chop with utf-8. fix #2967

src/string.c

@@ -1318,7 +1318,18 @@ mrb_str_chop_bang(mrb_state *mrb, mrb_value str)
   mrb_str_modify(mrb, s);
   if (RSTR_LEN(s) > 0) {
     mrb_int len;
+#ifdef MRB_UTF8_STRING
+    const char* t = RSTR_PTR(s), *p = t;
+    const char* e = p + RSTR_LEN(s);
+    while (p<e) {
+      mrb_int clen = utf8len(p, e);
+      if (p + clen>=e) break;
+      p += clen;
+    }
+    len = p - t;
+#else
     len = RSTR_LEN(s) - 1;
+#endif
     if (RSTR_PTR(s)[len] == '\n') {
       if (len > 0 &&
           RSTR_PTR(s)[len-1] == '\r') {

test/t/string.rb

@@ -262,6 +262,16 @@ assert('String#chop', '15.2.10.5.11') do
   assert_equal 'abc', c
 end
 
+assert('String#chop', '15.2.10.5.11') do
+  a = ''.chop
+  b = 'あいう'.chop
+  c = 'あ\nい'.chop.chop
+
+  assert_equal '', a
+  assert_equal 'あい', b
+  assert_equal 'あ', c
+end if UTF8STRING
+
 assert('String#chop!', '15.2.10.5.12') do
   a = ''
   b = 'abc'
@@ -273,6 +283,21 @@ assert('String#chop!', '15.2.10.5.12') do
   assert_equal b, 'ab'
 end
 
+assert('String#chop!', '15.2.10.5.12') do
+  a = ''
+  b = 'あいうえ\n'
+  c = 'あいうえ\n'
+
+  a.chop!
+  b.chop!
+  c.chop!
+  c.chop!
+
+  assert_equal a, ''
+  assert_equal b, 'あいうえ'
+  assert_equal c, 'あいう'
+end if UTF8STRING
+
 assert('String#downcase', '15.2.10.5.13') do
   a = 'ABC'.downcase
   b = 'ABC'