From 48e70046368e2fa24d4f0080af77df6b2d2b31c5 Mon Sep 17 00:00:00 2001 From: Nigel Tao Date: Sat, 22 Feb 2020 08:07:26 +1100 Subject: [PATCH] Add std/json character classes --- release/c/wuffs-unsupported-snapshot.c | 56 ++++++++++++++-------- std/json/common_consts.wuffs | 66 ++++++++++++++++++++++++++ std/json/decode_json.wuffs | 34 ++++++------- 3 files changed, 116 insertions(+), 40 deletions(-) diff --git a/release/c/wuffs-unsupported-snapshot.c b/release/c/wuffs-unsupported-snapshot.c index 06ec1d0b..e774b415 100644 --- a/release/c/wuffs-unsupported-snapshot.c +++ b/release/c/wuffs-unsupported-snapshot.c @@ -18458,6 +18458,26 @@ static const uint8_t // 0, 0, 0, 0, 0, 0, 0, 0, 0, }; +static const uint8_t // + wuffs_json__classes[256] // + WUFFS_BASE__POTENTIALLY_UNUSED = { + 15, 15, 15, 15, 15, 15, 15, 15, 15, 0, 0, 15, 15, 0, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 0, 15, 1, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 2, 4, 15, 15, 4, 4, 4, 4, 4, 4, + 4, 4, 4, 4, 3, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 7, 15, 8, 15, 15, 15, 15, 15, 15, 15, 15, 9, 15, 15, 15, 15, 15, + 15, 15, 11, 15, 15, 15, 15, 15, 10, 15, 15, 15, 15, 15, 15, 5, 15, 6, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, + 15, 15, 15, 15, +}; + static const uint8_t // wuffs_json__hexadecimal_digits[256] // WUFFS_BASE__POTENTIALLY_UNUSED = { @@ -18585,6 +18605,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, uint32_t v_stack_bit = 0; uint32_t v_match = 0; uint8_t v_c = 0; + uint8_t v_class = 0; uint8_t v_backslash = 0; uint8_t v_uni4_ok = 0; uint32_t v_uni4_string = 0; @@ -18637,6 +18658,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, } v_whitespace_length = 0; v_c = 0; + v_class = 0; while (true) { if (((uint64_t)(io2_a_src - iop_a_src)) <= 0) { if (v_whitespace_length > 0) { @@ -18657,17 +18679,9 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, goto label_0_continue; } v_c = wuffs_base__load_u8be(iop_a_src); - if (v_c > 32) { + v_class = wuffs_json__classes[v_c]; + if (v_class != 0) { goto label_1_break; - } else if ((v_c != 32) && (v_c != 10) && (v_c != 9) && (v_c != 13)) { - if (v_whitespace_length > 0) { - *iop_a_dst++ = wuffs_base__make_token( - (((uint64_t)(0)) << WUFFS_BASE__TOKEN__VALUE__SHIFT) | - (((uint64_t)(((uint64_t)(v_whitespace_length)))) - << WUFFS_BASE__TOKEN__LENGTH__SHIFT)); - } - status = wuffs_base__make_status(wuffs_json__error__bad_input); - goto exit; } (iop_a_src += 1, wuffs_base__make_empty_struct()); if (v_whitespace_length >= 65534) { @@ -18691,7 +18705,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, } } if (v_c <= 57) { - if (v_c == 34) { + if (v_class == 1) { if (0 == (v_expect & 2)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -18888,7 +18902,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, goto label_0_continue; } goto label_5_break; - } else if (v_c == 44) { + } else if (v_class == 2) { if (0 == (v_expect & 4)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -18903,7 +18917,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, v_expect = 3; } goto label_0_continue; - } else if ((v_c >= 48) || (v_c == 45)) { + } else if (v_class == 4) { if (0 == (v_expect & 1)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -18961,7 +18975,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, label_6_break:; goto label_5_break; } - } else if (v_c == 58) { + } else if (v_class == 3) { if (0 == (v_expect & 8)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -18972,7 +18986,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, (((uint64_t)(1)) << WUFFS_BASE__TOKEN__LENGTH__SHIFT)); v_expect = 3; goto label_0_continue; - } else if (v_c == 91) { + } else if (v_class == 7) { if (0 == (v_expect & 1)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -18994,7 +19008,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, v_expect = 19; v_expect_after_value = 20; goto label_0_continue; - } else if (v_c == 93) { + } else if (v_class == 8) { if (0 == (v_expect & 16)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -19024,7 +19038,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, v_expect_after_value = 36; } goto label_0_continue; - } else if (v_c == 123) { + } else if (v_class == 5) { if (0 == (v_expect & 1)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -19046,7 +19060,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, v_expect = 34; v_expect_after_value = 36; goto label_0_continue; - } else if (v_c == 125) { + } else if (v_class == 6) { if (0 == (v_expect & 32)) { status = wuffs_base__make_status(wuffs_json__error__bad_input); goto exit; @@ -19082,7 +19096,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, goto exit; } v_match = 2; - if (v_c == 110) { + if (v_class == 11) { v_match = wuffs_base__io_reader__match7(iop_a_src, io2_a_src, 465676103172); if (v_match == 0) { @@ -19097,7 +19111,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, (iop_a_src += 4, wuffs_base__make_empty_struct()); goto label_5_break; } - } else if (v_c == 102) { + } else if (v_class == 9) { v_match = wuffs_base__io_reader__match7(iop_a_src, io2_a_src, 111546413966853); if (v_match == 0) { @@ -19112,7 +19126,7 @@ wuffs_json__decoder__decode_tokens(wuffs_json__decoder* self, (iop_a_src += 5, wuffs_base__make_empty_struct()); goto label_5_break; } - } else if (v_c == 116) { + } else if (v_class == 10) { v_match = wuffs_base__io_reader__match7(iop_a_src, io2_a_src, 435762131972); if (v_match == 0) { diff --git a/std/json/common_consts.wuffs b/std/json/common_consts.wuffs index b7f2c347..239b562c 100644 --- a/std/json/common_consts.wuffs +++ b/std/json/common_consts.wuffs @@ -19,6 +19,8 @@ pub status "#unsupported recursion depth" pri status "#internal error: inconsistent I/O" pri const backslashes array[256] base.u8 = [ + // 0 1 2 3 4 5 6 7 + // 8 9 A B C D E F 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0x00 ..= 0x07. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0x08 ..= 0x0F. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0x10 ..= 0x17. @@ -54,9 +56,71 @@ pri const backslashes array[256] base.u8 = [ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0xE8 ..= 0xEF. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0xF0 ..= 0xF7. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0xF8 ..= 0xFF. + // 0 1 2 3 4 5 6 7 + // 8 9 A B C D E F +] + +// - 0x00 is CLASS_WHITESPACE. +// - 0x01 is CLASS_STRING. +// - 0x02 is CLASS_COMMA. +// - 0x03 is CLASS_COLON. +// - 0x04 is CLASS_NUMBER. +// - 0x05 is CLASS_OPEN_CURLY_BRACE. +// - 0x06 is CLASS_CLOSE_CURLY_BRACE. +// - 0x07 is CLASS_OPEN_SQUARE_BRACKET. +// - 0x08 is CLASS_CLOSE_SQUARE_BRACKET. +// - 0x09 is CLASS_FALSE. +// - 0x0A is CLASS_TRUE. +// - 0x0B is CLASS_NULL. +// - 0x0C is reserved. +// - 0x0D is reserved. +// - 0x0E is reserved. +// - 0x0F is CLASS_BAD_INPUT. +pri const classes array[256] base.u8[..= 0x0F] = [ + // 0 1 2 3 4 5 6 7 + // 8 9 A B C D E F + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x00 ..= 0x07. + 0x0F, 0x00, 0x00, 0x0F, 0x0F, 0x00, 0x0F, 0x0F, // 0x08 ..= 0x0F. '\t', '\n', '\r'. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x10 ..= 0x17. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x18 ..= 0x1F. + 0x00, 0x0F, 0x01, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x20 ..= 0x27. ' ', '"'. + 0x0F, 0x0F, 0x0F, 0x0F, 0x02, 0x04, 0x0F, 0x0F, // 0x28 ..= 0x2F. ',', '-'. + 0x04, 0x04, 0x04, 0x04, 0x04, 0x04, 0x04, 0x04, // 0x30 ..= 0x37. '0'-'7'. + 0x04, 0x04, 0x03, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x38 ..= 0x3F. '8'-'9', ':'. + + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x40 ..= 0x47. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x48 ..= 0x4F. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x50 ..= 0x57. + 0x0F, 0x0F, 0x0F, 0x07, 0x0F, 0x08, 0x0F, 0x0F, // 0x58 ..= 0x5F. '[', ']'. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x09, 0x0F, // 0x60 ..= 0x67. 'f'. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0B, 0x0F, // 0x68 ..= 0x6F. 'n'. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0A, 0x0F, 0x0F, 0x0F, // 0x70 ..= 0x77. 't'. + 0x0F, 0x0F, 0x0F, 0x05, 0x0F, 0x06, 0x0F, 0x0F, // 0x78 ..= 0x7F. '{', '}'. + + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x80 ..= 0x87. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x88 ..= 0x8F. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x90 ..= 0x97. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0x98 ..= 0x9F. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xA0 ..= 0xA7. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xA8 ..= 0xAF. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xB0 ..= 0xB7. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xB8 ..= 0xBF. + + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xC0 ..= 0xC7. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xC8 ..= 0xCF. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xD0 ..= 0xD7. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xD8 ..= 0xDF. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xE0 ..= 0xE7. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xE8 ..= 0xEF. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xF0 ..= 0xF7. + 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, 0x0F, // 0xF8 ..= 0xFF. + // 0 1 2 3 4 5 6 7 + // 8 9 A B C D E F ] pri const hexadecimal_digits array[256] base.u8 = [ + // 0 1 2 3 4 5 6 7 + // 8 9 A B C D E F 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0x00 ..= 0x07. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0x08 ..= 0x0F. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0x10 ..= 0x17. @@ -92,4 +156,6 @@ pri const hexadecimal_digits array[256] base.u8 = [ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0xE8 ..= 0xEF. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0xF0 ..= 0xF7. 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0xF8 ..= 0xFF. + // 0 1 2 3 4 5 6 7 + // 8 9 A B C D E F ] diff --git a/std/json/decode_json.wuffs b/std/json/decode_json.wuffs index a6a787a8..5eb9c770 100644 --- a/std/json/decode_json.wuffs +++ b/std/json/decode_json.wuffs @@ -57,6 +57,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { var stack_bit : base.u32[..= 31] var match : base.u32[..= 2] var c : base.u8 + var class : base.u8[..= 0x0F] var backslash : base.u8 var uni4_ok : base.u8 var uni4_string : base.u32 @@ -96,6 +97,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { // Consume whitespace. whitespace_length = 0 c = 0 + class = 0 while.ws true, inv args.dst.available() > 0, post args.src.available() > 0, @@ -116,15 +118,9 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { } c = args.src.peek_u8() - if c > 0x20 { // 0x20 is ' '. + class = classes[c] + if class <> 0x00 { // 0x00 is CLASS_WHITESPACE. break.ws - } else if (c <> 0x20) and (c <> 0x0A) and (c <> 0x09) and (c <> 0x0D) { - if whitespace_length > 0 { - args.dst.write_fast_token!( - value: 0, - length: whitespace_length as base.u64) - } - return "#bad input" } args.src.skip32_fast!(actual: 1, worst_case: 1) @@ -157,7 +153,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { assert args.src.available() > 0 if c <= 0x39 { // 0x39 is '9'. - if c == 0x22 { // 0x22 is '"'. + if class == 0x01 { // 0x01 is CLASS_STRING. // -------- BEGIN parse strings. if 0 == (expect & 0x02) { // 0x02 is EXPECT_STRING_VALUE. return "#bad input" @@ -333,7 +329,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { break.goto_parsed_a_leaf_value // -------- END parse strings. - } else if c == 0x2C { // 0x2C is ','. + } else if class == 0x02 { // 0x02 is CLASS_COMMA. if 0 == (expect & 0x04) { // 0x01 is EXPECT_COMMA. return "#bad input" } @@ -351,7 +347,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { } continue.outer - } else if (c >= 0x30) or (c == 0x2D) { // 0x30 is '0', 0x2D is '-'. + } else if class == 0x04 { // 0x04 is CLASS_NUMBER. // -------- BEGIN parse numbers. if 0 == (expect & 0x01) { // 0x02 is EXPECT_NON_STRING_VALUE. return "#bad input" @@ -399,7 +395,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { // -------- END parse numbers. } - } else if c == 0x3A { // 0x3A is ':'. + } else if class == 0x03 { // 0x03 is CLASS_COLON. if 0 == (expect & 0x08) { // 0x08 is EXPECT_COLON. return "#bad input" } @@ -411,7 +407,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { expect = 0x03 // 0x03 is EXPECT_VALUE. continue.outer - } else if c == 0x5B { // 0x5B is '['. + } else if class == 0x07 { // 0x07 is CLASS_OPEN_SQUARE_BRACKET. if 0 == (expect & 0x01) { // 0x01 is EXPECT_NON_STRING_VALUE. return "#bad input" } @@ -431,7 +427,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { expect_after_value = 0x14 // 0x14 is (EXPECT_CLOSE_BRACKET | EXPECT_COMMA). continue.outer - } else if c == 0x5D { // 0x5D is ']'. + } else if class == 0x08 { // 0x08 is CLASS_CLOSE_SQUARE_BRACKET. if 0 == (expect & 0x10) { // 0x10 is EXPECT_CLOSE_BRACKET return "#bad input" } @@ -462,7 +458,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { } continue.outer - } else if c == 0x7B { // 0x7B is '{'. + } else if class == 0x05 { // 0x05 is CLASS_OPEN_CURLY_BRACE. if 0 == (expect & 0x01) { // 0x01 is EXPECT_NON_STRING_VALUE. return "#bad input" } @@ -482,7 +478,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { expect_after_value = 0x24 // 0x24 is (EXPECT_CLOSE_BRACE | EXPECT_COMMA). continue.outer - } else if c == 0x7D { // 0x7D is '}'. + } else if class == 0x06 { // 0x06 is CLASS_CLOSE_CURLY_BRACE. if 0 == (expect & 0x20) { // 0x20 is EXPECT_CLOSE_BRACE. return "#bad input" } @@ -518,7 +514,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { return "#bad input" } match = 2 - if c == 0x6E { // 0x6E is 'n'. + if class == 0x0B { // 0x0B is CLASS_NULL. match = args.src.match7(a: 0x6C_6C75_6E04) // 4 bytes "null". if match == 0 { args.dst.write_fast_token!( @@ -531,7 +527,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { break.goto_parsed_a_leaf_value } - } else if c == 0x66 { // 0x66 is 'f'. + } else if class == 0x09 { // 0x09 is CLASS_FALSE. match = args.src.match7(a: 0x6573_6C61_6605) // 5 bytes "false". if match == 0 { args.dst.write_fast_token!( @@ -544,7 +540,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader) { break.goto_parsed_a_leaf_value } - } else if c == 0x74 { // 0x74 is 't'. + } else if class == 0x0A { // 0x0A is CLASS_TRUE. match = args.src.match7(a: 0x65_7572_7404) // 4 bytes "true". if match == 0 { args.dst.write_fast_token!(