From 66b0a12b6c2403386225deaa9e3cc73124ea9599 Mon Sep 17 00:00:00 2001 From: Nigel Tao Date: Tue, 9 Jun 2020 23:48:54 +1000 Subject: [PATCH] Add json.EXPECT_ETC constants --- release/c/wuffs-unsupported-snapshot.c | 16 +++++++++ std/json/common_consts.wuffs | 50 +++++++++++++------------- std/json/decode_json.wuffs | 46 +++++++++++------------- 3 files changed, 62 insertions(+), 50 deletions(-) diff --git a/release/c/wuffs-unsupported-snapshot.c b/release/c/wuffs-unsupported-snapshot.c index af8aab72..80c3f4f6 100644 --- a/release/c/wuffs-unsupported-snapshot.c +++ b/release/c/wuffs-unsupported-snapshot.c @@ -21749,6 +21749,22 @@ WUFFS_JSON__LUT_CHARS[256]WUFFS_BASE__POTENTIALLY_UNUSED = { #define WUFFS_JSON__CLASS_COMMENT 12 +#define WUFFS_JSON__EXPECT_VALUE 7858 + +#define WUFFS_JSON__EXPECT_NON_STRING_VALUE 7856 + +#define WUFFS_JSON__EXPECT_STRING 4098 + +#define WUFFS_JSON__EXPECT_COMMA 4100 + +#define WUFFS_JSON__EXPECT_COLON 4104 + +#define WUFFS_JSON__EXPECT_NUMBER 4112 + +#define WUFFS_JSON__EXPECT_CLOSE_CURLY_BRACE 4160 + +#define WUFFS_JSON__EXPECT_CLOSE_SQUARE_BRACKET 4352 + static const uint8_t WUFFS_JSON__LUT_CLASSES[256]WUFFS_BASE__POTENTIALLY_UNUSED = { 15, 15, 15, 15, 15, 15, 15, 15, diff --git a/std/json/common_consts.wuffs b/std/json/common_consts.wuffs index 356f052e..7c5b445e 100644 --- a/std/json/common_consts.wuffs +++ b/std/json/common_consts.wuffs @@ -219,6 +219,30 @@ pri const CLASS_TRUE : base.u8 = 0x0A pri const CLASS_NULL_NAN_INF : base.u8 = 0x0B pri const CLASS_COMMENT : base.u8 = 0x0C +// EXPECT_ETC are unions of LUT_CLASSES bitmasks. See LUT_CLASSES below. +// +// Bitwise or'ing these together gives 0x1FFE. Whitespace is never expected, as +// it is handled separately. +// +// EXPECT_VALUE is also defined to be 0x1EB2, equivalent to (EXPECT_STRING | +// EXPECT_NON_STRING_VALUE). +// +// "Non-string value" includes literals (false, true, null), numbers, arrays +// and objects. +// +// "String value" includes "this" and "th\u0061t". +// +// Comments are always expected. Whether the relevant quirks are enabled are +// checked elsewhere. +pri const EXPECT_VALUE : base.u32 = 0x1EB2 +pri const EXPECT_NON_STRING_VALUE : base.u32 = 0x1EB0 +pri const EXPECT_STRING : base.u32 = 0x1002 +pri const EXPECT_COMMA : base.u32 = 0x1004 +pri const EXPECT_COLON : base.u32 = 0x1008 +pri const EXPECT_NUMBER : base.u32 = 0x1010 +pri const EXPECT_CLOSE_CURLY_BRACE : base.u32 = 0x1040 +pri const EXPECT_CLOSE_SQUARE_BRACKET : base.u32 = 0x1100 + // LUT_CLASSES is: // - 0x00 (bitmask 0x0001) is CLASS_WHITESPACE. // - 0x01 (bitmask 0x0002) is CLASS_STRING. @@ -238,31 +262,7 @@ pri const CLASS_COMMENT : base.u8 = 0x0C // - 0x0F (bitmask 0x8000) is CLASS_BAD_INPUT. // // The bitmasks are used by the "expect" variable: what the next character -// class can be. -// -// - 0x1002 is EXPECT_STRING. -// - 0x1EB0 is EXPECT_NON_STRING_VALUE. -// - 0x1004 is EXPECT_COMMA. -// - 0x1008 is EXPECT_COLON. -// - 0x1040 is EXPECT_CLOSE_CURLY_BRACE. -// - 0x1100 is EXPECT_CLOSE_SQUARE_BRACKET. -// -// Bitwise or'ing these together gives 0x0FFE. Whitespace is never expected, as -// it is handled separately. -// -// EXPECT_VALUE is also defined to be 0x1EB2, equivalent to (EXPECT_STRING | -// EXPECT_NON_STRING_VALUE). -// -// EXPECT_NUMBER is also defined to be 0x1010. Testing for EXPECT_NUMBER's -// presence is equivalent to testing for EXPECT_NON_STRING_VALUE. -// -// "Non-string value" includes literals (false, true, null), numbers, arrays -// and objects. -// -// "String value" includes "this" and "th\u0061t". -// -// Comments are always expected. Whether the relevant quirks are enabled are -// checked elsewhere. +// class can be. See EXPECT_ETC above. pri const LUT_CLASSES : array[256] base.u8[..= 0x0F] = [ // 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 ce639b78..aac36681 100644 --- a/std/json/decode_json.wuffs +++ b/std/json/decode_json.wuffs @@ -121,7 +121,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor this.decode_leading?(dst: args.dst, src: args.src) } - expect = 0x1EB2 // EXPECT_VALUE + expect = EXPECT_VALUE while.outer true { while.goto_parsed_a_leaf_value true {{ @@ -838,10 +838,10 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor } endwhile // As above, expect must have contained EXPECT_STRING. If it didn't - // also contain EXPECT_NUMBER then we were parsing an object key - // and the next token should be ':'. + // also contain EXPECT_NUMBER (excluding EXPECT_COMMENT) then we + // were parsing an object key and the next token should be ':'. if 0 == (expect & ((1 as base.u32) << CLASS_NUMBER)) { - expect = 0x1008 // 0x1008 is EXPECT_COLON. + expect = EXPECT_COLON continue.outer } break.goto_parsed_a_leaf_value @@ -859,15 +859,15 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor // array or an object. if 0 == (expect & ((1 as base.u32) << CLASS_CLOSE_SQUARE_BRACKET)) { if this.quirks[QUIRK_ALLOW_EXTRA_COMMA - QUIRKS_BASE] { - expect = 0x1042 // 0x1042 is EXPECT_STRING | EXPECT_CLOSE_CURLY_BRACE. + expect = EXPECT_STRING | EXPECT_CLOSE_CURLY_BRACE } else { - expect = 0x1002 // 0x1002 is EXPECT_STRING. + expect = EXPECT_STRING } } else { if this.quirks[QUIRK_ALLOW_EXTRA_COMMA - QUIRKS_BASE] { - expect = 0x1FB2 // 0x0FB2 is EXPECT_VALUE | EXPECT_CLOSE_SQUARE_BRACKET. + expect = EXPECT_VALUE | EXPECT_CLOSE_SQUARE_BRACKET } else { - expect = 0x1EB2 // 0x0EB2 is EXPECT_VALUE. + expect = EXPECT_VALUE } } continue.outer @@ -880,7 +880,7 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor value_minor: 0, continued: 0, length: 1) - expect = 0x1EB2 // 0x1EB2 is EXPECT_VALUE. + expect = EXPECT_VALUE continue.outer } else if class == CLASS_NUMBER { @@ -970,8 +970,8 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor value_minor: vminor, continued: 0, length: 1) - expect = 0x1042 // 0x1042 is (EXPECT_CLOSE_CURLY_BRACE | EXPECT_STRING). - expect_after_value = 0x1044 // 0x1044 is (EXPECT_CURLY_CLOSE_BRACE | EXPECT_COMMA). + expect = EXPECT_CLOSE_CURLY_BRACE | EXPECT_STRING + expect_after_value = EXPECT_CLOSE_CURLY_BRACE | EXPECT_COMMA continue.outer } else if class == CLASS_CLOSE_CURLY_BRACE { @@ -999,9 +999,8 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor base.TOKEN__VBD__STRUCTURE__TO_LIST, continued: 0, length: 1) - // 0x1104 is (EXPECT_SQUARE_CLOSE_BRACKET | EXPECT_COMMA). - expect = 0x1104 - expect_after_value = 0x1104 + expect = EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_COMMA + expect_after_value = EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_COMMA } else { args.dst.write_simple_token_fast!( value_major: 0, @@ -1011,9 +1010,8 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor base.TOKEN__VBD__STRUCTURE__TO_DICT, continued: 0, length: 1) - // 0x1044 is (EXPECT_CURLY_CLOSE_BRACE | EXPECT_COMMA). - expect = 0x1044 - expect_after_value = 0x1044 + expect = EXPECT_CLOSE_CURLY_BRACE | EXPECT_COMMA + expect_after_value = EXPECT_CLOSE_CURLY_BRACE | EXPECT_COMMA } continue.outer @@ -1049,8 +1047,8 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor value_minor: vminor, continued: 0, length: 1) - expect = 0x1FB2 // 0x1FB2 is (EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_VALUE). - expect_after_value = 0x1104 // 0x1104 is (EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_COMMA). + expect = EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_VALUE + expect_after_value = EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_COMMA continue.outer } else if class == CLASS_CLOSE_SQUARE_BRACKET { @@ -1078,9 +1076,8 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor base.TOKEN__VBD__STRUCTURE__TO_LIST, continued: 0, length: 1) - // 0x1104 is (EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_COMMA). - expect = 0x1104 - expect_after_value = 0x1104 + expect = EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_COMMA + expect_after_value = EXPECT_CLOSE_SQUARE_BRACKET | EXPECT_COMMA } else { args.dst.write_simple_token_fast!( value_major: 0, @@ -1090,9 +1087,8 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor base.TOKEN__VBD__STRUCTURE__TO_DICT, continued: 0, length: 1) - // 0x1044 is (EXPECT_CLOSE_CURLY_BRACE | EXPECT_COMMA). - expect = 0x1044 - expect_after_value = 0x1044 + expect = EXPECT_CLOSE_CURLY_BRACE | EXPECT_COMMA + expect_after_value = EXPECT_CLOSE_CURLY_BRACE | EXPECT_COMMA } continue.outer