diff options
| -rw-r--r-- | NEWS | 2 | ||||
| -rw-r--r-- | ext/standard/html.c | 6 | ||||
| -rwxr-xr-x | ext/standard/tests/strings/htmlentities-utf-2.phpt | 4 | ||||
| -rw-r--r-- | ext/standard/tests/strings/htmlentities-utf-3.phpt | 83 | ||||
| -rwxr-xr-x | ext/standard/tests/strings/htmlentities-utf.phpt | 4 |
5 files changed, 92 insertions, 7 deletions
@@ -33,6 +33,8 @@ and when there was data in the buffer before the emulation started. Also made more consistent its behavior -- should return failure every time less data than was requested was skipped. (Gustavo) +- Fixed htmlentities/htmlspecialchars accepting certain ill-formed UTF-8 + sequences. (Gustavo) - Fixed bug #53021 (In html_entity_decode, failure to convert numeric entities with ENT_NOQUOTES and ISO-8859-1). Fixed and extended the fix of ENT_NOQUOTES diff --git a/ext/standard/html.c b/ext/standard/html.c index 9cc17836a1..8994f4809e 100644 --- a/ext/standard/html.c +++ b/ext/standard/html.c @@ -540,7 +540,7 @@ inline static unsigned int get_next_char(enum entity_charset charset, MB_WRITE(c); this_char = c; pos++; - } else if (c < 0xc0) { + } else if (c < 0xc2) { MB_FAILURE(pos); } else if (c < 0xe0) { CHECK_LEN(pos, 2); @@ -572,7 +572,7 @@ inline static unsigned int get_next_char(enum entity_charset charset, MB_WRITE((unsigned char)str[pos + 1]); MB_WRITE((unsigned char)str[pos + 2]); pos += 3; - } else if (c < 0xf8) { + } else if (c < 0xf5) { CHECK_LEN(pos, 4); if (str[pos + 1] < 0x80 || str[pos + 1] > 0xbf) { MB_FAILURE(pos); @@ -584,7 +584,7 @@ inline static unsigned int get_next_char(enum entity_charset charset, MB_FAILURE(pos); } this_char = ((c & 0x07) << 18) | ((str[pos + 1] & 0x3f) << 12) | ((str[pos + 2] & 0x3f) << 6) | (str[pos + 3] & 0x3f); - if (this_char < 0x10000) { + if (this_char < 0x10000 || this_char > 0x10FFFF) { MB_FAILURE(pos); } MB_WRITE((unsigned char)c); diff --git a/ext/standard/tests/strings/htmlentities-utf-2.phpt b/ext/standard/tests/strings/htmlentities-utf-2.phpt index c5f4ac4ea6..d515dc0ff1 100755 --- a/ext/standard/tests/strings/htmlentities-utf-2.phpt +++ b/ext/standard/tests/strings/htmlentities-utf-2.phpt @@ -50,8 +50,8 @@ foreach($strings as $string) { %unicode|string%(16) "266561637574653b" %unicode|string%(2) "79" %unicode|string%(2) "79" -%unicode|string%(8) "f7bfbfbf" -%unicode|string%(8) "f7bfbfbf" +%unicode|string%(0) "" +%unicode|string%(0) "" %unicode|string%(0) "" %unicode|string%(0) "" %unicode|string%(0) "" diff --git a/ext/standard/tests/strings/htmlentities-utf-3.phpt b/ext/standard/tests/strings/htmlentities-utf-3.phpt new file mode 100644 index 0000000000..c28917ba16 --- /dev/null +++ b/ext/standard/tests/strings/htmlentities-utf-3.phpt @@ -0,0 +1,83 @@ +--TEST--
+Test get_next_char(), used by htmlentities()/htmlspecialchars(): validity of UTF-8 sequences
+--FILE--
+<?php
+
+/* conformance to Unicode 5.2, section 3.9, D92 */
+
+$val_ranges = array(
+ array(array(0x00, 0x7F)),
+ array(array(0xC2, 0xDF), array(0x80, 0xBF)),
+ array(array(0xE0, 0xE0), array(0xA0, 0xBF), array(0x80, 0xBF)),
+ array(array(0xE1, 0xEC), array(0x80, 0xBF), array(0x80, 0xBF)),
+ array(array(0xED, 0xED), array(0x80, 0x9F), array(0x80, 0xBF)),
+ array(array(0xEE, 0xEF), array(0x80, 0xBF), array(0x80, 0xBF)),
+ array(array(0xF0, 0xF0), array(0x90, 0xBF), array(0x80, 0xBF), array(0x80, 0xBF)),
+ array(array(0xF1, 0xF3), array(0x80, 0xBF), array(0x80, 0xBF), array(0x80, 0xBF)),
+ array(array(0xF4, 0xF4), array(0x80, 0x8F), array(0x80, 0xBF), array(0x80, 0xBF)),
+);
+
+function is_valid($seq) {
+ global $val_ranges;
+ $b = ord($seq[0]);
+ foreach ($val_ranges as $l) {
+ if ($b >= $l[0][0] && $b <= $l[0][1]) {
+ if (count($l) != strlen($seq)) {
+ return false;
+ }
+ for ($n = 1; $n < strlen($seq); $n++) {
+ if (ord($seq[$n]) < $l[$n][0] || ord($seq[$n]) > $l[$n][1]) {
+ return false;
+ }
+ }
+ return true;
+ }
+ }
+ return false;
+}
+
+function concordance($s) {
+ $vhe = strlen(htmlspecialchars($s, ENT_QUOTES, "UTF-8")) > 0;
+ $v = is_valid($s);
+ return ($vhe === $v);
+}
+
+for ($b1 = 0xC0; $b1 < 0xE0; $b1++) {
+ for ($b2 = 0x80; $b2 < 0xBF; $b2++) {
+ $s = chr($b1).chr($b2);
+ if (!concordance($s))
+ echo "Discordance for ".bin2hex($s),"\n";
+ }
+}
+
+
+for ($b1 = 0xE0; $b1 < 0xEF; $b1++) {
+ for ($b2 = 0x80; $b2 < 0xBF; $b2++) {
+ $s = chr($b1).chr($b2)."\x80";
+ if (!concordance($s))
+ echo "Discordance for ".bin2hex($s),"\n";
+ $s = chr($b1).chr($b2)."\xBF";
+ if (!concordance($s))
+ echo "Discordance for ".bin2hex($s),"\n";
+ }
+}
+
+for ($b1 = 0xF0; $b1 < 0xFF; $b1++) {
+ for ($b2 = 0x80; $b2 < 0xBF; $b2++) {
+ $s = chr($b1).chr($b2)."\x80\x80";
+ if (!concordance($s))
+ echo "Discordance for ".bin2hex($s),"\n";
+ $s = chr($b1).chr($b2)."\xBF\x80";
+ if (!concordance($s))
+ echo "Discordance for ".bin2hex($s),"\n";
+ $s = chr($b1).chr($b2)."\x80\xBF";
+ if (!concordance($s))
+ echo "Discordance for ".bin2hex($s),"\n";
+ $s = chr($b1).chr($b2)."\xBF\xBF";
+ if (!concordance($s))
+ echo "Discordance for ".bin2hex($s),"\n";
+ }
+}
+echo "Done.\n";
+--EXPECT--
+Done.
diff --git a/ext/standard/tests/strings/htmlentities-utf.phpt b/ext/standard/tests/strings/htmlentities-utf.phpt index 1daafc61d8..6a66e4df45 100755 --- a/ext/standard/tests/strings/htmlentities-utf.phpt +++ b/ext/standard/tests/strings/htmlentities-utf.phpt @@ -50,8 +50,8 @@ foreach($strings as $string) { %unicode|string%(16) "266561637574653b" %unicode|string%(0) "" %unicode|string%(0) "" -%unicode|string%(8) "f7bfbfbf" -%unicode|string%(8) "f7bfbfbf" +%unicode|string%(0) "" +%unicode|string%(0) "" %unicode|string%(0) "" %unicode|string%(0) "" %unicode|string%(0) "" |