summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--generic/tclUtf.c27
-rw-r--r--tests/utf.test6
2 files changed, 19 insertions, 14 deletions
diff --git a/generic/tclUtf.c b/generic/tclUtf.c
index 52b4291..b13ad75 100644
--- a/generic/tclUtf.c
+++ b/generic/tclUtf.c
@@ -465,7 +465,6 @@ Tcl_NumUtfChars(
* for strlen(string). */
{
Tcl_UniChar ch;
- register Tcl_UniChar *chPtr = &ch;
register int i;
/*
@@ -478,23 +477,25 @@ Tcl_NumUtfChars(
i = 0;
if (length < 0) {
while (*src != '\0') {
- src += TclUtfToUniChar(src, chPtr);
+ src += TclUtfToUniChar(src, &ch);
i++;
}
+ if (i < 0) i = INT_MAX; /* Bug [2738427] */
} else {
- register int n;
-
- while (length > 0) {
- if (UCHAR(*src) < 0xC0) {
- length--;
- src++;
- } else {
- n = Tcl_UtfToUniChar(src, chPtr);
- length -= n;
- src += n;
- }
+ register const char *endPtr = src + length - TCL_UTF_MAX;
+
+ while (src < endPtr) {
+ src += TclUtfToUniChar(src, &ch);
i++;
}
+ endPtr += TCL_UTF_MAX;
+ while ((src < endPtr) && Tcl_UtfCharComplete(src, endPtr - src)) {
+ src += TclUtfToUniChar(src, &ch);
+ i++;
+ }
+ if (src < endPtr) {
+ i += endPtr - src;
+ }
}
return i;
}
diff --git a/tests/utf.test b/tests/utf.test
index f677438..422ab08 100644
--- a/tests/utf.test
+++ b/tests/utf.test
@@ -110,9 +110,13 @@ test utf-4.7 {Tcl_NumUtfChars: long string, calc len} {testnumutfchars testbytes
test utf-4.8 {Tcl_NumUtfChars: #u0000, calc len} {testnumutfchars testbytestring} {
testnumutfchars [testbytestring "\xC0\x80"] 2
} {1}
-test utf-4.9 {Tcl_NumUtfChars: #u20AC, calc len, incomplete} {knownBug testnumutfchars testbytestring} {
+# Bug [2738427]: Tcl_NumUtfChars(...) no overflow check
+test utf-4.9 {Tcl_NumUtfChars: #u20AC, calc len, incomplete} {testnumutfchars testbytestring} {
testnumutfchars [testbytestring "\xE2\x82\xAC"] 2
} {2}
+test utf-4.10 {Tcl_NumUtfChars: #u0000, calc len, overcomplete} {testnumutfchars testbytestring} {
+ testnumutfchars [testbytestring "\x00"] 2
+} {2}
test utf-5.1 {Tcl_UtfFindFirsts} {
} {}