aboutsummaryrefslogtreecommitdiffstats
path: root/erts/emulator/beam/erl_unicode.c
diff options
context:
space:
mode:
authorBjörn Gustavsson <[email protected]>2011-08-30 11:51:11 +0200
committerBjörn Gustavsson <[email protected]>2011-10-13 14:16:00 +0200
commit34db76765561487e526fe66d3d19ecf3b3fb9dc8 (patch)
tree9141e3c5729e46d03c8b27b14da3b29b1e54abca /erts/emulator/beam/erl_unicode.c
parent6ca6dd3c670fb8185ebb9a20c2a731a7375c1cac (diff)
downloadotp-34db76765561487e526fe66d3d19ecf3b3fb9dc8.tar.gz
otp-34db76765561487e526fe66d3d19ecf3b3fb9dc8.tar.bz2
otp-34db76765561487e526fe66d3d19ecf3b3fb9dc8.zip
Allow noncharacter code points in unicode encoding and decoding
The two noncharacter code points 16#FFFE and 16#FFFF were not allowed to be encoded or decoded using the unicode module or bit syntax. That causes an inconsistency, since the noncharacters 16#FDD0 to 16#FDEF could be encoded/decoded. There is two ways to fix that inconsistency. We have chosen to allow 16#FFFE and 16#FFFF to be encoded and decoded, because the noncharacters could be useful internally within an application and it will make encoding and decoding slightly faster. Reported-by: Alisdair Sullivan
Diffstat (limited to 'erts/emulator/beam/erl_unicode.c')
-rw-r--r--erts/emulator/beam/erl_unicode.c24
1 files changed, 5 insertions, 19 deletions
diff --git a/erts/emulator/beam/erl_unicode.c b/erts/emulator/beam/erl_unicode.c
index 158eb361a4..bd5f3cc4c1 100644
--- a/erts/emulator/beam/erl_unicode.c
+++ b/erts/emulator/beam/erl_unicode.c
@@ -348,12 +348,6 @@ static int copy_utf8_bin(byte *target, byte *source, Uint size,
return copied;
}
- if (((*source) == 0xEF) && (source[1] == 0xBF) &&
- ((source[2] == 0xBE) || (source[2] == 0xBF))) {
- *err_pos = source;
- return copied;
- }
-
*(target++) = *(source++);
*(target++) = *(source++);
*(target++) = *(source++);
@@ -714,9 +708,8 @@ L_Again: /* Restart with sublist, old listend was pushed on stack */
target[(*pos)++] = (((byte) (x & 0x3F)) |
((byte) 0x80));
} else if (x < 0x10000) {
- if ((x >= 0xD800 && x <= 0xDFFF) ||
- (x == 0xFFFE) ||
- (x == 0xFFFF)) { /* Invalid unicode range */
+ if (x >= 0xD800 && x <= 0xDFFF) {
+ /* Invalid unicode range */
*err = 1;
goto done;
}
@@ -1230,10 +1223,6 @@ int erts_analyze_utf8(byte *source, Uint size,
((source[1] & 0x20) != 0)) {
return ERTS_UTF8_ERROR;
}
- if (((*source) == 0xEF) && (source[1] == 0xBF) &&
- ((source[2] == 0xBE) || (source[2] == 0xBF))) {
- return ERTS_UTF8_ERROR;
- }
source += 3;
size -= 3;
} else if (((*source) & ((byte) 0xF8)) == 0xF0) {
@@ -2166,9 +2155,8 @@ L_Again: /* Restart with sublist, old listend was pushed on stack */
} else if (x < 0x800) {
need += 2;
} else if (x < 0x10000) {
- if ((x >= 0xD800 && x <= 0xDFFF) ||
- (x == 0xFFFE) ||
- (x == 0xFFFF)) { /* Invalid unicode range */
+ if (x >= 0xD800 && x <= 0xDFFF) {
+ /* Invalid unicode range */
DESTROY_ESTACK(stack);
return ((Sint) -1);
}
@@ -2314,9 +2302,7 @@ L_Again: /* Restart with sublist, old listend was pushed on stack */
*p++ = (((byte) (x & 0x3F)) |
((byte) 0x80));
} else if (x < 0x10000) {
- ASSERT(!((x >= 0xD800 && x <= 0xDFFF) ||
- (x == 0xFFFE) ||
- (x == 0xFFFF)));
+ ASSERT(!(x >= 0xD800 && x <= 0xDFFF));
*p++ = (((byte) (x >> 12)) |
((byte) 0xE0));
*p++ = ((((byte) (x >> 6)) & 0x3F) |