aniani/vim
0
0
Fork 1
mirror of https://github.com/vim/vim.git synced 2025-09-27 04:14:06 -04:00
Code Activity

patch 8.2.1435: Vim9: always converting to string for ".." leads to mistakes

Browse Source 
Problem:    Vim9: always converting to string for ".." leads to mistakes.
Solution:   Only automatically convert simple types.
This commit is contained in:
Bram Moolenaar
2020-08-12 21:34:49 +02:00
parent fd77748df2
commit 418f1df547
10 changed files with 157 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
src/eval.c
View File

@@ -2712,7 +2712,7 @@ eval5(char_u **arg, typval_T *rettv, evalarg_T *evalarg)
return FAIL;
}
*arg = skipwhite_and_linebreak(*arg + oplen, evalarg);
if (eval6(arg, &var2, evalarg, op == '.') == FAIL)
if (eval6(arg, &var2, evalarg, !in_vim9script() && op == '.') == FAIL)
{
clear_tv(rettv);
return FAIL;
@@ -2727,8 +2727,22 @@ eval5(char_u **arg, typval_T *rettv, evalarg_T *evalarg)
{
char_u buf1[NUMBUFLEN], buf2[NUMBUFLEN];
char_u *s1 = tv_get_string_buf(rettv, buf1);
char_u *s2 = tv_get_string_buf_chk(&var2, buf2);
char_u *s2 = NULL;
if (in_vim9script() && (var2.v_type == VAR_VOID
|| var2.v_type == VAR_CHANNEL
|| var2.v_type == VAR_JOB))
emsg(_(e_inval_string));
#ifdef FEAT_FLOAT
else if (var2.v_type == VAR_FLOAT)
{
vim_snprintf((char *)buf2, NUMBUFLEN, "%g",
var2.vval.v_float);
s2 = buf2;
}
#endif
else
s2 = tv_get_string_buf_chk(&var2, buf2);
if (s2 == NULL) // type error ?
{
clear_tv(rettv);