patch 9.1.0297: Patch 9.1.0296 causes too many issues

Problem:  Patch 9.1.0296 causes too many issues
          (Tony Mechelynck, @chdiza, CI)
Solution: Back out the change for now

Revert "patch 9.1.0296: regexp: engines do not handle case-folding well"

This reverts commit 7a27c108e0509f3255ebdcb6558e896c223e4d23 it causes
issues with syntax highlighting and breaks the FreeBSD and MacOS CI. It
needs more work.

fixes: #14487

Signed-off-by: Christian Brabandt <cb@256bit.org>

diff --git a/src/mbyte.c b/src/mbyte.c
index 3be75099f10e6269718a3c20830fe978700e3bcd..d6fb7ecc761d2800138bd70d5f66cd7d71ce4875 100644 (file)
--- a/src/mbyte.c
+++ b/src/mbyte.c
@@ -3800,15 +3800,6 @@ utf_strnicmp(
  * Returns zero if s1 and s2 are equal (ignoring case), the difference between
  * two characters otherwise.
  */
-    int
-mb_strnicmp2(char_u *s1, char_u *s2, int n1, int n2)
-{
-    if (n1 == n2 || !enc_utf8)
-       return mb_strnicmp(s1, s2, n1);
-    else
-       return utf_strnicmp(s1, s2, n1, n2);
-}
-
     int
 mb_strnicmp(char_u *s1, char_u *s2, size_t nn)
 {

diff --git a/src/proto/mbyte.pro b/src/proto/mbyte.pro
index c49f7e7072a68e3db3c7a8b8d61ee1773879d42e..7883b3b4c7331093ea3b3618fed492151eef9794 100644 (file)
--- a/src/proto/mbyte.pro
+++ b/src/proto/mbyte.pro
@@ -48,7 +48,6 @@ int utf_islower(int a);
 int utf_tolower(int a);
 int utf_isupper(int a);
 int mb_strnicmp(char_u *s1, char_u *s2, size_t nn);
-int mb_strnicmp2(char_u *s1, char_u *s2, int n1, int n2);
 void show_utf8(void);
 int latin_head_off(char_u *base, char_u *p);
 int dbcs_screen_head_off(char_u *base, char_u *p);

diff --git a/src/regexp.c b/src/regexp.c
index 4e85ebc29e728059112bab54daf8149023b5cdf9..4373ae0cfaf41f6e875db7990471e16fded85bc0 100644 (file)
--- a/src/regexp.c
+++ b/src/regexp.c
@@ -1606,9 +1606,7 @@ mb_decompose(int c, int *c1, int *c2, int *c3)
 /*
  * Compare two strings, ignore case if rex.reg_ic set.
  * Return 0 if strings match, non-zero otherwise.
- * Correct the length "*n" when composing characters are ignored
- * or for utf8 when both utf codepoints are considered equal because of
- * case-folding but have different length (e.g. 's' and 'ſ')
+ * Correct the length "*n" when composing characters are ignored.
  */
     static int
 cstrncmp(char_u *s1, char_u *s2, int *n)
@@ -1617,13 +1615,6 @@ cstrncmp(char_u *s1, char_u *s2, int *n)
 
     if (!rex.reg_ic)
        result = STRNCMP(s1, s2, *n);
-    else if (enc_utf8)
-    {
-       int l2 = mb_ptr2len(s2);
-       result = MB_STRNICMP2(s1, s2, *n, l2);
-       if (result == 0 && l2 < *n)
-           *n = l2;
-    }
     else
        result = MB_STRNICMP(s1, s2, *n);
 

diff --git a/src/regexp_bt.c b/src/regexp_bt.c
index 2a03fec579c82feb9e0e897e76650a422b522c38..5d9450d8712b3f284ca6c23e341ce142e07a58a8 100644 (file)
--- a/src/regexp_bt.c
+++ b/src/regexp_bt.c
@@ -3816,14 +3816,6 @@ regmatch(
                        }
                    }
                }
-               else if (enc_utf8)
-               {
-                   if (cstrncmp(opnd, rex.input, &len) != 0)
-                   {
-                       status = RA_NOMATCH;
-                       break;
-                   }
-               }
                else
                    for (i = 0; i < len; ++i)
                        if (opnd[i] != rex.input[i])

diff --git a/src/regexp_nfa.c b/src/regexp_nfa.c
index 451720a09dd88c192d19a9104bf42c69dea620fc..5e4fadd028bdcae30bc109a5c6d9f0315e291183 100644 (file)
--- a/src/regexp_nfa.c
+++ b/src/regexp_nfa.c
@@ -5666,12 +5666,7 @@ find_match_text(colnr_T *startcol, int regstart, char_u *match_text)
     for (;;)
     {
        match = TRUE;
-       // skip regstart
-       len2 = MB_CHAR2LEN(regstart);
-       if (enc_utf8 && len2 > 1 && MB_CHAR2LEN(PTR2CHAR(rex.line + col)) != len2)
-           // because of case-folding of the previously matched text, we may need
-           // to skip fewer bytes than mb_char2len(regstart)
-           len2 = mb_char2len(utf_fold(regstart));
+       len2 = MB_CHAR2LEN(regstart); // skip regstart
        for (len1 = 0; match_text[len1] != NUL; len1 += MB_CHAR2LEN(c1))
        {
            c1 = PTR2CHAR(match_text + len1);
@@ -7508,7 +7503,7 @@ nfa_regexec_both(
 
        // If match_text is set it contains the full text that must match.
        // Nothing else to try. Doesn't handle combining chars well.
-       if (prog->match_text != NULL && *prog->match_text != NUL && !rex.reg_icombine)
+       if (prog->match_text != NULL && !rex.reg_icombine)
        {
            retval = find_match_text(&col, prog->regstart, prog->match_text);
            if (REG_MULTI)

diff --git a/src/testdir/test_regexp_utf8.vim b/src/testdir/test_regexp_utf8.vim
index 9980e5b7f5505e47142b5825522f6f882bdb4305..6669dee57e4cc7af650ed2d5bb4ba297c6da5d11 100644 (file)
--- a/src/testdir/test_regexp_utf8.vim
+++ b/src/testdir/test_regexp_utf8.vim
@@ -587,32 +587,4 @@ func Test_combining_chars_in_collection()
   bw!
 endfunc
 
-func Test_search_multibyte_match_ascii()
-  new
-  " Match single 'ſ' and 's'
-  call setline(1,  'das abc heraus abc ſich abc ſind')
-  for i in range(0, 2)
-    exe "set re="..i
-    let ic_match = matchbufline('%', '\c\%u17f', 1, '$')->mapnew({idx, val -> val.text})
-    let noic_match = matchbufline('%', '\C\%u17f', 1, '$')->mapnew({idx, val -> val.text})
-    call assert_equal(['s', 's', 'ſ','ſ'], ic_match, "Ignorecase Regex-engine: " .. &re)
-    call assert_equal(['ſ','ſ'], noic_match, "No-Ignorecase Regex-engine: " .. &re)
-  endfor
-  " Match several 'ſſ' and 'ss'
-  call setline(1,  'das abc herauss abc ſſich abc ſind')
-  for i in range(0, 2)
-    exe "set re="..i
-    let ic_match = matchbufline('%', '\c\%u17f\%u17f', 1, '$')->mapnew({idx, val -> val.text})
-    let noic_match = matchbufline('%', '\C\%u17f\%u17f', 1, '$')->mapnew({idx, val -> val.text})
-    let ic_match2 = matchbufline('%', '\c\%u17f\+', 1, '$')->mapnew({idx, val -> val.text})
-    let noic_match2 = matchbufline('%', '\C\%u17f\+', 1, '$')->mapnew({idx, val -> val.text})
-
-    call assert_equal(['ss', 'ſſ'], ic_match, "Ignorecase Regex-engine: " .. &re)
-    call assert_equal(['ſſ'], noic_match, "No-Ignorecase Regex-engine: " .. &re)
-    call assert_equal(['s', 'ss', 'ſſ', 'ſ'], ic_match2, "Ignorecase Regex-engine: " .. &re)
-    call assert_equal(['ſſ','ſ'], noic_match2, "No-Ignorecase Regex-engine: " .. &re)
-  endfor
-  bw!
-endfunc
-
 " vim: shiftwidth=2 sts=2 expandtab

diff --git a/src/version.c b/src/version.c
index c63b141aa9399140318a292a91c4fcab639c8a46..5bf9581b56be2599e68292653f38b5c5746fa83f 100644 (file)
--- a/src/version.c
+++ b/src/version.c
@@ -704,6 +704,8 @@ static char *(features[]) =
 
 static int included_patches[] =
 {   /* Add new patch number below this line */
+/**/
+    297,
 /**/
     296,
 /**/

diff --git a/src/vim.h b/src/vim.h
index 33d592038d1f2dead21e883bf142566e69ca4661..4507674fcde6efb73c2ab9455f0f4ab969120d3c 100644 (file)
--- a/src/vim.h
+++ b/src/vim.h
@@ -1751,7 +1751,6 @@ void *vim_memset(void *, int, size_t);
 
 # define MB_STRICMP(d, s)      mb_strnicmp((char_u *)(d), (char_u *)(s), (int)MAXCOL)
 # define MB_STRNICMP(d, s, n)  mb_strnicmp((char_u *)(d), (char_u *)(s), (int)(n))
-# define MB_STRNICMP2(d, s, n1, n2)    mb_strnicmp2((char_u *)(d), (char_u *)(s), (int)(n1), (int)(n2))
 
 #define STRCAT(d, s)       strcat((char *)(d), (char *)(s))
 #define STRNCAT(d, s, n)    strncat((char *)(d), (char *)(s), (size_t)(n))