patch 9.1.0296: regexp: engines do not handle case-folding wellv9.1.0296

Problem:  Regex engines do not handle case-folding well
Solution: Correctly calculate byte length of characters to skip

When the regexp engine compares two utf-8 codepoints case insensitively
it may match an adjacent character, because it assumes it can step over
as many bytes as the pattern contains.

This however is not necessarily true because of case-folding, a
multi-byte UTF-8 character can be considered equal to some single-byte
value.

Let's consider the pattern 'ſ' and the string 's'. When comparing and
ignoring case, the single character 's' matches, and since it matches
Vim will try to step over the match (by the amount of bytes of the
pattern), assuming that since it matches, the length of both strings is
the same.

However in that case, it should only step over the single byte
value 's' so by 1 byte and try to start matching after it again. So for the
backtracking engine we need to ensure:
- we try to match the correct length for the pattern and the text
- in case of a match, we step over it correctly

The same thing can happen for the NFA engine, when skipping to the next
character to test for a match. We are skipping over the regstart
pointer, however we do not consider the case that because of
case-folding we may need to adjust the number of bytes to skip over. So
this needs to be adjusted in find_match_text() as well.

A related issue turned out, when prog->match_text is actually empty. In
that case we should try to find the next match and skip this condition.

fixes: #14294
closes: #14433

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

8 files changed, 66 insertions, 3 deletions

diff --git a/src/mbyte.c b/src/mbyte.c
index d6fb7ecc76..3be75099f1 100644
--- a/src/mbyte.c
+++ b/src/mbyte.c
@@ -3801,6 +3801,15 @@ utf_strnicmp(
  * 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)
 {
     int		i, l;
diff --git a/src/proto/mbyte.pro b/src/proto/mbyte.pro
index 7883b3b4c7..c49f7e7072 100644
--- a/src/proto/mbyte.pro
+++ b/src/proto/mbyte.pro
@@ -48,6 +48,7 @@ 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 4373ae0cfa..4e85ebc29e 100644
--- a/src/regexp.c
+++ b/src/regexp.c
@@ -1606,7 +1606,9 @@ 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.
+ * 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')
  */
     static int
 cstrncmp(char_u *s1, char_u *s2, int *n)
@@ -1615,6 +1617,13 @@ 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 5d9450d871..2a03fec579 100644
--- a/src/regexp_bt.c
+++ b/src/regexp_bt.c
@@ -3816,6 +3816,14 @@ 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 5e4fadd028..451720a09d 100644
--- a/src/regexp_nfa.c
+++ b/src/regexp_nfa.c
@@ -5666,7 +5666,12 @@ find_match_text(colnr_T *startcol, int regstart, char_u *match_text)
     for (;;)
     {
 	match = TRUE;
-	len2 = MB_CHAR2LEN(regstart); // skip regstart
+	// 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));
 	for (len1 = 0; match_text[len1] != NUL; len1 += MB_CHAR2LEN(c1))
 	{
 	    c1 = PTR2CHAR(match_text + len1);
@@ -7503,7 +7508,7 @@ 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 && !rex.reg_icombine)
+	if (prog->match_text != NULL && *prog->match_text != NUL && !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 6669dee57e..9980e5b7f5 100644
--- a/src/testdir/test_regexp_utf8.vim
+++ b/src/testdir/test_regexp_utf8.vim
@@ -587,4 +587,32 @@ 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 2c6e7d02fd..c63b141aa9 100644
--- a/src/version.c
+++ b/src/version.c
@@ -705,6 +705,8 @@ static char *(features[]) =
 static int included_patches[] =
 {   /* Add new patch number below this line */
 /**/
+    296,
+/**/
     295,
 /**/
     294,
diff --git a/src/vim.h b/src/vim.h
index 4507674fcd..33d592038d 100644
--- a/src/vim.h
+++ b/src/vim.h
@@ -1751,6 +1751,7 @@ 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))