Tweak JSON trailing comment/new-line quirks

This commit is contained in:
Nigel Tao committed 2020-09-22 23:03:49 +10:00
1 parent 3e8cb4ebf8
commit cd4cbc9102
7 files changed
+287 -205

No files matched your search

+38 -26
View File
@@ -15,14 +15,15 @@
pub struct decoder? implements base.token_decoder(
quirks : array[QUIRKS_COUNT] base.bool,
allow_leading_ars : base.bool,
allow_leading_ubom : base.bool,
allow_trailing_comment : base.bool,
allow_leading_ars : base.bool,
allow_leading_ubom : base.bool,
end_of_data : base.bool,
trailer_stop : base.u8,
// comment_type is set as a side-effect of decode_comment?.
// - 0 means failure.
// - 0 means no comment.
// - 1 means a block comment.
// - 2 means a line comment.
comment_type : base.u8,
@@ -123,6 +124,14 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor
return base."@end of data"
}
if this.quirks[QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF - QUIRKS_BASE] {
if this.quirks[QUIRK_ALLOW_COMMENT_BLOCK - QUIRKS_BASE] or
this.quirks[QUIRK_ALLOW_COMMENT_LINE - QUIRKS_BASE] or
this.quirks[QUIRK_ALLOW_TRAILING_FILLER - QUIRKS_BASE] {
return "#bad quirk combination"
}
}
if this.quirks[QUIRK_ALLOW_LEADING_ASCII_RECORD_SEPARATOR - QUIRKS_BASE] or
this.quirks[QUIRK_ALLOW_LEADING_UNICODE_BYTE_ORDER_MARK - QUIRKS_BASE] {
this.decode_leading?(dst: args.dst, src: args.src)
@@ -1176,7 +1185,9 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor
if this.quirks[QUIRK_ALLOW_COMMENT_BLOCK - QUIRKS_BASE] or
this.quirks[QUIRK_ALLOW_COMMENT_LINE - QUIRKS_BASE] {
this.decode_comment?(dst: args.dst, src: args.src)
continue.outer
if this.comment_type > 0 {
continue.outer
}
}
}
@@ -1191,8 +1202,9 @@ pub func decoder.decode_tokens?(dst: base.token_writer, src: base.io_reader, wor
expect = expect_after_value
} endwhile.outer
if this.quirks[QUIRK_ALLOW_TRAILING_NEW_LINE - QUIRKS_BASE] {
this.decode_trailing_new_line?(dst: args.dst, src: args.src)
if this.quirks[QUIRK_ALLOW_TRAILING_FILLER - QUIRKS_BASE] or
this.quirks[QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF - QUIRKS_BASE] {
this.decode_trailer?(dst: args.dst, src: args.src)
}
this.end_of_data = true
@@ -1438,7 +1450,7 @@ pri func decoder.decode_comment?(dst: base.token_writer, src: base.io_reader) {
continue
}
if args.src.is_closed() {
return "#bad input"
return ok
}
yield? base."$short read"
} endwhile
@@ -1562,8 +1574,6 @@ pri func decoder.decode_comment?(dst: base.token_writer, src: base.io_reader) {
} endwhile
} endwhile.comment_line
}
return "#bad input"
}
pri func decoder.decode_inf_nan?(dst: base.token_writer, src: base.io_reader) {
@@ -1677,11 +1687,15 @@ pri func decoder.decode_inf_nan?(dst: base.token_writer, src: base.io_reader) {
} endwhile
}
pri func decoder.decode_trailing_new_line?(dst: base.token_writer, src: base.io_reader) {
pri func decoder.decode_trailer?(dst: base.token_writer, src: base.io_reader) {
var c : base.u8
var whitespace_length : base.u32[..= 0xFFFE]
this.allow_trailing_comment = this.quirks[QUIRK_ALLOW_TRAILING_COMMENT - QUIRKS_BASE]
if this.quirks[QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF - QUIRKS_BASE] {
this.trailer_stop = '\n'
} else {
this.trailer_stop = 0
}
while.outer true {
if args.dst.length() <= 0 {
@@ -1714,27 +1728,25 @@ pri func decoder.decode_trailing_new_line?(dst: base.token_writer, src: base.io_
value_major: 0, value_minor: 0, continued: 0, length: whitespace_length)
whitespace_length = 0
}
if this.allow_trailing_comment {
this.decode_comment?(dst: args.dst, src: args.src)
c = 0
whitespace_length = 0
if this.comment_type == 1 { // Block comment.
this.allow_trailing_comment = false
continue.outer
} else if this.comment_type == 2 { // Line comment.
break.outer
}
if this.trailer_stop > 0 {
return "#bad input"
}
return "#bad input"
this.decode_comment?(dst: args.dst, src: args.src)
c = 0
whitespace_length = 0
if this.comment_type > 0 {
continue.outer
}
return ok
}
args.src.skip_u32_fast!(actual: 1, worst_case: 1)
if (whitespace_length >= 0xFFFE) or (c == '\n') {
if (whitespace_length >= 0xFFFE) or (c == this.trailer_stop) {
args.dst.write_simple_token_fast!(
value_major: 0, value_minor: 0, continued: 0, length: whitespace_length + 1)
whitespace_length = 0
if c == '\n' {
break.outer
if c == this.trailer_stop {
return ok
}
continue.outer
}
+42 -38
View File
@@ -97,16 +97,17 @@ pub const QUIRK_ALLOW_BACKSLASH_X_AS_CODE_POINTS : base.u32 = 0x4909_9400 | 0x09
pub const QUIRK_ALLOW_BACKSLASH_ZERO : base.u32 = 0x4909_9400 | 0x0A
// When this quirk is enabled, "/* C/C++ style block comments */" are accepted
// anywhere whitespace would be, although see the QUIRK_ALLOW_TRAILING_COMMENT
// comment for additional interaction when combining multiple quirks.
// anywhere whitespace would be. See also QUIRK_ALLOW_TRAILING_FILLER.
//
// They produce WUFFS_BASE__TOKEN__VBD__FILLER__COMMENT_BLOCK tokens. The token
// chain's source bytes includes the starting "/*" and the ending "*/".
//
// To avoid ambiguity (as comments can contain new lines), this quirk cannot be
// combined with QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF.
pub const QUIRK_ALLOW_COMMENT_BLOCK : base.u32 = 0x4909_9400 | 0x0B
// When this quirk is enabled, "// C/C++ style line comments\n" are accepted
// anywhere whitespace would be, although see the QUIRK_ALLOW_TRAILING_COMMENT
// comment for additional interaction when combining multiple quirks.
// anywhere whitespace would be. See also QUIRK_ALLOW_TRAILING_FILLER.
//
// A line comment may not omit the ending "\n", even if there is no input
// afterwards (i.e. the prospective line comment ends with the end-of-file).
@@ -117,6 +118,9 @@ pub const QUIRK_ALLOW_COMMENT_BLOCK : base.u32 = 0x4909_9400 | 0x0B
// Even if the line comments are on consecutive lines, each line comment is a
// separate token chain. There may be whitespace tokens between one line
// comment's ending "\n" and the next one's starting "//".
//
// To avoid ambiguity (as comments can contain new lines), this quirk cannot be
// combined with QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF.
pub const QUIRK_ALLOW_COMMENT_LINE : base.u32 = 0x4909_9400 | 0x0C
// When this quirk is enabled, there may be a comma after the final array
@@ -140,8 +144,9 @@ pub const QUIRK_ALLOW_INF_NAN_NUMBERS : base.u32 = 0x4909_9400 | 0x0E
// When combined with QUIRK_ALLOW_LEADING_UNICODE_BYTE_ORDER_MARK, either mark
// may come first in the byte stream.
//
// When combined with QUIRK_ALLOW_TRAILING_NEW_LINE, this format is also known
// as RFC 7464, Json Text Sequences and MIME type "application/json-seq".
// When combined with QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF, this format is
// also known as RFC 7464, Json Text Sequences and MIME type
// "application/json-seq".
pub const QUIRK_ALLOW_LEADING_ASCII_RECORD_SEPARATOR : base.u32 = 0x4909_9400 | 0x0F
// When this quirk is enabled, the input byte stream may optionally start with
@@ -152,40 +157,43 @@ pub const QUIRK_ALLOW_LEADING_ASCII_RECORD_SEPARATOR : base.u32 = 0x4909_9400 |
// may come first in the byte stream.
pub const QUIRK_ALLOW_LEADING_UNICODE_BYTE_ORDER_MARK : base.u32 = 0x4909_9400 | 0x10
// When this quirk is enabled and both:
// - QUIRK_ALLOW_TRAILING_NEW_LINE is enabled,
// - at least one of QUIRK_ALLOW_COMMENT_ETC is enabled,
// the trailing whitespace may optionally contain a single comment. As per
// QUIRK_ALLOW_TRAILING_NEW_LINE, processing will still stop at the first
// trailing '\n' (outside of a block comment), even if more comments followed.
// When this quirk is enabled, following a successful decoding of a top-level
// JSON value, any trailing whitespace (ASCII characters 0x09, 0x0A, 0x0D or
// 0x20) and/or comments (if QUIRK_ALLOW_COMMENT_ETC is enabled) are also
// consumed (and WUFFS_BASE__TOKEN__VBC__FILLER tokens emitted) up to but
// excluding the end-of-file or the next non-filler byte.
//
// For a trailing block comment, new lines within the comment are not counted
// and after the comment concludes, the decoder will continue consuming
// whitespace up to and including the next '\n' (or end-of-file).
// Trailing non-filler is not an error. Decoding simply stops before it.
//
// For a trailing line comment, the decoder stops immediately after the
// comment. If not stopped by end-of-file, this stops after the '\n' that
// concludes (and is part of) the comment. One implication is that if multiple
// line comments trail a JSON value, only the first one will be processed.
// To avoid ambiguity, this quirk cannot be combined with
// QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF. Unlike that quirk, enabling this
// quirk will consume multiple trailing '\n' bytes.
pub const QUIRK_ALLOW_TRAILING_FILLER : base.u32 = 0x4909_9400 | 0x11
// Deprecated: equivalent to QUIRK_ALLOW_TRAILING_FILLER.
pub const QUIRK_ALLOW_TRAILING_COMMENT : base.u32 = 0x4909_9400 | 0x11
// When this quirk is enabled, following a successful decoding of a top-level
// JSON value, any trailing whitespace (ASCII characters 0x09, 0x0A, 0x0D or
// 0x20) is also consumed (and WUFFS_BASE__TOKEN__VBC__FILLER tokens emitted)
// up to the end-of-file or up to and including a single new line (ASCII 0x0A,
// also known as '\n'), whichever comes first. This trailing whitespace is not
// mandatory, but it is consumed if present.
// up to but excluding the end-of-file or up to and including a single new line
// (ASCII 0x0A, also known as '\n'), whichever comes first. A trailing '\n' is
// not mandatory (if at end-of-file), but it is consumed if present and will
// stop decoding.
//
// When enabled, trailing non-whitespace (before a '\n') is an error (unless it
// is a comment and QUIRK_ALLOW_TRAILING_COMMENT is enabled; see above). For
// example, with "007" input, decoding with this quirk disabled (the default
// case) will consume just 1 byte ("0") and leave the rest ("07") unread
// without error, as "0" is a perfectly valid JSON value (but "00" is not).
// Decoding "007" (or "007\n") with this quirk enabled will return an error.
// Trailing non-whitespace after a trailing '\n' is ignored.
//
// When enabled, the decoder will not consume more than one trailing '\n'
// (outside of a block comment), nor will it consume any other whitespace
// immediately after a trailing '\n'.
// Trailing non-whitespace, before EOF or '\n', is an error. For example, with
// "007" input, decoding with this quirk disabled (the default case) will
// consume just 1 byte ("0") and leave the rest ("07") unread without error, as
// "0" is a perfectly valid JSON value (but "00" is not). Decoding "007" (or
// "007\n") with this quirk enabled will return an error.
//
// To avoid ambiguity (as comments can contain new lines), this quirk cannot be
// combined with any of:
// - QUIRK_ALLOW_COMMENT_BLOCK
// - QUIRK_ALLOW_COMMENT_LINE
// - QUIRK_ALLOW_TRAILING_FILLER
//
// If a JSON encoder avoids emitting (optional) '\n' bytes, other than a single
// '\n' after each top-level value in a multi-JSON-value stream, this format is
@@ -199,13 +207,9 @@ pub const QUIRK_ALLOW_TRAILING_COMMENT : base.u32 = 0x4909_9400 | 0x11
// When combined with QUIRK_ALLOW_LEADING_ASCII_RECORD_SEPARATOR, this format
// is also known as RFC 7464, Json Text Sequences and MIME type
// "application/json-seq".
//
// When combined with QUIRK_ALLOW_COMMENT_BLOCK or QUIRK_ALLOW_COMMENT_LINE, it
// is an error for a comment to occur in this trailing whitespace, unless
// QUIRK_ALLOW_TRAILING_COMMENT is also enabled. Be aware that block comments
// can contain multiple new lines, so combining such quirks can break the
// "exactly one JSON value per line" assumption for newline-delimited but
// otherwise compact (minified) JSON.
pub const QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF : base.u32 = 0x4909_9400 | 0x12
// Deprecated: equivalent to QUIRK_EXPECT_TRAILING_NEW_LINE_OR_EOF.
pub const QUIRK_ALLOW_TRAILING_NEW_LINE : base.u32 = 0x4909_9400 | 0x12
// When this quirk is enabled, JSON Pointer strings containing "~r" or "~n",