patch 8.2.1461: Vim9: string indexes are counted in bytes

Problem:    Vim9: string indexes are counted in bytes.
Solution:   Use character indexes. (closes #6574)

runtime/doc/eval.txt

@@ -1131,19 +1131,25 @@ Evaluation is always from left to right.
 
 expr8[expr1]		item of String or |List|	*expr-[]* *E111*
 							*E909* *subscript*
+In legacy Vim script:
 If expr8 is a Number or String this results in a String that contains the
-expr1'th single byte from expr8.  expr8 is used as a String, expr1 as a
+expr1'th single byte from expr8.  expr8 is used as a String (a number is
-Number.  This doesn't recognize multi-byte encodings, see `byteidx()` for
+automatically converted to a String), expr1 as a Number.  This doesn't
-an alternative, or use `split()` to turn the string into a list of characters.
+recognize multi-byte encodings, see `byteidx()` for an alternative, or use
-
+`split()` to turn the string into a list of characters.  Example, to get the
-Index zero gives the first byte.  This is like it works in C.  Careful:
+byte under the cursor: >
-text column numbers start with one!  Example, to get the byte under the
-cursor: >
 	:let c = getline(".")[col(".") - 1]
 
+In Vim9 script:
+If expr8 is a String this results in a String that contains the expr1'th
+single character from expr8.  To use byte indexes use |strpart()|.
+
+Index zero gives the first byte or character.  Careful: text column numbers
+start with one!
+
 If the length of the String is less than the index, the result is an empty
 String.  A negative index always results in an empty string (reason: backward
-compatibility).  Use [-1:] to get the last byte.
+compatibility).  Use [-1:] to get the last byte or character.
 
 If expr8 is a |List| then it results the item at index expr1.  See |list-index|
 for possible index values.  If the index is out of range this results in an
@@ -1157,10 +1163,16 @@ error.
 
 expr8[expr1a : expr1b]	substring or sublist		*expr-[:]*
 
-If expr8 is a Number or String this results in the substring with the bytes
+If expr8 is a String this results in the substring with the bytes from expr1a
-from expr1a to and including expr1b.  expr8 is used as a String, expr1a and
+to and including expr1b.  expr8 is used as a String, expr1a and expr1b are
-expr1b are used as a Number.  This doesn't recognize multi-byte encodings, see
+used as a Number.
-|byteidx()| for computing the indexes.
+
+In legacy Vim script the indexes are byte indexes.  This doesn't recognize
+multi-byte encodings, see |byteidx()| for computing the indexes.  If expr8 is
+a Number it is first converted to a String.
+
+In Vim9 script the indexes are character indexes.  To use byte indexes use
+|strpart()|.
 
 If expr1a is omitted zero is used.  If expr1b is omitted the length of the
 string minus one is used.

src/eval.c

@@ -3718,6 +3718,10 @@ eval_index(
 		    else
 			s = vim_strnsave(s + n1, n2 - n1 + 1);
 		}
+		else if (in_vim9script())
+		{
+		    s = char_from_string(s, n1);
+		}
 		else
 		{
 		    // The resulting variable is a string of a single
@@ -5284,6 +5288,30 @@ eval_isdictc(int c)
     return ASCII_ISALNUM(c) || c == '_';
 }
 
+/*
+ * Return the character "str[index]" where "index" is the character index.  If
+ * "index" is out of range NULL is returned.
+ */
+    char_u *
+char_from_string(char_u *str, varnumber_T index)
+{
+    size_t	    nbyte = 0;
+    varnumber_T	    nchar = index;
+    size_t	    slen;
+
+    if (str == NULL || index < 0)
+	return NULL;
+    slen = STRLEN(str);
+    while (nchar > 0 && nbyte < slen)
+    {
+	nbyte += MB_CPTR2LEN(str + nbyte);
+	--nchar;
+    }
+    if (nbyte >= slen)
+	return NULL;
+    return vim_strnsave(str + nbyte, MB_CPTR2LEN(str + nbyte));
+}
+
 /*
  * Handle:
  * - expr[expr], expr[expr:expr] subscript

src/proto/eval.pro

@@ -59,6 +59,7 @@ char_u *find_name_end(char_u *arg, char_u **expr_start, char_u **expr_end, int f
 int eval_isnamec(int c);
 int eval_isnamec1(int c);
 int eval_isdictc(int c);
+char_u *char_from_string(char_u *str, varnumber_T index);
 int handle_subscript(char_u **arg, typval_T *rettv, evalarg_T *evalarg, int verbose);
 int item_copy(typval_T *from, typval_T *to, int deep, int copyID);
 void echo_one(typval_T *rettv, int with_space, int *atstart, int *needclr);

src/testdir/test_vim9_expr.vim

@@ -2075,12 +2075,28 @@ def Test_expr7_trailing()
 enddef
 
 def Test_expr7_subscript()
-  let text = 'abcdef'
+  let lines =<< trim END
-  assert_equal('', text[-1])
+    let text = 'abcdef'
-  assert_equal('a', text[0])
+    assert_equal('', text[-1])
-  assert_equal('e', text[4])
+    assert_equal('a', text[0])
-  assert_equal('f', text[5])
+    assert_equal('e', text[4])
-  assert_equal('', text[6])
+    assert_equal('f', text[5])
+    assert_equal('', text[6])
+
+    text = 'ábçdëf'
+    assert_equal('', text[-999])
+    assert_equal('', text[-1])
+    assert_equal('á', text[0])
+    assert_equal('b', text[1])
+    assert_equal('ç', text[2])
+    assert_equal('d', text[3])
+    assert_equal('ë', text[4])
+    assert_equal('f', text[5])
+    assert_equal('', text[6])
+    assert_equal('', text[999])
+  END
+  CheckDefSuccess(lines)
+  CheckScriptSuccess(['vim9script'] + lines)
 enddef
 
 def Test_expr7_subscript_linebreak()

src/version.c

@@ -754,6 +754,8 @@ static char *(features[]) =
 
 static int included_patches[] =
 {   /* Add new patch number below this line */
+/**/
+    1461,
 /**/
     1460,
 /**/

src/vim9execute.c

@@ -2233,7 +2233,6 @@ call_def_function(
 
 	    case ISN_STRINDEX:
 		{
-		    char_u	*s;
 		    varnumber_T	n;
 		    char_u	*res;
 
@@ -2245,7 +2244,6 @@ call_def_function(
 			emsg(_(e_stringreq));
 			goto on_error;
 		    }
-		    s = tv->vval.v_string;
 
 		    tv = STACK_TV_BOT(-1);
 		    if (tv->v_type != VAR_NUMBER)
@@ -2259,12 +2257,9 @@ call_def_function(
 		    // The resulting variable is a string of a single
 		    // character.  If the index is too big or negative the
 		    // result is empty.
-		    if (n < 0 || n >= (varnumber_T)STRLEN(s))
-			res = NULL;
-		    else
-			res = vim_strnsave(s + n, 1);
 		    --ectx.ec_stack.ga_len;
 		    tv = STACK_TV_BOT(-1);
+		    res = char_from_string(tv->vval.v_string, n);
 		    vim_free(tv->vval.v_string);
 		    tv->vval.v_string = res;
 		}