diff options
author | Scott MacVicar <scottmac@php.net> | 2009-01-02 03:02:22 +0000 |
---|---|---|
committer | Scott MacVicar <scottmac@php.net> | 2009-01-02 03:02:22 +0000 |
commit | 0bdbc4e356d81fd2fd7e1f2391e24e76c2f3b8f6 (patch) | |
tree | b62b4615b40808b430f3b0c8f69fbe5aa416c3b3 /ext/json | |
parent | 8d62f3dd0201094f589fa8c64ab337e5edcf43e6 (diff) | |
download | php-git-0bdbc4e356d81fd2fd7e1f2391e24e76c2f3b8f6.tar.gz |
MFH Fix bug #46944 - UTF-8 characters outside the BMP aren't encoded correctly.
Diffstat (limited to 'ext/json')
-rw-r--r-- | ext/json/tests/bug46944.phpt | 32 | ||||
-rw-r--r-- | ext/json/utf8_decode.c | 2 | ||||
-rw-r--r-- | ext/json/utf8_to_utf16.c | 2 |
3 files changed, 34 insertions, 2 deletions
diff --git a/ext/json/tests/bug46944.phpt b/ext/json/tests/bug46944.phpt new file mode 100644 index 0000000000..735de04435 --- /dev/null +++ b/ext/json/tests/bug46944.phpt @@ -0,0 +1,32 @@ +--TEST-- +Bug #46944 (json_encode() doesn't handle 3 byte utf8 correctly) +--SKIPIF-- +<?php if (!extension_loaded('json')) print 'skip'; ?> +--FILE-- +<?php + +for ($i = 1; $i <= 16; $i++) { + echo json_encode(b"aa" . (0xf0|($i >> 2)) . (0x8f|($i & 3) << 4) . "\xbf\xbdzz") . "\n"; +} + + +echo "Done\n"; +?> +--EXPECT-- +"aa\ud83f\udffdzz" +"aa\ud87f\udffdzz" +"aa\ud8bf\udffdzz" +"aa\ud8ff\udffdzz" +"aa\ud93f\udffdzz" +"aa\ud97f\udffdzz" +"aa\ud9bf\udffdzz" +"aa\ud9ff\udffdzz" +"aa\uda3f\udffdzz" +"aa\uda7f\udffdzz" +"aa\udabf\udffdzz" +"aa\udaff\udffdzz" +"aa\udb3f\udffdzz" +"aa\udb7f\udffdzz" +"aa\udbbf\udffdzz" +"aa\udbff\udffdzz" +Done diff --git a/ext/json/utf8_decode.c b/ext/json/utf8_decode.c index cea1f8cec8..2d0422bedb 100644 --- a/ext/json/utf8_decode.c +++ b/ext/json/utf8_decode.c @@ -165,7 +165,7 @@ utf8_decode_next(json_utf8_decode *utf8) /* Three continuation (65536 to 1114111) */ - if ((c & 0xF1) == 0xF0) { + if ((c & 0xF8) == 0xF0) { int c1 = cont(utf8); int c2 = cont(utf8); int c3 = cont(utf8); diff --git a/ext/json/utf8_to_utf16.c b/ext/json/utf8_to_utf16.c index 42ea9e5d8e..599f0e13b4 100644 --- a/ext/json/utf8_to_utf16.c +++ b/ext/json/utf8_to_utf16.c @@ -46,7 +46,7 @@ utf8_to_utf16(unsigned short w[], char p[], int length) w[the_index] = (unsigned short)c; the_index += 1; } else { - c &= 0xFFFF; + c -= 0x10000; w[the_index] = (unsigned short)(0xD800 | (c >> 10)); the_index += 1; w[the_index] = (unsigned short)(0xDC00 | (c & 0x3FF)); |