was present, then ignore composing characters.
! * When ignoring the base character this always matches. */
/* TODO: How about negated? */
! if (len == 0 && sta->c != c)
result = FAIL;
len = n;
while (sta->c != NFA_END_COMPOSING)
*** ../vim-7.3.1014/src/testdir/test95.in 2013-05-24 23:10:45.000000000
+0200
--- src/testdir/test95.in 2013-05-25 14:36:50.000000000 +0200
***************
*** 38,43 ****
--- 38,52 ----
:"""" Test composing character matching
:call add(tl, ['.ม', 'xม่x yมy', 'yม'])
:call add(tl, ['.ม่', 'xม่x yมy', 'xม่'])
+ :call add(tl, ["\u05b9", " x\u05b9 ", "x\u05b9"])
+ :call add(tl, [".\u05b9", " x\u05b9 ", "x\u05b9"])
+ :call add(tl, ["\u05b9\u05bb", " x\u05b9\u05bb ", "x\u05b9\u05bb"])
+ :call add(tl, [".\u05b9\u05bb", " x\u05b9\u05bb ", "x\u05b9\u05bb"])
+ :call add(tl, ["\u05bb\u05b9", " x\u05b9\u05bb "])
+ :call add(tl, [".\u05bb\u05b9", " x\u05b9\u05bb "])
+ :call add(tl, ["\u05b9", " y\u05bb x\u05b9 ", "x\u05b9"])
+ :call add(tl, [".\u05b9", " y\u05bb x\u05b9 ", "x\u05b9"])
+
:"""" Test \Z
:call add(tl, ['ú\Z', 'x'])
***************
*** 50,55 ****
--- 59,66 ----
:call add(tl, ["ק\u200d\u05b9x\\Z", "xק\u200dxy", "ק\u200dx"])
:call add(tl, ["ק\u200dx\\Z", "xק\u200d\u05b9xy", "ק\u200d\u05b9x"])
:call add(tl, ["ק\u200dx\\Z", "xק\u200dxy", "ק\u200dx"])
+ :call add(tl, ["\u05b9\\+\\Z", "xyz", "xyz"])
+ :call add(tl, ["\\Z\u05b9\\+", "xyz", "xyz"])
:"""" Combining different tests and features
:call add(tl, ['[^[=a=]]\+', 'ddaãâbcd', 'dd'])
*** ../vim-7.3.1014/src/testdir/test95.ok 2013-05-24 23:10:45.000000000
+0200
--- src/testdir/test95.ok 2013-05-25 14:36:54.000000000 +0200
***************
*** 11,16 ****
--- 11,24 ----
OK - \%#=1\f\+
OK - .ม
OK - .ม่
+ OK - ֹ
+ OK - .ֹ
+ OK - ֹֻ
+ OK - .ֹֻ
+ OK - ֹֻ
+ OK - .ֹֻ
+ OK - ֹ
+ OK - .ֹ
OK - ú\Z
OK - יהוה\Z
OK - יְהוָה\Z
***************
*** 21,24 ****
--- 29,34 ----
OK - קֹx\Z
OK - קx\Z
OK - קx\Z
+ OK - ֹ\+\Z
+ OK - \Zֹ\+
OK - [^[=a=]]\+
*** ../vim-7.3.1014/src/version.c 2013-05-25 12:28:08.000000000 +0200
--- src/version.c 2013-05-25 14:41:05.000000000 +0200
***************
*** 730,731 ****
--- 730,733 ----
{ /* Add new patch number below this line */
+ /**/
+ 1015,
/**/
--
If your company is not involved in something called "ISO 9000" you probably
have no idea what it is. If your company _is_ involved in ISO 9000 then you
definitely have no idea what it is.
(Scott Adams - The Dilbert principle)
/// 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].
For more options, visit https://groups.google.com/groups/opt_out.
