From 9eba9e87fc3ed257bf82c1bc7e2f77dba1aaf9a5 Mon Sep 17 00:00:00 2001
From: Steve Bennett <steveb@workware.net.au>
Date: Sun, 27 Oct 2019 21:17:42 +1000
Subject: regexp -indices should return character indices
Not byte indices
Reported-by: dbohdan <dbohdan@dbohdan.com>
Signed-off-by: Steve Bennett <steveb@workware.net.au>
---
jim-regexp.c | 15 +++++++--------
1 file changed, 7 insertions(+), 8 deletions(-)
(limited to 'jim-regexp.c')
diff --git a/jim-regexp.c b/jim-regexp.c
index 81f3207..3134598 100644
--- a/jim-regexp.c
+++ b/jim-regexp.c
@@ -281,16 +281,15 @@ int Jim_RegexpCmd(Jim_Interp *interp, int argc, Jim_Obj *const *argv)
}
}
else {
- int len = pmatch[j].rm_eo - pmatch[j].rm_so;
-
if (opt_indices) {
- Jim_ListAppendElement(interp, resultObj, Jim_NewIntObj(interp,
- offset + pmatch[j].rm_so));
- Jim_ListAppendElement(interp, resultObj, Jim_NewIntObj(interp,
- offset + pmatch[j].rm_so + len - 1));
+ /* rm_so and rm_eo are byte offsets. We need char offsets */
+ int so = utf8_strlen(source_str, pmatch[j].rm_so);
+ int eo = utf8_strlen(source_str + pmatch[j].rm_so, pmatch[j].rm_eo);
+ Jim_ListAppendElement(interp, resultObj, Jim_NewIntObj(interp, offset + so));
+ Jim_ListAppendElement(interp, resultObj, Jim_NewIntObj(interp, offset + eo - 1));
}
else {
- Jim_AppendString(interp, resultObj, source_str + pmatch[j].rm_so, len);
+ Jim_AppendString(interp, resultObj, source_str + pmatch[j].rm_so, pmatch[j].rm_eo - pmatch[j].rm_so);
}
}
@@ -311,7 +310,7 @@ int Jim_RegexpCmd(Jim_Interp *interp, int argc, Jim_Obj *const *argv)
try_next_match:
if (opt_all && (pattern[0] != '^' || (regcomp_flags & REG_NEWLINE)) && *source_str) {
if (pmatch[0].rm_eo) {
- offset += pmatch[0].rm_eo;
+ offset += utf8_strlen(source_str, pmatch[0].rm_eo);
source_str += pmatch[0].rm_eo;
}
else {
--
cgit v1.1