diff options
Diffstat (limited to 'core/encoding')
| -rw-r--r-- | core/encoding/ansi/ansi.odin | 137 | ||||
| -rw-r--r-- | core/encoding/ansi/doc.odin | 20 | ||||
| -rw-r--r-- | core/encoding/cbor/cbor.odin | 4 | ||||
| -rw-r--r-- | core/encoding/cbor/coding.odin | 107 | ||||
| -rw-r--r-- | core/encoding/cbor/marshal.odin | 16 | ||||
| -rw-r--r-- | core/encoding/cbor/unmarshal.odin | 102 | ||||
| -rw-r--r-- | core/encoding/entity/entity.odin | 123 | ||||
| -rw-r--r-- | core/encoding/hex/hex.odin | 11 | ||||
| -rw-r--r-- | core/encoding/hxa/hxa.odin | 31 | ||||
| -rw-r--r-- | core/encoding/hxa/read.odin | 42 | ||||
| -rw-r--r-- | core/encoding/ini/ini.odin | 189 | ||||
| -rw-r--r-- | core/encoding/json/marshal.odin | 4 | ||||
| -rw-r--r-- | core/encoding/json/parser.odin | 65 | ||||
| -rw-r--r-- | core/encoding/json/types.odin | 14 | ||||
| -rw-r--r-- | core/encoding/xml/tokenizer.odin | 43 | ||||
| -rw-r--r-- | core/encoding/xml/xml_reader.odin | 24 |
16 files changed, 624 insertions, 308 deletions
diff --git a/core/encoding/ansi/ansi.odin b/core/encoding/ansi/ansi.odin new file mode 100644 index 000000000..5550a1671 --- /dev/null +++ b/core/encoding/ansi/ansi.odin @@ -0,0 +1,137 @@ +package ansi + +BEL :: "\a" // Bell +BS :: "\b" // Backspace +ESC :: "\e" // Escape + +// Fe Escape sequences + +CSI :: ESC + "[" // Control Sequence Introducer +OSC :: ESC + "]" // Operating System Command +ST :: ESC + "\\" // String Terminator + +// CSI sequences + +CUU :: "A" // Cursor Up +CUD :: "B" // Cursor Down +CUF :: "C" // Cursor Forward +CUB :: "D" // Cursor Back +CNL :: "E" // Cursor Next Line +CPL :: "F" // Cursor Previous Line +CHA :: "G" // Cursor Horizontal Absolute +CUP :: "H" // Cursor Position +ED :: "J" // Erase in Display +EL :: "K" // Erase in Line +SU :: "S" // Scroll Up +SD :: "T" // Scroll Down +HVP :: "f" // Horizontal Vertical Position +SGR :: "m" // Select Graphic Rendition +AUX_ON :: "5i" // AUX Port On +AUX_OFF :: "4i" // AUX Port Off +DSR :: "6n" // Device Status Report + +// CSI: private sequences + +SCP :: "s" // Save Current Cursor Position +RCP :: "u" // Restore Saved Cursor Position +DECAWM_ON :: "?7h" // Auto Wrap Mode (Enabled) +DECAWM_OFF :: "?7l" // Auto Wrap Mode (Disabled) +DECTCEM_SHOW :: "?25h" // Text Cursor Enable Mode (Visible) +DECTCEM_HIDE :: "?25l" // Text Cursor Enable Mode (Invisible) + +// SGR sequences + +RESET :: "0" +BOLD :: "1" +FAINT :: "2" +ITALIC :: "3" // Not widely supported. +UNDERLINE :: "4" +BLINK_SLOW :: "5" +BLINK_RAPID :: "6" // Not widely supported. +INVERT :: "7" // Also known as reverse video. +HIDE :: "8" // Not widely supported. +STRIKE :: "9" +FONT_PRIMARY :: "10" +FONT_ALT1 :: "11" +FONT_ALT2 :: "12" +FONT_ALT3 :: "13" +FONT_ALT4 :: "14" +FONT_ALT5 :: "15" +FONT_ALT6 :: "16" +FONT_ALT7 :: "17" +FONT_ALT8 :: "18" +FONT_ALT9 :: "19" +FONT_FRAKTUR :: "20" // Rarely supported. +UNDERLINE_DOUBLE :: "21" // May be interpreted as "disable bold." +NO_BOLD_FAINT :: "22" +NO_ITALIC_BLACKLETTER :: "23" +NO_UNDERLINE :: "24" +NO_BLINK :: "25" +PROPORTIONAL_SPACING :: "26" +NO_REVERSE :: "27" +NO_HIDE :: "28" +NO_STRIKE :: "29" + +FG_BLACK :: "30" +FG_RED :: "31" +FG_GREEN :: "32" +FG_YELLOW :: "33" +FG_BLUE :: "34" +FG_MAGENTA :: "35" +FG_CYAN :: "36" +FG_WHITE :: "37" +FG_COLOR :: "38" +FG_COLOR_8_BIT :: "38;5" // Followed by ";n" where n is in 0..=255 +FG_COLOR_24_BIT :: "38;2" // Followed by ";r;g;b" where r,g,b are in 0..=255 +FG_DEFAULT :: "39" + +BG_BLACK :: "40" +BG_RED :: "41" +BG_GREEN :: "42" +BG_YELLOW :: "43" +BG_BLUE :: "44" +BG_MAGENTA :: "45" +BG_CYAN :: "46" +BG_WHITE :: "47" +BG_COLOR :: "48" +BG_COLOR_8_BIT :: "48;5" // Followed by ";n" where n is in 0..=255 +BG_COLOR_24_BIT :: "48;2" // Followed by ";r;g;b" where r,g,b are in 0..=255 +BG_DEFAULT :: "49" + +NO_PROPORTIONAL_SPACING :: "50" +FRAMED :: "51" +ENCIRCLED :: "52" +OVERLINED :: "53" +NO_FRAME_ENCIRCLE :: "54" +NO_OVERLINE :: "55" + +// SGR: non-standard bright colors + +FG_BRIGHT_BLACK :: "90" // Also known as grey. +FG_BRIGHT_RED :: "91" +FG_BRIGHT_GREEN :: "92" +FG_BRIGHT_YELLOW :: "93" +FG_BRIGHT_BLUE :: "94" +FG_BRIGHT_MAGENTA :: "95" +FG_BRIGHT_CYAN :: "96" +FG_BRIGHT_WHITE :: "97" + +BG_BRIGHT_BLACK :: "100" // Also known as grey. +BG_BRIGHT_RED :: "101" +BG_BRIGHT_GREEN :: "102" +BG_BRIGHT_YELLOW :: "103" +BG_BRIGHT_BLUE :: "104" +BG_BRIGHT_MAGENTA :: "105" +BG_BRIGHT_CYAN :: "106" +BG_BRIGHT_WHITE :: "107" + +// Fp Escape sequences + +DECSC :: ESC + "7" // DEC Save Cursor +DECRC :: ESC + "8" // DEC Restore Cursor + +// OSC sequences + +WINDOW_TITLE :: "2" // Followed by ";<text>" ST. +HYPERLINK :: "8" // Followed by ";[params];<URI>" ST. Closed by OSC HYPERLINK ";;" ST. +CLIPBOARD :: "52" // Followed by ";c;<Base64-encoded string>" ST. diff --git a/core/encoding/ansi/doc.odin b/core/encoding/ansi/doc.odin new file mode 100644 index 000000000..a0945c581 --- /dev/null +++ b/core/encoding/ansi/doc.odin @@ -0,0 +1,20 @@ +/* +package ansi implements constant references to many widely-supported ANSI +escape codes, primarily used in terminal emulators for enhanced graphics, such +as colors, text styling, and animated displays. + +For example, you can print out a line of cyan text like this: + fmt.println(ansi.CSI + ansi.FG_CYAN + ansi.SGR + "Hellope!" + ansi.CSI + ansi.RESET + ansi.SGR) + +Multiple SGR (Select Graphic Rendition) codes can be joined by semicolons: + fmt.println(ansi.CSI + ansi.BOLD + ";" + ansi.FG_BLUE + ansi.SGR + "Hellope!" + ansi.CSI + ansi.RESET + ansi.SGR) + +If your terminal supports 24-bit true color mode, you can also do this: + fmt.println(ansi.CSI + ansi.FG_COLOR_24_BIT + ";0;255;255" + ansi.SGR + "Hellope!" + ansi.CSI + ansi.RESET + ansi.SGR) + +For more information, see: + 1. https://en.wikipedia.org/wiki/ANSI_escape_code + 2. https://www.vt100.net/docs/vt102-ug/chapter5.html + 3. https://invisible-island.net/xterm/ctlseqs/ctlseqs.html +*/ +package ansi diff --git a/core/encoding/cbor/cbor.odin b/core/encoding/cbor/cbor.odin index d0e406ab1..7897b2a37 100644 --- a/core/encoding/cbor/cbor.odin +++ b/core/encoding/cbor/cbor.odin @@ -320,8 +320,8 @@ to_diagnostic_format :: proc { // Turns the given CBOR value into a human-readable string. // See docs on the proc group `diagnose` for more info. -to_diagnostic_format_string :: proc(val: Value, padding := 0, allocator := context.allocator) -> (string, mem.Allocator_Error) #optional_allocator_error { - b := strings.builder_make(allocator) +to_diagnostic_format_string :: proc(val: Value, padding := 0, allocator := context.allocator, loc := #caller_location) -> (string, mem.Allocator_Error) #optional_allocator_error { + b := strings.builder_make(allocator, loc) w := strings.to_stream(&b) err := to_diagnostic_format_writer(w, val, padding) if err == .EOF { diff --git a/core/encoding/cbor/coding.odin b/core/encoding/cbor/coding.odin index 0d276a7a1..07f0637a6 100644 --- a/core/encoding/cbor/coding.odin +++ b/core/encoding/cbor/coding.odin @@ -95,24 +95,25 @@ decode :: decode_from // Decodes the given string as CBOR. // See docs on the proc group `decode` for more information. -decode_from_string :: proc(s: string, flags: Decoder_Flags = {}, allocator := context.allocator) -> (v: Value, err: Decode_Error) { +decode_from_string :: proc(s: string, flags: Decoder_Flags = {}, allocator := context.allocator, loc := #caller_location) -> (v: Value, err: Decode_Error) { r: strings.Reader strings.reader_init(&r, s) - return decode_from_reader(strings.reader_to_stream(&r), flags, allocator) + return decode_from_reader(strings.reader_to_stream(&r), flags, allocator, loc) } // Reads a CBOR value from the given reader. // See docs on the proc group `decode` for more information. -decode_from_reader :: proc(r: io.Reader, flags: Decoder_Flags = {}, allocator := context.allocator) -> (v: Value, err: Decode_Error) { +decode_from_reader :: proc(r: io.Reader, flags: Decoder_Flags = {}, allocator := context.allocator, loc := #caller_location) -> (v: Value, err: Decode_Error) { return decode_from_decoder( Decoder{ DEFAULT_MAX_PRE_ALLOC, flags, r }, allocator=allocator, + loc = loc, ) } // Reads a CBOR value from the given decoder. // See docs on the proc group `decode` for more information. -decode_from_decoder :: proc(d: Decoder, allocator := context.allocator) -> (v: Value, err: Decode_Error) { +decode_from_decoder :: proc(d: Decoder, allocator := context.allocator, loc := #caller_location) -> (v: Value, err: Decode_Error) { context.allocator = allocator d := d @@ -121,13 +122,13 @@ decode_from_decoder :: proc(d: Decoder, allocator := context.allocator) -> (v: V d.max_pre_alloc = DEFAULT_MAX_PRE_ALLOC } - v, err = _decode_from_decoder(d) + v, err = _decode_from_decoder(d, {}, allocator, loc) // Normal EOF does not exist here, we try to read the exact amount that is said to be provided. if err == .EOF { err = .Unexpected_EOF } return } -_decode_from_decoder :: proc(d: Decoder, hdr: Header = Header(0)) -> (v: Value, err: Decode_Error) { +_decode_from_decoder :: proc(d: Decoder, hdr: Header = Header(0), allocator := context.allocator, loc := #caller_location) -> (v: Value, err: Decode_Error) { hdr := hdr r := d.reader if hdr == Header(0) { hdr = _decode_header(r) or_return } @@ -161,11 +162,11 @@ _decode_from_decoder :: proc(d: Decoder, hdr: Header = Header(0)) -> (v: Value, switch maj { case .Unsigned: return _decode_tiny_u8(add) case .Negative: return Negative_U8(_decode_tiny_u8(add) or_return), nil - case .Bytes: return _decode_bytes_ptr(d, add) - case .Text: return _decode_text_ptr(d, add) - case .Array: return _decode_array_ptr(d, add) - case .Map: return _decode_map_ptr(d, add) - case .Tag: return _decode_tag_ptr(d, add) + case .Bytes: return _decode_bytes_ptr(d, add, .Bytes, allocator, loc) + case .Text: return _decode_text_ptr(d, add, allocator, loc) + case .Array: return _decode_array_ptr(d, add, allocator, loc) + case .Map: return _decode_map_ptr(d, add, allocator, loc) + case .Tag: return _decode_tag_ptr(d, add, allocator, loc) case .Other: return _decode_tiny_simple(add) case: return nil, .Bad_Major } @@ -203,27 +204,27 @@ encode :: encode_into // Encodes the CBOR value into binary CBOR allocated on the given allocator. // See the docs on the proc group `encode_into` for more info. -encode_into_bytes :: proc(v: Value, flags := ENCODE_SMALL, allocator := context.allocator, temp_allocator := context.temp_allocator) -> (data: []byte, err: Encode_Error) { - b := strings.builder_make(allocator) or_return +encode_into_bytes :: proc(v: Value, flags := ENCODE_SMALL, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> (data: []byte, err: Encode_Error) { + b := strings.builder_make(allocator, loc) or_return encode_into_builder(&b, v, flags, temp_allocator) or_return return b.buf[:], nil } // Encodes the CBOR value into binary CBOR written to the given builder. // See the docs on the proc group `encode_into` for more info. -encode_into_builder :: proc(b: ^strings.Builder, v: Value, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator) -> Encode_Error { - return encode_into_writer(strings.to_stream(b), v, flags, temp_allocator) +encode_into_builder :: proc(b: ^strings.Builder, v: Value, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator, loc := #caller_location) -> Encode_Error { + return encode_into_writer(strings.to_stream(b), v, flags, temp_allocator, loc=loc) } // Encodes the CBOR value into binary CBOR written to the given writer. // See the docs on the proc group `encode_into` for more info. -encode_into_writer :: proc(w: io.Writer, v: Value, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator) -> Encode_Error { - return encode_into_encoder(Encoder{flags, w, temp_allocator}, v) +encode_into_writer :: proc(w: io.Writer, v: Value, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator, loc := #caller_location) -> Encode_Error { + return encode_into_encoder(Encoder{flags, w, temp_allocator}, v, loc=loc) } // Encodes the CBOR value into binary CBOR written to the given encoder. // See the docs on the proc group `encode_into` for more info. -encode_into_encoder :: proc(e: Encoder, v: Value) -> Encode_Error { +encode_into_encoder :: proc(e: Encoder, v: Value, loc := #caller_location) -> Encode_Error { e := e if e.temp_allocator.procedure == nil { @@ -366,21 +367,21 @@ _encode_u64_exact :: proc(w: io.Writer, v: u64, major: Major = .Unsigned) -> (er return } -_decode_bytes_ptr :: proc(d: Decoder, add: Add, type: Major = .Bytes) -> (v: ^Bytes, err: Decode_Error) { - v = new(Bytes) or_return - defer if err != nil { free(v) } +_decode_bytes_ptr :: proc(d: Decoder, add: Add, type: Major = .Bytes, allocator := context.allocator, loc := #caller_location) -> (v: ^Bytes, err: Decode_Error) { + v = new(Bytes, allocator, loc) or_return + defer if err != nil { free(v, allocator, loc) } - v^ = _decode_bytes(d, add, type) or_return + v^ = _decode_bytes(d, add, type, allocator, loc) or_return return } -_decode_bytes :: proc(d: Decoder, add: Add, type: Major = .Bytes, allocator := context.allocator) -> (v: Bytes, err: Decode_Error) { +_decode_bytes :: proc(d: Decoder, add: Add, type: Major = .Bytes, allocator := context.allocator, loc := #caller_location) -> (v: Bytes, err: Decode_Error) { context.allocator = allocator add := add n, scap := _decode_len_str(d, add) or_return - buf := strings.builder_make(0, scap) or_return + buf := strings.builder_make(0, scap, allocator, loc) or_return defer if err != nil { strings.builder_destroy(&buf) } buf_stream := strings.to_stream(&buf) @@ -426,40 +427,40 @@ _encode_bytes :: proc(e: Encoder, val: Bytes, major: Major = .Bytes) -> (err: En return } -_decode_text_ptr :: proc(d: Decoder, add: Add) -> (v: ^Text, err: Decode_Error) { - v = new(Text) or_return +_decode_text_ptr :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: ^Text, err: Decode_Error) { + v = new(Text, allocator, loc) or_return defer if err != nil { free(v) } - v^ = _decode_text(d, add) or_return + v^ = _decode_text(d, add, allocator, loc) or_return return } -_decode_text :: proc(d: Decoder, add: Add, allocator := context.allocator) -> (v: Text, err: Decode_Error) { - return (Text)(_decode_bytes(d, add, .Text, allocator) or_return), nil +_decode_text :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: Text, err: Decode_Error) { + return (Text)(_decode_bytes(d, add, .Text, allocator, loc) or_return), nil } _encode_text :: proc(e: Encoder, val: Text) -> Encode_Error { return _encode_bytes(e, transmute([]byte)val, .Text) } -_decode_array_ptr :: proc(d: Decoder, add: Add) -> (v: ^Array, err: Decode_Error) { - v = new(Array) or_return +_decode_array_ptr :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: ^Array, err: Decode_Error) { + v = new(Array, allocator, loc) or_return defer if err != nil { free(v) } - v^ = _decode_array(d, add) or_return + v^ = _decode_array(d, add, allocator, loc) or_return return } -_decode_array :: proc(d: Decoder, add: Add) -> (v: Array, err: Decode_Error) { +_decode_array :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: Array, err: Decode_Error) { n, scap := _decode_len_container(d, add) or_return - array := make([dynamic]Value, 0, scap) or_return + array := make([dynamic]Value, 0, scap, allocator, loc) or_return defer if err != nil { - for entry in array { destroy(entry) } - delete(array) + for entry in array { destroy(entry, allocator) } + delete(array, loc) } for i := 0; n == -1 || i < n; i += 1 { - val, verr := _decode_from_decoder(d) + val, verr := _decode_from_decoder(d, {}, allocator, loc) if n == -1 && verr == .Break { break } else if verr != nil { @@ -485,39 +486,39 @@ _encode_array :: proc(e: Encoder, arr: Array) -> Encode_Error { return nil } -_decode_map_ptr :: proc(d: Decoder, add: Add) -> (v: ^Map, err: Decode_Error) { - v = new(Map) or_return +_decode_map_ptr :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: ^Map, err: Decode_Error) { + v = new(Map, allocator, loc) or_return defer if err != nil { free(v) } - v^ = _decode_map(d, add) or_return + v^ = _decode_map(d, add, allocator, loc) or_return return } -_decode_map :: proc(d: Decoder, add: Add) -> (v: Map, err: Decode_Error) { +_decode_map :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: Map, err: Decode_Error) { n, scap := _decode_len_container(d, add) or_return - items := make([dynamic]Map_Entry, 0, scap) or_return + items := make([dynamic]Map_Entry, 0, scap, allocator, loc) or_return defer if err != nil { for entry in items { destroy(entry.key) destroy(entry.value) } - delete(items) + delete(items, loc) } for i := 0; n == -1 || i < n; i += 1 { - key, kerr := _decode_from_decoder(d) + key, kerr := _decode_from_decoder(d, {}, allocator, loc) if n == -1 && kerr == .Break { break } else if kerr != nil { return nil, kerr } - value := _decode_from_decoder(d) or_return + value := _decode_from_decoder(d, {}, allocator, loc) or_return append(&items, Map_Entry{ key = key, value = value, - }) or_return + }, loc) or_return } if .Shrink_Excess in d.flags { shrink(&items) } @@ -578,20 +579,20 @@ _encode_map :: proc(e: Encoder, m: Map) -> (err: Encode_Error) { return nil } -_decode_tag_ptr :: proc(d: Decoder, add: Add) -> (v: Value, err: Decode_Error) { - tag := _decode_tag(d, add) or_return +_decode_tag_ptr :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: Value, err: Decode_Error) { + tag := _decode_tag(d, add, allocator, loc) or_return if t, ok := tag.?; ok { defer if err != nil { destroy(t.value) } - tp := new(Tag) or_return + tp := new(Tag, allocator, loc) or_return tp^ = t return tp, nil } // no error, no tag, this was the self described CBOR tag, skip it. - return _decode_from_decoder(d) + return _decode_from_decoder(d, {}, allocator, loc) } -_decode_tag :: proc(d: Decoder, add: Add) -> (v: Maybe(Tag), err: Decode_Error) { +_decode_tag :: proc(d: Decoder, add: Add, allocator := context.allocator, loc := #caller_location) -> (v: Maybe(Tag), err: Decode_Error) { num := _decode_uint_as_u64(d.reader, add) or_return // CBOR can be wrapped in a tag that decoders can use to see/check if the binary data is CBOR. @@ -602,7 +603,7 @@ _decode_tag :: proc(d: Decoder, add: Add) -> (v: Maybe(Tag), err: Decode_Error) t := Tag{ number = num, - value = _decode_from_decoder(d) or_return, + value = _decode_from_decoder(d, {}, allocator, loc) or_return, } if nested, ok := t.value.(^Tag); ok { @@ -883,4 +884,4 @@ _encode_deterministic_f64 :: proc(w: io.Writer, v: f64) -> io.Error { } return _encode_f64_exact(w, v) -} +}
\ No newline at end of file diff --git a/core/encoding/cbor/marshal.odin b/core/encoding/cbor/marshal.odin index 37c9dd180..775eafd9c 100644 --- a/core/encoding/cbor/marshal.odin +++ b/core/encoding/cbor/marshal.odin @@ -45,8 +45,8 @@ marshal :: marshal_into // Marshals the given value into a CBOR byte stream (allocated using the given allocator). // See docs on the `marshal_into` proc group for more info. -marshal_into_bytes :: proc(v: any, flags := ENCODE_SMALL, allocator := context.allocator, temp_allocator := context.temp_allocator) -> (bytes: []byte, err: Marshal_Error) { - b, alloc_err := strings.builder_make(allocator) +marshal_into_bytes :: proc(v: any, flags := ENCODE_SMALL, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> (bytes: []byte, err: Marshal_Error) { + b, alloc_err := strings.builder_make(allocator, loc=loc) // The builder as a stream also returns .EOF if it ran out of memory so this is consistent. if alloc_err != nil { return nil, .EOF @@ -54,7 +54,7 @@ marshal_into_bytes :: proc(v: any, flags := ENCODE_SMALL, allocator := context.a defer if err != nil { strings.builder_destroy(&b) } - if err = marshal_into_builder(&b, v, flags, temp_allocator); err != nil { + if err = marshal_into_builder(&b, v, flags, temp_allocator, loc=loc); err != nil { return } @@ -63,20 +63,20 @@ marshal_into_bytes :: proc(v: any, flags := ENCODE_SMALL, allocator := context.a // Marshals the given value into a CBOR byte stream written to the given builder. // See docs on the `marshal_into` proc group for more info. -marshal_into_builder :: proc(b: ^strings.Builder, v: any, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator) -> Marshal_Error { - return marshal_into_writer(strings.to_writer(b), v, flags, temp_allocator) +marshal_into_builder :: proc(b: ^strings.Builder, v: any, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator, loc := #caller_location) -> Marshal_Error { + return marshal_into_writer(strings.to_writer(b), v, flags, temp_allocator, loc=loc) } // Marshals the given value into a CBOR byte stream written to the given writer. // See docs on the `marshal_into` proc group for more info. -marshal_into_writer :: proc(w: io.Writer, v: any, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator) -> Marshal_Error { +marshal_into_writer :: proc(w: io.Writer, v: any, flags := ENCODE_SMALL, temp_allocator := context.temp_allocator, loc := #caller_location) -> Marshal_Error { encoder := Encoder{flags, w, temp_allocator} - return marshal_into_encoder(encoder, v) + return marshal_into_encoder(encoder, v, loc=loc) } // Marshals the given value into a CBOR byte stream written to the given encoder. // See docs on the `marshal_into` proc group for more info. -marshal_into_encoder :: proc(e: Encoder, v: any) -> (err: Marshal_Error) { +marshal_into_encoder :: proc(e: Encoder, v: any, loc := #caller_location) -> (err: Marshal_Error) { e := e if e.temp_allocator.procedure == nil { diff --git a/core/encoding/cbor/unmarshal.odin b/core/encoding/cbor/unmarshal.odin index a1524d9f4..c31ba1d92 100644 --- a/core/encoding/cbor/unmarshal.odin +++ b/core/encoding/cbor/unmarshal.odin @@ -31,8 +31,8 @@ unmarshal :: proc { unmarshal_from_string, } -unmarshal_from_reader :: proc(r: io.Reader, ptr: ^$T, flags := Decoder_Flags{}, allocator := context.allocator, temp_allocator := context.temp_allocator) -> (err: Unmarshal_Error) { - err = unmarshal_from_decoder(Decoder{ DEFAULT_MAX_PRE_ALLOC, flags, r }, ptr, allocator, temp_allocator) +unmarshal_from_reader :: proc(r: io.Reader, ptr: ^$T, flags := Decoder_Flags{}, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> (err: Unmarshal_Error) { + err = unmarshal_from_decoder(Decoder{ DEFAULT_MAX_PRE_ALLOC, flags, r }, ptr, allocator, temp_allocator, loc) // Normal EOF does not exist here, we try to read the exact amount that is said to be provided. if err == .EOF { err = .Unexpected_EOF } @@ -40,21 +40,21 @@ unmarshal_from_reader :: proc(r: io.Reader, ptr: ^$T, flags := Decoder_Flags{}, } // Unmarshals from a string, see docs on the proc group `Unmarshal` for more info. -unmarshal_from_string :: proc(s: string, ptr: ^$T, flags := Decoder_Flags{}, allocator := context.allocator, temp_allocator := context.temp_allocator) -> (err: Unmarshal_Error) { +unmarshal_from_string :: proc(s: string, ptr: ^$T, flags := Decoder_Flags{}, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> (err: Unmarshal_Error) { sr: strings.Reader r := strings.to_reader(&sr, s) - err = unmarshal_from_reader(r, ptr, flags, allocator, temp_allocator) + err = unmarshal_from_reader(r, ptr, flags, allocator, temp_allocator, loc) // Normal EOF does not exist here, we try to read the exact amount that is said to be provided. if err == .EOF { err = .Unexpected_EOF } return } -unmarshal_from_decoder :: proc(d: Decoder, ptr: ^$T, allocator := context.allocator, temp_allocator := context.temp_allocator) -> (err: Unmarshal_Error) { +unmarshal_from_decoder :: proc(d: Decoder, ptr: ^$T, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> (err: Unmarshal_Error) { d := d - err = _unmarshal_any_ptr(d, ptr, nil, allocator, temp_allocator) + err = _unmarshal_any_ptr(d, ptr, nil, allocator, temp_allocator, loc) // Normal EOF does not exist here, we try to read the exact amount that is said to be provided. if err == .EOF { err = .Unexpected_EOF } @@ -62,7 +62,7 @@ unmarshal_from_decoder :: proc(d: Decoder, ptr: ^$T, allocator := context.alloca } -_unmarshal_any_ptr :: proc(d: Decoder, v: any, hdr: Maybe(Header) = nil, allocator := context.allocator, temp_allocator := context.temp_allocator) -> Unmarshal_Error { +_unmarshal_any_ptr :: proc(d: Decoder, v: any, hdr: Maybe(Header) = nil, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> Unmarshal_Error { context.allocator = allocator context.temp_allocator = temp_allocator v := v @@ -78,10 +78,10 @@ _unmarshal_any_ptr :: proc(d: Decoder, v: any, hdr: Maybe(Header) = nil, allocat } data := any{(^rawptr)(v.data)^, ti.variant.(reflect.Type_Info_Pointer).elem.id} - return _unmarshal_value(d, data, hdr.? or_else (_decode_header(d.reader) or_return)) + return _unmarshal_value(d, data, hdr.? or_else (_decode_header(d.reader) or_return), allocator, temp_allocator, loc) } -_unmarshal_value :: proc(d: Decoder, v: any, hdr: Header) -> (err: Unmarshal_Error) { +_unmarshal_value :: proc(d: Decoder, v: any, hdr: Header, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> (err: Unmarshal_Error) { v := v ti := reflect.type_info_base(type_info_of(v.id)) r := d.reader @@ -104,7 +104,7 @@ _unmarshal_value :: proc(d: Decoder, v: any, hdr: Header) -> (err: Unmarshal_Err // Allow generic unmarshal by doing it into a `Value`. switch &dst in v { case Value: - dst = err_conv(_decode_from_decoder(d, hdr)) or_return + dst = err_conv(_decode_from_decoder(d, hdr, allocator, loc)) or_return return } @@ -308,7 +308,7 @@ _unmarshal_value :: proc(d: Decoder, v: any, hdr: Header) -> (err: Unmarshal_Err if impl, ok := _tag_implementations_nr[nr]; ok { return impl->unmarshal(d, nr, v) } else if nr == TAG_OBJECT_TYPE { - return _unmarshal_union(d, v, ti, hdr) + return _unmarshal_union(d, v, ti, hdr, loc=loc) } else { // Discard the tag info and unmarshal as its value. return _unmarshal_value(d, v, _decode_header(r) or_return) @@ -316,19 +316,19 @@ _unmarshal_value :: proc(d: Decoder, v: any, hdr: Header) -> (err: Unmarshal_Err return _unsupported(v, hdr, add) - case .Bytes: return _unmarshal_bytes(d, v, ti, hdr, add) - case .Text: return _unmarshal_string(d, v, ti, hdr, add) - case .Array: return _unmarshal_array(d, v, ti, hdr, add) - case .Map: return _unmarshal_map(d, v, ti, hdr, add) + case .Bytes: return _unmarshal_bytes(d, v, ti, hdr, add, allocator=allocator, loc=loc) + case .Text: return _unmarshal_string(d, v, ti, hdr, add, allocator=allocator, loc=loc) + case .Array: return _unmarshal_array(d, v, ti, hdr, add, allocator=allocator, loc=loc) + case .Map: return _unmarshal_map(d, v, ti, hdr, add, allocator=allocator, loc=loc) case: return .Bad_Major } } -_unmarshal_bytes :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add) -> (err: Unmarshal_Error) { +_unmarshal_bytes :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add, allocator := context.allocator, loc := #caller_location) -> (err: Unmarshal_Error) { #partial switch t in ti.variant { case reflect.Type_Info_String: - bytes := err_conv(_decode_bytes(d, add)) or_return + bytes := err_conv(_decode_bytes(d, add, allocator=allocator, loc=loc)) or_return if t.is_cstring { raw := (^cstring)(v.data) @@ -347,7 +347,7 @@ _unmarshal_bytes :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header if elem_base.id != byte { return _unsupported(v, hdr) } - bytes := err_conv(_decode_bytes(d, add)) or_return + bytes := err_conv(_decode_bytes(d, add, allocator=allocator, loc=loc)) or_return raw := (^mem.Raw_Slice)(v.data) raw^ = transmute(mem.Raw_Slice)bytes return @@ -357,12 +357,12 @@ _unmarshal_bytes :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header if elem_base.id != byte { return _unsupported(v, hdr) } - bytes := err_conv(_decode_bytes(d, add)) or_return + bytes := err_conv(_decode_bytes(d, add, allocator=allocator, loc=loc)) or_return raw := (^mem.Raw_Dynamic_Array)(v.data) raw.data = raw_data(bytes) raw.len = len(bytes) raw.cap = len(bytes) - raw.allocator = context.allocator + raw.allocator = allocator return case reflect.Type_Info_Array: @@ -385,10 +385,10 @@ _unmarshal_bytes :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header return _unsupported(v, hdr) } -_unmarshal_string :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add) -> (err: Unmarshal_Error) { +_unmarshal_string :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add, allocator := context.allocator, temp_allocator := context.temp_allocator, loc := #caller_location) -> (err: Unmarshal_Error) { #partial switch t in ti.variant { case reflect.Type_Info_String: - text := err_conv(_decode_text(d, add)) or_return + text := err_conv(_decode_text(d, add, allocator, loc)) or_return if t.is_cstring { raw := (^cstring)(v.data) @@ -403,8 +403,8 @@ _unmarshal_string :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Heade // Enum by its variant name. case reflect.Type_Info_Enum: - text := err_conv(_decode_text(d, add, allocator=context.temp_allocator)) or_return - defer delete(text, context.temp_allocator) + text := err_conv(_decode_text(d, add, allocator=temp_allocator, loc=loc)) or_return + defer delete(text, temp_allocator, loc) for name, i in t.names { if name == text { @@ -414,8 +414,8 @@ _unmarshal_string :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Heade } case reflect.Type_Info_Rune: - text := err_conv(_decode_text(d, add, allocator=context.temp_allocator)) or_return - defer delete(text, context.temp_allocator) + text := err_conv(_decode_text(d, add, allocator=temp_allocator, loc=loc)) or_return + defer delete(text, temp_allocator, loc) r := (^rune)(v.data) dr, n := utf8.decode_rune(text) @@ -430,13 +430,15 @@ _unmarshal_string :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Heade return _unsupported(v, hdr) } -_unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add) -> (err: Unmarshal_Error) { +_unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add, allocator := context.allocator, loc := #caller_location) -> (err: Unmarshal_Error) { assign_array :: proc( d: Decoder, da: ^mem.Raw_Dynamic_Array, elemt: ^reflect.Type_Info, length: int, growable := true, + allocator := context.allocator, + loc := #caller_location, ) -> (out_of_space: bool, err: Unmarshal_Error) { for idx: uintptr = 0; length == -1 || idx < uintptr(length); idx += 1 { elem_ptr := rawptr(uintptr(da.data) + idx*uintptr(elemt.size)) @@ -450,13 +452,13 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header if !growable { return true, .Out_Of_Memory } cap := 2 * da.cap - ok := runtime.__dynamic_array_reserve(da, elemt.size, elemt.align, cap) + ok := runtime.__dynamic_array_reserve(da, elemt.size, elemt.align, cap, loc) // NOTE: Might be lying here, but it is at least an allocator error. if !ok { return false, .Out_Of_Memory } } - err = _unmarshal_value(d, elem, hdr) + err = _unmarshal_value(d, elem, hdr, allocator=allocator, loc=loc) if length == -1 && err == .Break { break } if err != nil { return } @@ -469,10 +471,10 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header // Allow generically storing the values array. switch &dst in v { case ^Array: - dst = err_conv(_decode_array_ptr(d, add)) or_return + dst = err_conv(_decode_array_ptr(d, add, allocator=allocator, loc=loc)) or_return return case Array: - dst = err_conv(_decode_array(d, add)) or_return + dst = err_conv(_decode_array(d, add, allocator=allocator, loc=loc)) or_return return } @@ -480,8 +482,8 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header case reflect.Type_Info_Slice: length, scap := err_conv(_decode_len_container(d, add)) or_return - data := mem.alloc_bytes_non_zeroed(t.elem.size * scap, t.elem.align) or_return - defer if err != nil { mem.free_bytes(data) } + data := mem.alloc_bytes_non_zeroed(t.elem.size * scap, t.elem.align, allocator=allocator, loc=loc) or_return + defer if err != nil { mem.free_bytes(data, allocator=allocator, loc=loc) } da := mem.Raw_Dynamic_Array{raw_data(data), 0, length, context.allocator } @@ -489,7 +491,7 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header if .Shrink_Excess in d.flags { // Ignoring an error here, but this is not critical to succeed. - _ = runtime.__dynamic_array_shrink(&da, t.elem.size, t.elem.align, da.len) + _ = runtime.__dynamic_array_shrink(&da, t.elem.size, t.elem.align, da.len, loc=loc) } raw := (^mem.Raw_Slice)(v.data) @@ -500,8 +502,8 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header case reflect.Type_Info_Dynamic_Array: length, scap := err_conv(_decode_len_container(d, add)) or_return - data := mem.alloc_bytes_non_zeroed(t.elem.size * scap, t.elem.align) or_return - defer if err != nil { mem.free_bytes(data) } + data := mem.alloc_bytes_non_zeroed(t.elem.size * scap, t.elem.align, loc=loc) or_return + defer if err != nil { mem.free_bytes(data, allocator=allocator, loc=loc) } raw := (^mem.Raw_Dynamic_Array)(v.data) raw.data = raw_data(data) @@ -513,7 +515,7 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header if .Shrink_Excess in d.flags { // Ignoring an error here, but this is not critical to succeed. - _ = runtime.__dynamic_array_shrink(raw, t.elem.size, t.elem.align, raw.len) + _ = runtime.__dynamic_array_shrink(raw, t.elem.size, t.elem.align, raw.len, loc=loc) } return @@ -525,7 +527,7 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header return _unsupported(v, hdr) } - da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, length, context.allocator } + da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, length, allocator } out_of_space := assign_array(d, &da, t.elem, length, growable=false) or_return if out_of_space { return _unsupported(v, hdr) } @@ -539,7 +541,7 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header return _unsupported(v, hdr) } - da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, length, context.allocator } + da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, length, allocator } out_of_space := assign_array(d, &da, t.elem, length, growable=false) or_return if out_of_space { return _unsupported(v, hdr) } @@ -553,7 +555,7 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header return _unsupported(v, hdr) } - da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, 2, context.allocator } + da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, 2, allocator } info: ^runtime.Type_Info switch ti.id { @@ -575,7 +577,7 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header return _unsupported(v, hdr) } - da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, 4, context.allocator } + da := mem.Raw_Dynamic_Array{rawptr(v.data), 0, 4, allocator } info: ^runtime.Type_Info switch ti.id { @@ -593,17 +595,17 @@ _unmarshal_array :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header } } -_unmarshal_map :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add) -> (err: Unmarshal_Error) { +_unmarshal_map :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, add: Add, allocator := context.allocator, loc := #caller_location) -> (err: Unmarshal_Error) { r := d.reader - decode_key :: proc(d: Decoder, v: any, allocator := context.allocator) -> (k: string, err: Unmarshal_Error) { + decode_key :: proc(d: Decoder, v: any, allocator := context.allocator, loc := #caller_location) -> (k: string, err: Unmarshal_Error) { entry_hdr := _decode_header(d.reader) or_return entry_maj, entry_add := _header_split(entry_hdr) #partial switch entry_maj { case .Text: - k = err_conv(_decode_text(d, entry_add, allocator)) or_return + k = err_conv(_decode_text(d, entry_add, allocator=allocator, loc=loc)) or_return return case .Bytes: - bytes := err_conv(_decode_bytes(d, entry_add, allocator=allocator)) or_return + bytes := err_conv(_decode_bytes(d, entry_add, allocator=allocator, loc=loc)) or_return k = string(bytes) return case: @@ -615,10 +617,10 @@ _unmarshal_map :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, // Allow generically storing the map array. switch &dst in v { case ^Map: - dst = err_conv(_decode_map_ptr(d, add)) or_return + dst = err_conv(_decode_map_ptr(d, add, allocator=allocator, loc=loc)) or_return return case Map: - dst = err_conv(_decode_map(d, add)) or_return + dst = err_conv(_decode_map(d, add, allocator=allocator, loc=loc)) or_return return } @@ -754,7 +756,7 @@ _unmarshal_map :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, // Unmarshal into a union, based on the `TAG_OBJECT_TYPE` tag of the spec, it denotes a tag which // contains an array of exactly two elements, the first is a textual representation of the following // CBOR value's type. -_unmarshal_union :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header) -> (err: Unmarshal_Error) { +_unmarshal_union :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header, loc := #caller_location) -> (err: Unmarshal_Error) { r := d.reader #partial switch t in ti.variant { case reflect.Type_Info_Union: @@ -792,7 +794,7 @@ _unmarshal_union :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header case reflect.Type_Info_Named: if vti.name == target_name { reflect.set_union_variant_raw_tag(v, tag) - return _unmarshal_value(d, any{v.data, variant.id}, _decode_header(r) or_return) + return _unmarshal_value(d, any{v.data, variant.id}, _decode_header(r) or_return, loc=loc) } case: @@ -804,7 +806,7 @@ _unmarshal_union :: proc(d: Decoder, v: any, ti: ^reflect.Type_Info, hdr: Header if variant_name == target_name { reflect.set_union_variant_raw_tag(v, tag) - return _unmarshal_value(d, any{v.data, variant.id}, _decode_header(r) or_return) + return _unmarshal_value(d, any{v.data, variant.id}, _decode_header(r) or_return, loc=loc) } } } diff --git a/core/encoding/entity/entity.odin b/core/encoding/entity/entity.odin index cee6230ef..f5208ad6f 100644 --- a/core/encoding/entity/entity.odin +++ b/core/encoding/entity/entity.odin @@ -56,38 +56,27 @@ CDATA_END :: "]]>" COMMENT_START :: "<!--" COMMENT_END :: "-->" -/* - Default: CDATA and comments are passed through unchanged. -*/ +// Default: CDATA and comments are passed through unchanged. XML_Decode_Option :: enum u8 { - /* - Do not decode & entities. It decodes by default. - If given, overrides `Decode_CDATA`. - */ + // Do not decode & entities. It decodes by default. If given, overrides `Decode_CDATA`. No_Entity_Decode, - /* - CDATA is unboxed. - */ + // CDATA is unboxed. Unbox_CDATA, - /* - Unboxed CDATA is decoded as well. - Ignored if `.Unbox_CDATA` is not given. - */ + // Unboxed CDATA is decoded as well. Ignored if `.Unbox_CDATA` is not given. Decode_CDATA, - /* - Comments are stripped. - */ + // Comments are stripped. Comment_Strip, + + // Normalize whitespace + Normalize_Whitespace, } XML_Decode_Options :: bit_set[XML_Decode_Option; u8] -/* - Decode a string that may include SGML/XML/HTML entities. - The caller has to free the result. -*/ +// Decode a string that may include SGML/XML/HTML entities. +// The caller has to free the result. decode_xml :: proc(input: string, options := XML_Decode_Options{}, allocator := context.allocator) -> (decoded: string, err: Error) { context.allocator = allocator @@ -100,14 +89,14 @@ decode_xml :: proc(input: string, options := XML_Decode_Options{}, allocator := t := Tokenizer{src=input} in_data := false + prev: rune = ' ' + loop: for { advance(&t) or_return if t.r < 0 { break loop } - /* - Below here we're never inside a CDATA tag. - At most we'll see the start of one, but that doesn't affect the logic. - */ + // Below here we're never inside a CDATA tag. At most we'll see the start of one, + // but that doesn't affect the logic. switch t.r { case '<': /* @@ -126,9 +115,7 @@ decode_xml :: proc(input: string, options := XML_Decode_Options{}, allocator := in_data = _handle_xml_special(&t, &builder, options) or_return case ']': - /* - If we're unboxing _and_ decoding CDATA, we'll have to check for the end tag. - */ + // If we're unboxing _and_ decoding CDATA, we'll have to check for the end tag. if in_data { if t.read_offset + len(CDATA_END) < len(t.src) { if string(t.src[t.offset:][:len(CDATA_END)]) == CDATA_END { @@ -143,22 +130,16 @@ decode_xml :: proc(input: string, options := XML_Decode_Options{}, allocator := case: if in_data && .Decode_CDATA not_in options { - /* - Unboxed, but undecoded. - */ + // Unboxed, but undecoded. write_rune(&builder, t.r) continue } if t.r == '&' { if entity, entity_err := _extract_xml_entity(&t); entity_err != .None { - /* - We read to the end of the string without closing the entity. - Pass through as-is. - */ + // We read to the end of the string without closing the entity. Pass through as-is. write_string(&builder, entity) } else { - if .No_Entity_Decode not_in options { if decoded, ok := xml_decode_entity(entity); ok { write_rune(&builder, decoded) @@ -166,19 +147,41 @@ decode_xml :: proc(input: string, options := XML_Decode_Options{}, allocator := } } - /* - Literal passthrough because the decode failed or we want entities not decoded. - */ + // Literal passthrough because the decode failed or we want entities not decoded. write_string(&builder, "&") write_string(&builder, entity) write_string(&builder, ";") } } else { - write_rune(&builder, t.r) + // Handle AV Normalization: https://www.w3.org/TR/2006/REC-xml11-20060816/#AVNormalize + if .Normalize_Whitespace in options { + switch t.r { + case ' ', '\r', '\n', '\t': + if prev != ' ' { + write_rune(&builder, ' ') + prev = ' ' + } + case: + write_rune(&builder, t.r) + prev = t.r + } + } else { + // https://www.w3.org/TR/2006/REC-xml11-20060816/#sec-line-ends + switch t.r { + case '\n', 0x85, 0x2028: + write_rune(&builder, '\n') + case '\r': // Do nothing until next character + case: + if prev == '\r' { // Turn a single carriage return into a \n + write_rune(&builder, '\n') + } + write_rune(&builder, t.r) + } + prev = t.r + } } } } - return strings.clone(strings.to_string(builder), allocator), err } @@ -253,24 +256,18 @@ xml_decode_entity :: proc(entity: string) -> (decoded: rune, ok: bool) { return rune(val), true case: - /* - Named entity. - */ + // Named entity. return named_xml_entity_to_rune(entity) } } -/* - Private XML helper to extract `&<stuff>;` entity. -*/ +// Private XML helper to extract `&<stuff>;` entity. @(private="file") _extract_xml_entity :: proc(t: ^Tokenizer) -> (entity: string, err: Error) { assert(t != nil && t.r == '&') - /* - All of these would be in the ASCII range. - Even if one is not, it doesn't matter. All characters we need to compare to extract are. - */ + // All of these would be in the ASCII range. + // Even if one is not, it doesn't matter. All characters we need to compare to extract are. length := len(t.src) found := false @@ -292,9 +289,7 @@ _extract_xml_entity :: proc(t: ^Tokenizer) -> (entity: string, err: Error) { return string(t.src[t.offset : t.read_offset]), .Invalid_Entity_Encoding } -/* - Private XML helper for CDATA and comments. -*/ +// Private XML helper for CDATA and comments. @(private="file") _handle_xml_special :: proc(t: ^Tokenizer, builder: ^strings.Builder, options: XML_Decode_Options) -> (in_data: bool, err: Error) { assert(t != nil && t.r == '<') @@ -304,20 +299,14 @@ _handle_xml_special :: proc(t: ^Tokenizer, builder: ^strings.Builder, options: X t.read_offset += len(CDATA_START) - 1 if .Unbox_CDATA in options && .Decode_CDATA in options { - /* - We're unboxing _and_ decoding CDATA - */ + // We're unboxing _and_ decoding CDATA return true, .None } - /* - CDATA is passed through. - */ + // CDATA is passed through. offset := t.offset - /* - Scan until end of CDATA. - */ + // Scan until end of CDATA. for { advance(t) or_return if t.r < 0 { return true, .CDATA_Not_Terminated } @@ -341,14 +330,10 @@ _handle_xml_special :: proc(t: ^Tokenizer, builder: ^strings.Builder, options: X } else if string(t.src[t.offset:][:len(COMMENT_START)]) == COMMENT_START { t.read_offset += len(COMMENT_START) - /* - Comment is passed through by default. - */ + // Comment is passed through by default. offset := t.offset - /* - Scan until end of Comment. - */ + // Scan until end of Comment. for { advance(t) or_return if t.r < 0 { return true, .Comment_Not_Terminated } diff --git a/core/encoding/hex/hex.odin b/core/encoding/hex/hex.odin index dbffe216b..c2cd89c5b 100644 --- a/core/encoding/hex/hex.odin +++ b/core/encoding/hex/hex.odin @@ -2,8 +2,8 @@ package encoding_hex import "core:strings" -encode :: proc(src: []byte, allocator := context.allocator) -> []byte #no_bounds_check { - dst := make([]byte, len(src) * 2, allocator) +encode :: proc(src: []byte, allocator := context.allocator, loc := #caller_location) -> []byte #no_bounds_check { + dst := make([]byte, len(src) * 2, allocator, loc) for i, j := 0, 0; i < len(src); i += 1 { v := src[i] dst[j] = HEXTABLE[v>>4] @@ -15,12 +15,12 @@ encode :: proc(src: []byte, allocator := context.allocator) -> []byte #no_bounds } -decode :: proc(src: []byte, allocator := context.allocator) -> (dst: []byte, ok: bool) #no_bounds_check { +decode :: proc(src: []byte, allocator := context.allocator, loc := #caller_location) -> (dst: []byte, ok: bool) #no_bounds_check { if len(src) % 2 == 1 { return } - dst = make([]byte, len(src) / 2, allocator) + dst = make([]byte, len(src) / 2, allocator, loc) for i, j := 0, 1; j < len(src); j += 2 { p := src[j-1] q := src[j] @@ -69,5 +69,4 @@ hex_digit :: proc(char: byte) -> (u8, bool) { case 'A' ..= 'F': return char - 'A' + 10, true case: return 0, false } -} - +}
\ No newline at end of file diff --git a/core/encoding/hxa/hxa.odin b/core/encoding/hxa/hxa.odin index 9b24ede9c..9d0c58196 100644 --- a/core/encoding/hxa/hxa.odin +++ b/core/encoding/hxa/hxa.odin @@ -160,34 +160,35 @@ CONVENTION_SOFT_TRANSFORM :: "transform" /* destroy procedures */ -meta_destroy :: proc(meta: Meta, allocator := context.allocator) { +meta_destroy :: proc(meta: Meta, allocator := context.allocator, loc := #caller_location) { if nested, ok := meta.value.([]Meta); ok { for m in nested { - meta_destroy(m) + meta_destroy(m, loc=loc) } - delete(nested, allocator) + delete(nested, allocator, loc=loc) } } -nodes_destroy :: proc(nodes: []Node, allocator := context.allocator) { +nodes_destroy :: proc(nodes: []Node, allocator := context.allocator, loc := #caller_location) { for node in nodes { for meta in node.meta_data { - meta_destroy(meta) + meta_destroy(meta, loc=loc) } - delete(node.meta_data, allocator) + delete(node.meta_data, allocator, loc=loc) switch n in node.content { case Node_Geometry: - delete(n.corner_stack, allocator) - delete(n.edge_stack, allocator) - delete(n.face_stack, allocator) + delete(n.corner_stack, allocator, loc=loc) + delete(n.vertex_stack, allocator, loc=loc) + delete(n.edge_stack, allocator, loc=loc) + delete(n.face_stack, allocator, loc=loc) case Node_Image: - delete(n.image_stack, allocator) + delete(n.image_stack, allocator, loc=loc) } } - delete(nodes, allocator) + delete(nodes, allocator, loc=loc) } -file_destroy :: proc(file: File) { - nodes_destroy(file.nodes, file.allocator) - delete(file.backing, file.allocator) -} +file_destroy :: proc(file: File, loc := #caller_location) { + nodes_destroy(file.nodes, file.allocator, loc=loc) + delete(file.backing, file.allocator, loc=loc) +}
\ No newline at end of file diff --git a/core/encoding/hxa/read.odin b/core/encoding/hxa/read.odin index f37dc3193..5c8503229 100644 --- a/core/encoding/hxa/read.odin +++ b/core/encoding/hxa/read.odin @@ -11,24 +11,21 @@ Read_Error :: enum { Unable_To_Read_File, } -read_from_file :: proc(filename: string, print_error := false, allocator := context.allocator) -> (file: File, err: Read_Error) { +read_from_file :: proc(filename: string, print_error := false, allocator := context.allocator, loc := #caller_location) -> (file: File, err: Read_Error) { context.allocator = allocator - data, ok := os.read_entire_file(filename) + data, ok := os.read_entire_file(filename, allocator, loc) if !ok { err = .Unable_To_Read_File + delete(data, allocator, loc) return } - defer if !ok { - delete(data) - } else { - file.backing = data - } - file, err = read(data, filename, print_error, allocator) + file, err = read(data, filename, print_error, allocator, loc) + file.backing = data return } -read :: proc(data: []byte, filename := "<input>", print_error := false, allocator := context.allocator) -> (file: File, err: Read_Error) { +read :: proc(data: []byte, filename := "<input>", print_error := false, allocator := context.allocator, loc := #caller_location) -> (file: File, err: Read_Error) { Reader :: struct { filename: string, data: []byte, @@ -79,8 +76,8 @@ read :: proc(data: []byte, filename := "<input>", print_error := false, allocato return string(data[:len]), nil } - read_meta :: proc(r: ^Reader, capacity: u32le) -> (meta_data: []Meta, err: Read_Error) { - meta_data = make([]Meta, int(capacity)) + read_meta :: proc(r: ^Reader, capacity: u32le, allocator := context.allocator, loc := #caller_location) -> (meta_data: []Meta, err: Read_Error) { + meta_data = make([]Meta, int(capacity), allocator=allocator) count := 0 defer meta_data = meta_data[:count] for &m in meta_data { @@ -111,10 +108,10 @@ read :: proc(data: []byte, filename := "<input>", print_error := false, allocato return } - read_layer_stack :: proc(r: ^Reader, capacity: u32le) -> (layers: Layer_Stack, err: Read_Error) { + read_layer_stack :: proc(r: ^Reader, capacity: u32le, allocator := context.allocator, loc := #caller_location) -> (layers: Layer_Stack, err: Read_Error) { stack_count := read_value(r, u32le) or_return layer_count := 0 - layers = make(Layer_Stack, stack_count) + layers = make(Layer_Stack, stack_count, allocator=allocator, loc=loc) defer layers = layers[:layer_count] for &layer in layers { layer.name = read_name(r) or_return @@ -170,7 +167,8 @@ read :: proc(data: []byte, filename := "<input>", print_error := false, allocato node_count := 0 file.header = header^ - file.nodes = make([]Node, header.internal_node_count) + file.nodes = make([]Node, header.internal_node_count, allocator=allocator, loc=loc) + file.allocator = allocator defer if err != nil { nodes_destroy(file.nodes) file.nodes = nil @@ -198,15 +196,15 @@ read :: proc(data: []byte, filename := "<input>", print_error := false, allocato case .Geometry: g: Node_Geometry - g.vertex_count = read_value(r, u32le) or_return - g.vertex_stack = read_layer_stack(r, g.vertex_count) or_return - g.edge_corner_count = read_value(r, u32le) or_return - g.corner_stack = read_layer_stack(r, g.edge_corner_count) or_return + g.vertex_count = read_value(r, u32le) or_return + g.vertex_stack = read_layer_stack(r, g.vertex_count, loc=loc) or_return + g.edge_corner_count = read_value(r, u32le) or_return + g.corner_stack = read_layer_stack(r, g.edge_corner_count, loc=loc) or_return if header.version > 2 { - g.edge_stack = read_layer_stack(r, g.edge_corner_count) or_return + g.edge_stack = read_layer_stack(r, g.edge_corner_count, loc=loc) or_return } - g.face_count = read_value(r, u32le) or_return - g.face_stack = read_layer_stack(r, g.face_count) or_return + g.face_count = read_value(r, u32le) or_return + g.face_stack = read_layer_stack(r, g.face_count, loc=loc) or_return node.content = g @@ -233,4 +231,4 @@ read :: proc(data: []byte, filename := "<input>", print_error := false, allocato } return -} +}
\ No newline at end of file diff --git a/core/encoding/ini/ini.odin b/core/encoding/ini/ini.odin new file mode 100644 index 000000000..eb0ad9e7c --- /dev/null +++ b/core/encoding/ini/ini.odin @@ -0,0 +1,189 @@ +package encoding_ini + +import "base:runtime" +import "base:intrinsics" +import "core:strings" +import "core:strconv" +import "core:io" +import "core:os" +import "core:fmt" +_ :: fmt + +Options :: struct { + comment: string, + key_lower_case: bool, +} + +DEFAULT_OPTIONS :: Options { + comment = ";", + key_lower_case = false, +} + +Iterator :: struct { + section: string, + _src: string, + options: Options, +} + +iterator_from_string :: proc(src: string, options := DEFAULT_OPTIONS) -> Iterator { + return { + section = "", + options = options, + _src = src, + } +} + + +// Returns the raw `key` and `value`. `ok` will be false if no more key=value pairs cannot be found. +// They key and value may be quoted, which may require the use of `strconv.unquote_string`. +iterate :: proc(it: ^Iterator) -> (key, value: string, ok: bool) { + for line_ in strings.split_lines_iterator(&it._src) { + line := strings.trim_space(line_) + + if len(line) == 0 { + continue + } + + if line[0] == '[' { + end_idx := strings.index_byte(line, ']') + if end_idx < 0 { + end_idx = len(line) + } + it.section = line[1:end_idx] + continue + } + + if it.options.comment != "" && strings.has_prefix(line, it.options.comment) { + continue + } + + equal := strings.index(line, " =") // check for things keys that `ctrl+= = zoom_in` + quote := strings.index_byte(line, '"') + if equal < 0 || quote > 0 && quote < equal { + equal = strings.index_byte(line, '=') + if equal < 0 { + continue + } + } else { + equal += 1 + } + + key = strings.trim_space(line[:equal]) + value = strings.trim_space(line[equal+1:]) + ok = true + return + } + + it.section = "" + return +} + +Map :: distinct map[string]map[string]string + +load_map_from_string :: proc(src: string, allocator: runtime.Allocator, options := DEFAULT_OPTIONS) -> (m: Map, err: runtime.Allocator_Error) { + unquote :: proc(val: string) -> (string, runtime.Allocator_Error) { + v, allocated, ok := strconv.unquote_string(val) + if !ok { + return strings.clone(val) + } + if allocated { + return v, nil + } + return strings.clone(v) + + } + + context.allocator = allocator + + it := iterator_from_string(src, options) + + for key, value in iterate(&it) { + section := it.section + if section not_in m { + section = strings.clone(section) or_return + m[section] = {} + } + + // store key-value pair + pairs := &m[section] + new_key := unquote(key) or_return + if options.key_lower_case { + old_key := new_key + new_key = strings.to_lower(key) or_return + delete(old_key) or_return + } + pairs[new_key] = unquote(value) or_return + } + return +} + +load_map_from_path :: proc(path: string, allocator: runtime.Allocator, options := DEFAULT_OPTIONS) -> (m: Map, err: runtime.Allocator_Error, ok: bool) { + data := os.read_entire_file(path, allocator) or_return + defer delete(data, allocator) + m, err = load_map_from_string(string(data), allocator, options) + ok = err != nil + defer if !ok { + delete_map(m) + } + return +} + +save_map_to_string :: proc(m: Map, allocator: runtime.Allocator) -> (data: string) { + b := strings.builder_make(allocator) + _, _ = write_map(strings.to_writer(&b), m) + return strings.to_string(b) +} + +delete_map :: proc(m: Map) { + allocator := m.allocator + for section, pairs in m { + for key, value in pairs { + delete(key, allocator) + delete(value, allocator) + } + delete(section) + } + delete(m) +} + +write_section :: proc(w: io.Writer, name: string, n_written: ^int = nil) -> (n: int, err: io.Error) { + defer if n_written != nil { n_written^ += n } + io.write_byte (w, '[', &n) or_return + io.write_string(w, name, &n) or_return + io.write_byte (w, ']', &n) or_return + return +} + +write_pair :: proc(w: io.Writer, key: string, value: $T, n_written: ^int = nil) -> (n: int, err: io.Error) { + defer if n_written != nil { n_written^ += n } + io.write_string(w, key, &n) or_return + io.write_string(w, " = ", &n) or_return + when intrinsics.type_is_string(T) { + val := string(value) + if len(val) > 0 && (val[0] == ' ' || val[len(val)-1] == ' ') { + io.write_quoted_string(w, val, n_written=&n) or_return + } else { + io.write_string(w, val, &n) or_return + } + } else { + n += fmt.wprint(w, value) + } + io.write_byte(w, '\n', &n) or_return + return +} + +write_map :: proc(w: io.Writer, m: Map) -> (n: int, err: io.Error) { + section_index := 0 + for section, pairs in m { + if section_index == 0 && section == "" { + // ignore section + } else { + write_section(w, section, &n) or_return + } + for key, value in pairs { + write_pair(w, key, value, &n) or_return + } + section_index += 1 + } + return +} diff --git a/core/encoding/json/marshal.odin b/core/encoding/json/marshal.odin index b41a76856..dfca8b9db 100644 --- a/core/encoding/json/marshal.odin +++ b/core/encoding/json/marshal.odin @@ -62,8 +62,8 @@ Marshal_Options :: struct { mjson_skipped_first_braces_end: bool, } -marshal :: proc(v: any, opt: Marshal_Options = {}, allocator := context.allocator) -> (data: []byte, err: Marshal_Error) { - b := strings.builder_make(allocator) +marshal :: proc(v: any, opt: Marshal_Options = {}, allocator := context.allocator, loc := #caller_location) -> (data: []byte, err: Marshal_Error) { + b := strings.builder_make(allocator, loc) defer if err != nil { strings.builder_destroy(&b) } diff --git a/core/encoding/json/parser.odin b/core/encoding/json/parser.odin index 3973725dc..38f71edf6 100644 --- a/core/encoding/json/parser.odin +++ b/core/encoding/json/parser.odin @@ -28,27 +28,27 @@ make_parser_from_string :: proc(data: string, spec := DEFAULT_SPECIFICATION, par } -parse :: proc(data: []byte, spec := DEFAULT_SPECIFICATION, parse_integers := false, allocator := context.allocator) -> (Value, Error) { - return parse_string(string(data), spec, parse_integers, allocator) +parse :: proc(data: []byte, spec := DEFAULT_SPECIFICATION, parse_integers := false, allocator := context.allocator, loc := #caller_location) -> (Value, Error) { + return parse_string(string(data), spec, parse_integers, allocator, loc) } -parse_string :: proc(data: string, spec := DEFAULT_SPECIFICATION, parse_integers := false, allocator := context.allocator) -> (Value, Error) { +parse_string :: proc(data: string, spec := DEFAULT_SPECIFICATION, parse_integers := false, allocator := context.allocator, loc := #caller_location) -> (Value, Error) { context.allocator = allocator p := make_parser_from_string(data, spec, parse_integers, allocator) switch p.spec { case .JSON: - return parse_object(&p) + return parse_object(&p, loc) case .JSON5: - return parse_value(&p) + return parse_value(&p, loc) case .SJSON: #partial switch p.curr_token.kind { case .Ident, .String: - return parse_object_body(&p, .EOF) + return parse_object_body(&p, .EOF, loc) } - return parse_value(&p) + return parse_value(&p, loc) } - return parse_object(&p) + return parse_object(&p, loc) } token_end_pos :: proc(tok: Token) -> Pos { @@ -106,7 +106,7 @@ parse_comma :: proc(p: ^Parser) -> (do_break: bool) { return false } -parse_value :: proc(p: ^Parser) -> (value: Value, err: Error) { +parse_value :: proc(p: ^Parser, loc := #caller_location) -> (value: Value, err: Error) { err = .None token := p.curr_token #partial switch token.kind { @@ -142,13 +142,13 @@ parse_value :: proc(p: ^Parser) -> (value: Value, err: Error) { case .String: advance_token(p) - return unquote_string(token, p.spec, p.allocator) + return unquote_string(token, p.spec, p.allocator, loc) case .Open_Brace: - return parse_object(p) + return parse_object(p, loc) case .Open_Bracket: - return parse_array(p) + return parse_array(p, loc) case: if p.spec != .JSON { @@ -176,7 +176,7 @@ parse_value :: proc(p: ^Parser) -> (value: Value, err: Error) { return } -parse_array :: proc(p: ^Parser) -> (value: Value, err: Error) { +parse_array :: proc(p: ^Parser, loc := #caller_location) -> (value: Value, err: Error) { err = .None expect_token(p, .Open_Bracket) or_return @@ -184,14 +184,14 @@ parse_array :: proc(p: ^Parser) -> (value: Value, err: Error) { array.allocator = p.allocator defer if err != nil { for elem in array { - destroy_value(elem) + destroy_value(elem, loc=loc) } - delete(array) + delete(array, loc) } for p.curr_token.kind != .Close_Bracket { - elem := parse_value(p) or_return - append(&array, elem) + elem := parse_value(p, loc) or_return + append(&array, elem, loc) if parse_comma(p) { break @@ -228,38 +228,39 @@ clone_string :: proc(s: string, allocator: mem.Allocator, loc := #caller_locatio return } -parse_object_key :: proc(p: ^Parser, key_allocator: mem.Allocator) -> (key: string, err: Error) { +parse_object_key :: proc(p: ^Parser, key_allocator: mem.Allocator, loc := #caller_location) -> (key: string, err: Error) { tok := p.curr_token if p.spec != .JSON { if allow_token(p, .Ident) { - return clone_string(tok.text, key_allocator) + return clone_string(tok.text, key_allocator, loc) } } if tok_err := expect_token(p, .String); tok_err != nil { err = .Expected_String_For_Object_Key return } - return unquote_string(tok, p.spec, key_allocator) + return unquote_string(tok, p.spec, key_allocator, loc) } -parse_object_body :: proc(p: ^Parser, end_token: Token_Kind) -> (obj: Object, err: Error) { - obj.allocator = p.allocator +parse_object_body :: proc(p: ^Parser, end_token: Token_Kind, loc := #caller_location) -> (obj: Object, err: Error) { + obj = make(Object, allocator=p.allocator, loc=loc) + defer if err != nil { for key, elem in obj { - delete(key, p.allocator) - destroy_value(elem) + delete(key, p.allocator, loc) + destroy_value(elem, loc=loc) } - delete(obj) + delete(obj, loc) } for p.curr_token.kind != end_token { - key := parse_object_key(p, p.allocator) or_return + key := parse_object_key(p, p.allocator, loc) or_return parse_colon(p) or_return - elem := parse_value(p) or_return + elem := parse_value(p, loc) or_return if key in obj { err = .Duplicate_Object_Key - delete(key, p.allocator) + delete(key, p.allocator, loc) return } @@ -267,7 +268,7 @@ parse_object_body :: proc(p: ^Parser, end_token: Token_Kind) -> (obj: Object, er // inserting empty key/values into the object and for those we do not // want to allocate anything if key != "" { - reserve_error := reserve(&obj, len(obj) + 1) + reserve_error := reserve(&obj, len(obj) + 1, loc) if reserve_error == mem.Allocator_Error.Out_Of_Memory { return nil, .Out_Of_Memory } @@ -281,9 +282,9 @@ parse_object_body :: proc(p: ^Parser, end_token: Token_Kind) -> (obj: Object, er return obj, .None } -parse_object :: proc(p: ^Parser) -> (value: Value, err: Error) { +parse_object :: proc(p: ^Parser, loc := #caller_location) -> (value: Value, err: Error) { expect_token(p, .Open_Brace) or_return - obj := parse_object_body(p, .Close_Brace) or_return + obj := parse_object_body(p, .Close_Brace, loc) or_return expect_token(p, .Close_Brace) or_return return obj, .None } @@ -480,4 +481,4 @@ unquote_string :: proc(token: Token, spec: Specification, allocator := context.a } return string(b[:w]), nil -} +}
\ No newline at end of file diff --git a/core/encoding/json/types.odin b/core/encoding/json/types.odin index 73e183615..41eb21377 100644 --- a/core/encoding/json/types.odin +++ b/core/encoding/json/types.odin @@ -89,22 +89,22 @@ Error :: enum { -destroy_value :: proc(value: Value, allocator := context.allocator) { +destroy_value :: proc(value: Value, allocator := context.allocator, loc := #caller_location) { context.allocator = allocator #partial switch v in value { case Object: for key, elem in v { - delete(key) - destroy_value(elem) + delete(key, loc=loc) + destroy_value(elem, loc=loc) } - delete(v) + delete(v, loc=loc) case Array: for elem in v { - destroy_value(elem) + destroy_value(elem, loc=loc) } - delete(v) + delete(v, loc=loc) case String: - delete(v) + delete(v, loc=loc) } } diff --git a/core/encoding/xml/tokenizer.odin b/core/encoding/xml/tokenizer.odin index 0f87c366b..2d06038b7 100644 --- a/core/encoding/xml/tokenizer.odin +++ b/core/encoding/xml/tokenizer.odin @@ -218,9 +218,7 @@ scan_identifier :: proc(t: ^Tokenizer) -> string { for is_valid_identifier_rune(t.ch) { advance_rune(t) if t.ch == ':' { - /* - A namespaced attr can have at most two parts, `namespace:ident`. - */ + // A namespaced attr can have at most two parts, `namespace:ident`. if namespaced { break } @@ -268,14 +266,10 @@ scan_comment :: proc(t: ^Tokenizer) -> (comment: string, err: Error) { return string(t.src[offset : t.offset - 1]), .None } -/* - Skip CDATA -*/ +// Skip CDATA skip_cdata :: proc(t: ^Tokenizer) -> (err: Error) { if t.read_offset + len(CDATA_START) >= len(t.src) { - /* - Can't be the start of a CDATA tag. - */ + // Can't be the start of a CDATA tag. return .None } @@ -290,9 +284,7 @@ skip_cdata :: proc(t: ^Tokenizer) -> (err: Error) { return .Premature_EOF } - /* - Scan until the end of a CDATA tag. - */ + // Scan until the end of a CDATA tag. if t.read_offset + len(CDATA_END) < len(t.src) { if string(t.src[t.offset:][:len(CDATA_END)]) == CDATA_END { t.read_offset += len(CDATA_END) @@ -319,14 +311,10 @@ scan_string :: proc(t: ^Tokenizer, offset: int, close: rune = '<', consume_close case '<': if peek_byte(t) == '!' { if peek_byte(t, 1) == '[' { - /* - Might be the start of a CDATA tag. - */ + // Might be the start of a CDATA tag. skip_cdata(t) or_return } else if peek_byte(t, 1) == '-' && peek_byte(t, 2) == '-' { - /* - Comment start. Eat comment. - */ + // Comment start. Eat comment. t.read_offset += 3 _ = scan_comment(t) or_return } @@ -342,17 +330,13 @@ scan_string :: proc(t: ^Tokenizer, offset: int, close: rune = '<', consume_close } if t.ch == close { - /* - If it's not a CDATA or comment, it's the end of this body. - */ + // If it's not a CDATA or comment, it's the end of this body. break loop } advance_rune(t) } - /* - Strip trailing whitespace. - */ + // Strip trailing whitespace. lit := string(t.src[offset : t.offset]) end := len(lit) @@ -369,11 +353,6 @@ scan_string :: proc(t: ^Tokenizer, offset: int, close: rune = '<', consume_close if consume_close { advance_rune(t) } - - /* - TODO: Handle decoding escape characters and unboxing CDATA. - */ - return lit, err } @@ -384,7 +363,7 @@ peek :: proc(t: ^Tokenizer) -> (token: Token) { return token } -scan :: proc(t: ^Tokenizer) -> Token { +scan :: proc(t: ^Tokenizer, multiline_string := false) -> Token { skip_whitespace(t) offset := t.offset @@ -418,7 +397,7 @@ scan :: proc(t: ^Tokenizer) -> Token { case '"', '\'': kind = .Invalid - lit, err = scan_string(t, t.offset, ch, true, false) + lit, err = scan_string(t, t.offset, ch, true, multiline_string) if err == .None { kind = .String } @@ -435,4 +414,4 @@ scan :: proc(t: ^Tokenizer) -> Token { lit = string(t.src[offset : t.offset]) } return Token{kind, lit, pos} -} +}
\ No newline at end of file diff --git a/core/encoding/xml/xml_reader.odin b/core/encoding/xml/xml_reader.odin index 5b4b12948..b9656900f 100644 --- a/core/encoding/xml/xml_reader.odin +++ b/core/encoding/xml/xml_reader.odin @@ -203,9 +203,7 @@ parse_bytes :: proc(data: []u8, options := DEFAULT_OPTIONS, path := "", error_ha doc.elements = make([dynamic]Element, 1024, 1024, allocator) - // strings.intern_init(&doc.intern, allocator, allocator) - - err = .Unexpected_Token + err = .Unexpected_Token element, parent: Element_ID open: Token @@ -259,8 +257,8 @@ parse_bytes :: proc(data: []u8, options := DEFAULT_OPTIONS, path := "", error_ha case .Slash: // Empty tag. Close it. expect(t, .Gt) or_return - parent = doc.elements[element].parent - element = parent + parent = doc.elements[element].parent + element = parent case: error(t, t.offset, "Expected close tag, got: %#v\n", end_token) @@ -276,8 +274,8 @@ parse_bytes :: proc(data: []u8, options := DEFAULT_OPTIONS, path := "", error_ha error(t, t.offset, "Mismatched Closing Tag. Expected %v, got %v\n", doc.elements[element].ident, ident.text) return doc, .Mismatched_Closing_Tag } - parent = doc.elements[element].parent - element = parent + parent = doc.elements[element].parent + element = parent } else if open.kind == .Exclaim { // <! @@ -463,8 +461,8 @@ validate_options :: proc(options: Options) -> (validated: Options, err: Error) { return validated, .None } -expect :: proc(t: ^Tokenizer, kind: Token_Kind) -> (tok: Token, err: Error) { - tok = scan(t) +expect :: proc(t: ^Tokenizer, kind: Token_Kind, multiline_string := false) -> (tok: Token, err: Error) { + tok = scan(t, multiline_string=multiline_string) if tok.kind == kind { return tok, .None } error(t, t.offset, "Expected \"%v\", got \"%v\".", kind, tok.kind) @@ -480,7 +478,13 @@ parse_attribute :: proc(doc: ^Document) -> (attr: Attribute, offset: int, err: E offset = t.offset - len(key.text) _ = expect(t, .Eq) or_return - value := expect(t, .String) or_return + value := expect(t, .String, multiline_string=true) or_return + + normalized, normalize_err := entity.decode_xml(value.text, {.Normalize_Whitespace}, doc.allocator) + if normalize_err == .None { + append(&doc.strings_to_free, normalized) + value.text = normalized + } attr.key = key.text attr.val = value.text |