From 634b5459db03b4cd00539ec45b6a3b23f29481de Mon Sep 17 00:00:00 2001 From: David Robillard Date: Mon, 24 Jan 2011 23:59:03 +0000 Subject: Remove support for 5 and 6 byte UTF-8 characters. The UTF-8 spec supports this, but I can't even find real characters to test it with. There don't seem to be actual Unicode code points this high yet. If it can't be tested, it can't be useful. Gone. git-svn-id: http://svn.drobilla.net/serd/trunk@55 490d8e77-9747-427b-9fa3-0b8f29cee8a0 --- src/reader.c | 16 ---------------- src/writer.c | 8 -------- tests/UTF-8.ttl | 5 ++--- 3 files changed, 2 insertions(+), 27 deletions(-) diff --git a/src/reader.c b/src/reader.c index 773dd99b..3a4ef452 100644 --- a/src/reader.c +++ b/src/reader.c @@ -349,10 +349,6 @@ read_hex_escape(SerdReader reader, unsigned length, Ref dest) size = 3; } else if (c < 0x00200000) { size = 4; - } else if (c < 0x04000000) { - size = 5; - } else if (c < 0x80000000) { - size = 6; } else { return false; } @@ -360,14 +356,6 @@ read_hex_escape(SerdReader reader, unsigned length, Ref dest) // Build output in buf // (Note # of bytes = # of leading 1 bits in first byte) switch (size) { - case 6: - buf[5] = 0x80 | (uint8_t)(c & 0x3F); - c >>= 6; - c |= (4 << 24); // set bit 2 - case 5: - buf[4] = 0x80 | (uint8_t)(c & 0x3F); - c >>= 6; - c |= (8 << 18); // set bit 3 case 4: buf[3] = 0x80 | (uint8_t)(c & 0x3F); c >>= 6; @@ -488,10 +476,6 @@ read_character(SerdReader reader, Ref dest) size = 3; } else if ((c & 0xF8) == 0xF0) { // Starts with `11110' size = 4; - } else if ((c & 0xFC) == 0xF8) { // Starts with `111110' - size = 5; - } else if ((c & 0xFE) == 0xFC) { // Starts with `1111110' - size = 6; } else { error(reader, "invalid character\n"); return SERD_ERROR; diff --git a/src/writer.c b/src/writer.c index 1de08a4c..d7d38baa 100644 --- a/src/writer.c +++ b/src/writer.c @@ -101,12 +101,6 @@ write_text(SerdWriter writer, TextContext ctx, } else if ((in & 0xF8) == 0xF0) { // Starts with `11110' size = 4; c = in & 0x07; - } else if ((in & 0xFC) == 0xF8) { // Starts with `111110' - size = 5; - c = in & 0x03; - } else if ((in & 0xFE) == 0xFC) { // Starts with `1111110' - size = 6; - c = in & 0x01; } else { fprintf(stderr, "invalid UTF-8 at offset %zu: %X\n", i, in); return false; @@ -128,8 +122,6 @@ write_text(SerdWriter writer, TextContext ctx, } while (0) switch (size) { - case 6: READ_BYTE(); - case 5: READ_BYTE(); case 4: READ_BYTE(); case 3: READ_BYTE(); case 2: READ_BYTE(); diff --git a/tests/UTF-8.ttl b/tests/UTF-8.ttl index 66be456e..9e9a7b33 100644 --- a/tests/UTF-8.ttl +++ b/tests/UTF-8.ttl @@ -214,7 +214,6 @@ Box drawing alignment tests: █ ▝▀▘▙▄▟ """ . <> rdfs:comment """ - 5 byte character: \U3FFFFFF - 6 byte character: \U7FFFFFFF + Two byte Unicode escape: \U00E0 + Largest Unicode escape in Turtle: \U0010FFFF """ . - -- cgit v1.2.1