diff options
| -rw-r--r-- | generic/tclUtf.c | 19 | ||||
| -rw-r--r-- | tests/utf.test | 3 |
2 files changed, 13 insertions, 9 deletions
diff --git a/generic/tclUtf.c b/generic/tclUtf.c index 4147bbc..80f3be8 100644 --- a/generic/tclUtf.c +++ b/generic/tclUtf.c @@ -548,17 +548,18 @@ Tcl_NumUtfChars( i++; } /* Loop over the remaining string where call must happen */ - while ((src < endPtr) && Tcl_UtfCharComplete(src, endPtr - src)) { - src += TclUtfToUniChar(src, &ch); + while (src < endPtr) { + if (Tcl_UtfCharComplete(src, endPtr - src)) { + src += TclUtfToUniChar(src, &ch); + } else { + /* + * src points to incomplete UTF-8 sequence + * Treat first byte as character and count it + */ + src++; + } i++; } - if (src < endPtr) { - /* - * String ends in an incomplete UTF-8 sequence. - * Count every byte in it. - */ - i += endPtr - src; - } } return i; } diff --git a/tests/utf.test b/tests/utf.test index 8aa3757..f48299d 100644 --- a/tests/utf.test +++ b/tests/utf.test @@ -175,6 +175,9 @@ test utf-4.12.1 {Tcl_NumUtfChars: #4-byte UTF-8 character} {testnumutfchars test test utf-4.12.2 {Tcl_NumUtfChars: #4-byte UTF-8 character} {testnumutfchars testbytestring utf16} { testnumutfchars [testbytestring \xF0\x9F\x92\xA9] end } 2 +test utf-4.13 {Tcl_NumUtfChars: end of string} {testnumutfchars testbytestring} { + testnumutfchars foobar[testbytestring \xF2\xC2\xA0] end +} 8 test utf-5.1 {Tcl_UtfFindFirst} {testfindfirst testbytestring} { testfindfirst [testbytestring abcbc] 98 |