lang/builtin/builtin.go - external/github.com/google/wuffs - Git at Google

 // Copyright 2017 The Wuffs Authors.
 //
 // Licensed under the Apache License, Version 2.0 (the "License");
 // you may not use this file except in compliance with the License.
 // You may obtain a copy of the License at
 //
 //    https://www.apache.org/licenses/LICENSE-2.0
 //
 // Unless required by applicable law or agreed to in writing, software
 // distributed under the License is distributed on an "AS IS" BASIS,
 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 // See the License for the specific language governing permissions and
 // limitations under the License.

 // Package builtin lists Wuffs' built-in concepts such as status codes.
 package builtin

 import (
 	"fmt"

 	"github.com/google/wuffs/lang/parse"

 	a "github.com/google/wuffs/lang/ast"
 	t "github.com/google/wuffs/lang/token"
 )

 var FourCCs = [...][2]string{
 	{"ICCP", "International Color Consortium Profile"},
 	{"XMP ", "Extensible Metadata Platform"},
 }

 var Statuses = [...]string{
 	// Notes.
 	`"@end of data"`,
 	`"@metadata reported"`,

 	// Suspensions.
 	`"$short read"`,
 	`"$short write"`,

 	// Errors.
 	`"#bad I/O position"`,
 	`"#bad argument (length too short)"`,
 	`"#bad argument"`,
 	`"#bad call sequence"`,
 	`"#bad receiver"`,
 	`"#bad restart"`,
 	`"#bad sizeof receiver"`,
 	`"#bad vtable"`,
 	`"#bad workbuf length"`,
 	`"#bad wuffs version"`,
 	`"#cannot return a suspension"`,
 	`"#disabled by previous error"`,
 	`"#initialize falsely claimed already zeroed"`,
 	`"#initialize not called"`,
 	`"#interleaved coroutine calls"`,
 	`"#not enough data"`,
 	`"#unsupported option"`,
 	`"#too much data"`,
 }

 // TODO: a collection of forbidden variable names like and, or, not, as, false,
 // true, in, out, this, u8, u16, etc?

 var Types = []string{
 	// TODO: i8, i16, i32, i64.
 	"u8",
 	"u16",
 	"u32",
 	"u64",

 	"empty_struct",
 	"bool",
 	"utility",

 	"range_ii_u32",
 	"range_ie_u32",
 	"range_ii_u64",
 	"range_ie_u64",
 	"rect_ie_u32",
 	"rect_ii_u32",

 	"io_reader",
 	"io_writer",
 	"status",

 	"frame_config",
 	"image_config",
 	"pixel_buffer",
 	"pixel_config",
 	"pixel_format",
 	"pixel_swizzler",

 	"decode_frame_options",
 }

 var Funcs = []string{
 	"u8.high_bits(n: u32[..= 8]) u8",
 	"u8.low_bits(n: u32[..= 8]) u8",
 	"u8.max(a: u8) u8",
 	"u8.min(a: u8) u8",

 	"u16.high_bits(n: u32[..= 16]) u16",
 	"u16.low_bits(n: u32[..= 16]) u16",
 	"u16.max(a: u16) u16",
 	"u16.min(a: u16) u16",

 	"u32.high_bits(n: u32[..= 32]) u32",
 	"u32.low_bits(n: u32[..= 32]) u32",
 	"u32.max(a: u32) u32",
 	"u32.min(a: u32) u32",

 	"u64.high_bits(n: u32[..= 64]) u64",
 	"u64.low_bits(n: u32[..= 64]) u64",
 	"u64.max(a: u64) u64",
 	"u64.min(a: u64) u64",

 	// ---- utility

 	"utility.empty_io_reader() io_reader",
 	"utility.empty_io_writer() io_writer",
 	"utility.empty_slice_u8() slice u8",
 	"utility.make_pixel_format(repr: u32) pixel_format",
 	"utility.make_range_ii_u32(min_incl: u32, max_incl: u32) range_ii_u32",
 	"utility.make_range_ie_u32(min_incl: u32, max_excl: u32) range_ie_u32",
 	"utility.make_range_ii_u64(min_incl: u64, max_incl: u64) range_ii_u64",
 	"utility.make_range_ie_u64(min_incl: u64, max_excl: u64) range_ie_u64",
 	"utility.make_rect_ii_u32(" +
 		"min_incl_x: u32, min_incl_y: u32, max_incl_x: u32, max_incl_y: u32) rect_ii_u32",
 	"utility.make_rect_ie_u32(" +
 		"min_incl_x: u32, min_incl_y: u32, max_excl_x: u32, max_excl_y: u32) rect_ie_u32",

 	// ---- ranges

 	"range_ie_u32.reset!()",
 	"range_ie_u32.get_min_incl() u32",
 	"range_ie_u32.get_max_excl() u32",
 	"range_ie_u32.intersect(r: range_ie_u32) range_ie_u32",
 	"range_ie_u32.unite(r: range_ie_u32) range_ie_u32",

 	"range_ii_u32.reset!()",
 	"range_ii_u32.get_min_incl() u32",
 	"range_ii_u32.get_max_incl() u32",
 	"range_ii_u32.intersect(r: range_ii_u32) range_ii_u32",
 	"range_ii_u32.unite(r: range_ii_u32) range_ii_u32",

 	"range_ie_u64.reset!()",
 	"range_ie_u64.get_min_incl() u64",
 	"range_ie_u64.get_max_excl() u64",
 	"range_ie_u64.intersect(r: range_ie_u64) range_ie_u64",
 	"range_ie_u64.unite(r: range_ie_u64) range_ie_u64",

 	"range_ii_u64.reset!()",
 	"range_ii_u64.get_min_incl() u64",
 	"range_ii_u64.get_max_incl() u64",
 	"range_ii_u64.intersect(r: range_ii_u64) range_ii_u64",
 	"range_ii_u64.unite(r: range_ii_u64) range_ii_u64",

 	// ---- io_reader

 	"io_reader.can_undo_byte() bool",
 	"io_reader.undo_byte!()",

 	"io_reader.read_u8?() u8",

 	"io_reader.read_u16be?() u16",
 	"io_reader.read_u16le?() u16",

 	"io_reader.read_u8_as_u32?() u32[..= 0xFF]",
 	"io_reader.read_u16be_as_u32?() u32[..= 0xFFFF]",
 	"io_reader.read_u16le_as_u32?() u32[..= 0xFFFF]",
 	"io_reader.read_u24be_as_u32?() u32[..= 0xFFFFFF]",
 	"io_reader.read_u24le_as_u32?() u32[..= 0xFFFFFF]",
 	"io_reader.read_u32be?() u32",
 	"io_reader.read_u32le?() u32",

 	"io_reader.read_u8_as_u64?() u64[..= 0xFF]",
 	"io_reader.read_u16be_as_u64?() u64[..= 0xFFFF]",
 	"io_reader.read_u16le_as_u64?() u64[..= 0xFFFF]",
 	"io_reader.read_u24be_as_u64?() u64[..= 0xFFFFFF]",
 	"io_reader.read_u24le_as_u64?() u64[..= 0xFFFFFF]",
 	"io_reader.read_u32be_as_u64?() u64[..= 0xFFFFFFFF]",
 	"io_reader.read_u32le_as_u64?() u64[..= 0xFFFFFFFF]",
 	"io_reader.read_u40be_as_u64?() u64[..= 0xFFFFFFFFFF]",
 	"io_reader.read_u40le_as_u64?() u64[..= 0xFFFFFFFFFF]",
 	"io_reader.read_u48be_as_u64?() u64[..= 0xFFFFFFFFFFFF]",
 	"io_reader.read_u48le_as_u64?() u64[..= 0xFFFFFFFFFFFF]",
 	"io_reader.read_u56be_as_u64?() u64[..= 0xFFFFFFFFFFFFFF]",
 	"io_reader.read_u56le_as_u64?() u64[..= 0xFFFFFFFFFFFFFF]",
 	"io_reader.read_u64be?() u64",
 	"io_reader.read_u64le?() u64",

 	// TODO: these should have an explicit pre-condition "available() >= N".
 	// For now, that's implicitly checked (i.e. hard coded).
 	//
 	// The io_reader has peek_etc methods and skip_fast, not read_etc_fast,
 	// because we sometimes advance the pointer by less than what's read. See
 	// https://fgiesen.wordpress.com/2018/02/20/reading-bits-in-far-too-many-ways-part-2/

 	"io_reader.peek_u8() u8",

 	"io_reader.peek_u16be() u16",
 	"io_reader.peek_u16le() u16",

 	"io_reader.peek_u8_as_u32() u32[..= 0xFF]",
 	"io_reader.peek_u16be_as_u32() u32[..= 0xFFFF]",
 	"io_reader.peek_u16le_as_u32() u32[..= 0xFFFF]",
 	"io_reader.peek_u24be_as_u32() u32[..= 0xFFFFFF]",
 	"io_reader.peek_u24le_as_u32() u32[..= 0xFFFFFF]",
 	"io_reader.peek_u32be() u32",
 	"io_reader.peek_u32le() u32",

 	"io_reader.peek_u8_as_u64() u64[..= 0xFF]",
 	"io_reader.peek_u16be_as_u64() u64[..= 0xFFFF]",
 	"io_reader.peek_u16le_as_u64() u64[..= 0xFFFF]",
 	"io_reader.peek_u24be_as_u64() u64[..= 0xFFFFFF]",
 	"io_reader.peek_u24le_as_u64() u64[..= 0xFFFFFF]",
 	"io_reader.peek_u32be_as_u64() u64[..= 0xFFFFFFFF]",
 	"io_reader.peek_u32le_as_u64() u64[..= 0xFFFFFFFF]",
 	"io_reader.peek_u40be_as_u64() u64[..= 0xFFFFFFFFFF]",
 	"io_reader.peek_u40le_as_u64() u64[..= 0xFFFFFFFFFF]",
 	"io_reader.peek_u48be_as_u64() u64[..= 0xFFFFFFFFFFFF]",
 	"io_reader.peek_u48le_as_u64() u64[..= 0xFFFFFFFFFFFF]",
 	"io_reader.peek_u56be_as_u64() u64[..= 0xFFFFFFFFFFFFFF]",
 	"io_reader.peek_u56le_as_u64() u64[..= 0xFFFFFFFFFFFFFF]",
 	"io_reader.peek_u64be() u64",
 	"io_reader.peek_u64le() u64",

 	"io_reader.available() u64",
 	"io_reader.count_since(mark: u64) u64",
 	"io_reader.mark() u64",
 	"io_reader.position() u64",
 	"io_reader.since(mark: u64) slice u8",
 	"io_reader.take!(n: u64) slice u8",

 	"io_reader.skip?(n: u32)",

 	// TODO: this should have explicit pre-conditions "actual <= worst_case"
 	// and "worst_case <= available()". As an implementation restriction, we
 	// also require that worst_case has a constant value. For now, that's all
 	// implicitly checked (i.e. hard coded).
 	"io_reader.skip_fast!(actual: u32, worst_case: u32)",

 	// ---- io_writer

 	"io_writer.write_u8?(a: u8)",
 	"io_writer.write_u16be?(a: u16)",
 	"io_writer.write_u16le?(a: u16)",
 	"io_writer.write_u24be?(a: u32[..= 0xFFFFFF])",
 	"io_writer.write_u24le?(a: u32[..= 0xFFFFFF])",
 	"io_writer.write_u32be?(a: u32)",
 	"io_writer.write_u32le?(a: u32)",
 	"io_writer.write_u40be?(a: u64[..= 0xFFFFFFFFFF])",
 	"io_writer.write_u40le?(a: u64[..= 0xFFFFFFFFFF])",
 	"io_writer.write_u48be?(a: u64[..= 0xFFFFFFFFFFFF])",
 	"io_writer.write_u48le?(a: u64[..= 0xFFFFFFFFFFFF])",
 	"io_writer.write_u56be?(a: u64[..= 0xFFFFFFFFFFFFFF])",
 	"io_writer.write_u56le?(a: u64[..= 0xFFFFFFFFFFFFFF])",
 	"io_writer.write_u64be?(a: u64)",
 	"io_writer.write_u64le?(a: u64)",

 	// TODO: these should have an explicit pre-condition "available() >= N".
 	// For now, that's implicitly checked (i.e. hard coded).
 	//
 	// The io_writer has write_fast_etc methods, not poke_etc and skip_fast,
 	// because skip_fast could leave uninitialized bytes in the io_buffer.
 	"io_writer.write_fast_u8!(a: u8)",
 	"io_writer.write_fast_u16be!(a: u16)",
 	"io_writer.write_fast_u16le!(a: u16)",
 	"io_writer.write_fast_u24be!(a: u32[..= 0xFFFFFF])",
 	"io_writer.write_fast_u24le!(a: u32[..= 0xFFFFFF])",
 	"io_writer.write_fast_u32be!(a: u32)",
 	"io_writer.write_fast_u32le!(a: u32)",
 	"io_writer.write_fast_u40be!(a: u64[..= 0xFFFFFFFFFF])",
 	"io_writer.write_fast_u40le!(a: u64[..= 0xFFFFFFFFFF])",
 	"io_writer.write_fast_u48be!(a: u64[..= 0xFFFFFFFFFFFF])",
 	"io_writer.write_fast_u48le!(a: u64[..= 0xFFFFFFFFFFFF])",
 	"io_writer.write_fast_u56be!(a: u64[..= 0xFFFFFFFFFFFFFF])",
 	"io_writer.write_fast_u56le!(a: u64[..= 0xFFFFFFFFFFFFFF])",
 	"io_writer.write_fast_u64be!(a: u64)",
 	"io_writer.write_fast_u64le!(a: u64)",

 	"io_writer.available() u64",
 	"io_writer.count_since(mark: u64) u64",
 	"io_writer.history_available() u64",
 	"io_writer.mark() u64",
 	"io_writer.position() u64",
 	"io_writer.since(mark: u64) slice u8",

 	"io_writer.copy_from_slice!(s: slice u8) u64",
 	"io_writer.copy_n_from_history!(n: u32, distance: u32) u32",
 	"io_writer.copy_n_from_reader!(n: u32, r: io_reader) u32",
 	"io_writer.copy_n_from_slice!(n: u32, s: slice u8) u32",

 	// TODO: this should have explicit pre-conditions:
 	//  - n <= this.available()
 	//  - distance > 0
 	//  - distance <= this.since_mark().length()
 	// For now, that's all implicitly checked (i.e. hard coded).
 	"io_writer.copy_n_from_history_fast!(n: u32, distance: u32) u32",

 	// ---- status

 	// TODO: should we add is_complete?
 	"status.is_error() bool",
 	"status.is_note() bool",
 	"status.is_ok() bool",
 	"status.is_suspension() bool",

 	// ---- frame_config
 	// Duration's upper bound is the maximum possible i64 value.

 	"frame_config.blend() u8",
 	"frame_config.disposal() u8",
 	"frame_config.duration() u64[..= 0x7FFFFFFFFFFFFFFF]",
 	"frame_config.index() u64",
 	"frame_config.io_position() u64",

 	"frame_config.update!(bounds: rect_ie_u32, duration: u64[..= 0x7FFFFFFFFFFFFFFF], " +
 		"index: u64, io_position: u64, blend: u8, disposal: u8, background_color: u32)",

 	// ---- image_config

 	"image_config.set!(pixfmt: u32, pixsub: u32, width: u32, height: u32, " +
 		"first_frame_io_position: u64, first_frame_is_opaque: bool)",

 	// ---- pixel_buffer

 	"pixel_buffer.palette() slice u8",
 	"pixel_buffer.pixel_format() pixel_format",
 	"pixel_buffer.plane(p: u32[..= 3]) table u8",

 	// ---- pixel_format

 	"pixel_format.bits_per_pixel() u32",

 	// ---- pixel_swizzler

 	"pixel_swizzler.prepare!(" +
 		"dst_pixfmt: pixel_format, dst_palette: slice u8, " +
 		"src_pixfmt: pixel_format, src_palette: slice u8) status",
 	"pixel_swizzler.swizzle_interleaved!(" +
 		"dst: slice u8, dst_palette: slice u8, src: slice u8) u64",
 }

 var Interfaces = []string{
 	"hasher_u32",
 }

 var InterfaceFuncs = []string{
 	// ---- hasher_u32

 	"hasher_u32.update_u32!(x: slice u8) u32",
 }

 // The "T1" and "T2" types here are placeholders for generic "slice T" or
 // "table T" types. After tokenizing (but before parsing) these XxxFunc strings
 // (e.g. in the lang/check package), replace "T1" and "T2" with "†" or "‡"
 // daggers, to avoid collision with a user-defined "T1" or "T2" type.

 const (
 	GenericOldName1 = t.IDT1
 	GenericOldName2 = t.IDT2
 	GenericNewName1 = t.IDDagger1
 	GenericNewName2 = t.IDDagger2
 )

 var SliceFuncs = []string{
 	"T1.copy_from_slice!(s: T1) u64",
 	"T1.length() u64",
 	"T1.prefix(up_to: u64) T1",
 	"T1.suffix(up_to: u64) T1",
 }

 var TableFuncs = []string{
 	"T2.height() u64",
 	"T2.stride() u64",
 	"T2.width() u64",

 	"T2.row(y: u32) T1",
 }

 func ParseFuncs(tm *t.Map, ss []string, generic bool, callback func(*a.Func) error) error {
 	buf := []byte(nil)
 	for _, s := range ss {
 		buf = buf[:0]
 		buf = append(buf, "pub func "...)
 		buf = append(buf, s...)
 		buf = append(buf, "{}\n"...)

 		const filename = "builtin.wuffs"
 		tokens, _, err := t.Tokenize(tm, filename, buf)
 		if err != nil {
 			return fmt.Errorf("parsing %q: could not tokenize built-in funcs: %v", s, err)
 		}
 		if generic {
 			for i := range tokens {
 				if tokens[i].ID == GenericOldName1 {
 					tokens[i].ID = GenericNewName1
 				} else if tokens[i].ID == GenericOldName2 {
 					tokens[i].ID = GenericNewName2
 				}
 			}
 		}
 		file, err := parse.Parse(tm, filename, tokens, &parse.Options{
 			AllowBuiltInNames: true,
 		})
 		if err != nil {
 			return fmt.Errorf("parsing %q: could not parse built-in funcs: %v", s, err)
 		}

 		tlds := file.TopLevelDecls()
 		if len(tlds) != 1 || tlds[0].Kind() != a.KFunc {
 			return fmt.Errorf("parsing %q: got %d top level decls, want %d", s, len(tlds), 1)
 		}
 		f := tlds[0].AsFunc()
 		f.AsNode().AsRaw().SetPackage(tm, t.IDBase)

 		if callback != nil {
 			if err := callback(f); err != nil {
 				return err
 			}
 		}
 	}
 	return nil
 }
	// Copyright 2017 The Wuffs Authors.
	//
	// Licensed under the Apache License, Version 2.0 (the "License");
	// you may not use this file except in compliance with the License.
	// You may obtain a copy of the License at
	//
	// https://www.apache.org/licenses/LICENSE-2.0
	//
	// Unless required by applicable law or agreed to in writing, software
	// distributed under the License is distributed on an "AS IS" BASIS,
	// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
	// See the License for the specific language governing permissions and
	// limitations under the License.

	// Package builtin lists Wuffs' built-in concepts such as status codes.
	package builtin

	import (
	"fmt"

	"github.com/google/wuffs/lang/parse"

	a "github.com/google/wuffs/lang/ast"
	t "github.com/google/wuffs/lang/token"
	)

	var FourCCs = [...][2]string{
	{"ICCP", "International Color Consortium Profile"},
	{"XMP ", "Extensible Metadata Platform"},
	}

	var Statuses = [...]string{
	// Notes.
	`"@end of data"`,
	`"@metadata reported"`,

	// Suspensions.
	`"$short read"`,
	`"$short write"`,

	// Errors.
	`"#bad I/O position"`,
	`"#bad argument (length too short)"`,
	`"#bad argument"`,
	`"#bad call sequence"`,
	`"#bad receiver"`,
	`"#bad restart"`,
	`"#bad sizeof receiver"`,
	`"#bad vtable"`,
	`"#bad workbuf length"`,
	`"#bad wuffs version"`,
	`"#cannot return a suspension"`,
	`"#disabled by previous error"`,
	`"#initialize falsely claimed already zeroed"`,
	`"#initialize not called"`,
	`"#interleaved coroutine calls"`,
	`"#not enough data"`,
	`"#unsupported option"`,
	`"#too much data"`,
	}

	// TODO: a collection of forbidden variable names like and, or, not, as, false,
	// true, in, out, this, u8, u16, etc?

	var Types = []string{
	// TODO: i8, i16, i32, i64.
	"u8",
	"u16",
	"u32",
	"u64",

	"empty_struct",
	"bool",
	"utility",

	"range_ii_u32",
	"range_ie_u32",
	"range_ii_u64",
	"range_ie_u64",
	"rect_ie_u32",
	"rect_ii_u32",

	"io_reader",
	"io_writer",
	"status",

	"frame_config",
	"image_config",
	"pixel_buffer",
	"pixel_config",
	"pixel_format",
	"pixel_swizzler",

	"decode_frame_options",
	}

	var Funcs = []string{
	"u8.high_bits(n: u32[..= 8]) u8",
	"u8.low_bits(n: u32[..= 8]) u8",
	"u8.max(a: u8) u8",
	"u8.min(a: u8) u8",

	"u16.high_bits(n: u32[..= 16]) u16",
	"u16.low_bits(n: u32[..= 16]) u16",
	"u16.max(a: u16) u16",
	"u16.min(a: u16) u16",

	"u32.high_bits(n: u32[..= 32]) u32",
	"u32.low_bits(n: u32[..= 32]) u32",
	"u32.max(a: u32) u32",
	"u32.min(a: u32) u32",

	"u64.high_bits(n: u32[..= 64]) u64",
	"u64.low_bits(n: u32[..= 64]) u64",
	"u64.max(a: u64) u64",
	"u64.min(a: u64) u64",

	// ---- utility

	"utility.empty_io_reader() io_reader",
	"utility.empty_io_writer() io_writer",
	"utility.empty_slice_u8() slice u8",
	"utility.make_pixel_format(repr: u32) pixel_format",
	"utility.make_range_ii_u32(min_incl: u32, max_incl: u32) range_ii_u32",
	"utility.make_range_ie_u32(min_incl: u32, max_excl: u32) range_ie_u32",
	"utility.make_range_ii_u64(min_incl: u64, max_incl: u64) range_ii_u64",
	"utility.make_range_ie_u64(min_incl: u64, max_excl: u64) range_ie_u64",
	"utility.make_rect_ii_u32(" +
	"min_incl_x: u32, min_incl_y: u32, max_incl_x: u32, max_incl_y: u32) rect_ii_u32",
	"utility.make_rect_ie_u32(" +
	"min_incl_x: u32, min_incl_y: u32, max_excl_x: u32, max_excl_y: u32) rect_ie_u32",

	// ---- ranges

	"range_ie_u32.reset!()",
	"range_ie_u32.get_min_incl() u32",
	"range_ie_u32.get_max_excl() u32",
	"range_ie_u32.intersect(r: range_ie_u32) range_ie_u32",
	"range_ie_u32.unite(r: range_ie_u32) range_ie_u32",

	"range_ii_u32.reset!()",
	"range_ii_u32.get_min_incl() u32",
	"range_ii_u32.get_max_incl() u32",
	"range_ii_u32.intersect(r: range_ii_u32) range_ii_u32",
	"range_ii_u32.unite(r: range_ii_u32) range_ii_u32",

	"range_ie_u64.reset!()",
	"range_ie_u64.get_min_incl() u64",
	"range_ie_u64.get_max_excl() u64",
	"range_ie_u64.intersect(r: range_ie_u64) range_ie_u64",
	"range_ie_u64.unite(r: range_ie_u64) range_ie_u64",

	"range_ii_u64.reset!()",
	"range_ii_u64.get_min_incl() u64",
	"range_ii_u64.get_max_incl() u64",
	"range_ii_u64.intersect(r: range_ii_u64) range_ii_u64",
	"range_ii_u64.unite(r: range_ii_u64) range_ii_u64",

	// ---- io_reader

	"io_reader.can_undo_byte() bool",
	"io_reader.undo_byte!()",

	"io_reader.read_u8?() u8",

	"io_reader.read_u16be?() u16",
	"io_reader.read_u16le?() u16",

	"io_reader.read_u8_as_u32?() u32[..= 0xFF]",
	"io_reader.read_u16be_as_u32?() u32[..= 0xFFFF]",
	"io_reader.read_u16le_as_u32?() u32[..= 0xFFFF]",
	"io_reader.read_u24be_as_u32?() u32[..= 0xFFFFFF]",
	"io_reader.read_u24le_as_u32?() u32[..= 0xFFFFFF]",
	"io_reader.read_u32be?() u32",
	"io_reader.read_u32le?() u32",

	"io_reader.read_u8_as_u64?() u64[..= 0xFF]",
	"io_reader.read_u16be_as_u64?() u64[..= 0xFFFF]",
	"io_reader.read_u16le_as_u64?() u64[..= 0xFFFF]",
	"io_reader.read_u24be_as_u64?() u64[..= 0xFFFFFF]",
	"io_reader.read_u24le_as_u64?() u64[..= 0xFFFFFF]",
	"io_reader.read_u32be_as_u64?() u64[..= 0xFFFFFFFF]",
	"io_reader.read_u32le_as_u64?() u64[..= 0xFFFFFFFF]",
	"io_reader.read_u40be_as_u64?() u64[..= 0xFFFFFFFFFF]",
	"io_reader.read_u40le_as_u64?() u64[..= 0xFFFFFFFFFF]",
	"io_reader.read_u48be_as_u64?() u64[..= 0xFFFFFFFFFFFF]",
	"io_reader.read_u48le_as_u64?() u64[..= 0xFFFFFFFFFFFF]",
	"io_reader.read_u56be_as_u64?() u64[..= 0xFFFFFFFFFFFFFF]",
	"io_reader.read_u56le_as_u64?() u64[..= 0xFFFFFFFFFFFFFF]",
	"io_reader.read_u64be?() u64",
	"io_reader.read_u64le?() u64",

	// TODO: these should have an explicit pre-condition "available() >= N".
	// For now, that's implicitly checked (i.e. hard coded).
	//
	// The io_reader has peek_etc methods and skip_fast, not read_etc_fast,
	// because we sometimes advance the pointer by less than what's read. See
	// https://fgiesen.wordpress.com/2018/02/20/reading-bits-in-far-too-many-ways-part-2/

	"io_reader.peek_u8() u8",

	"io_reader.peek_u16be() u16",
	"io_reader.peek_u16le() u16",

	"io_reader.peek_u8_as_u32() u32[..= 0xFF]",
	"io_reader.peek_u16be_as_u32() u32[..= 0xFFFF]",
	"io_reader.peek_u16le_as_u32() u32[..= 0xFFFF]",
	"io_reader.peek_u24be_as_u32() u32[..= 0xFFFFFF]",
	"io_reader.peek_u24le_as_u32() u32[..= 0xFFFFFF]",
	"io_reader.peek_u32be() u32",
	"io_reader.peek_u32le() u32",

	"io_reader.peek_u8_as_u64() u64[..= 0xFF]",
	"io_reader.peek_u16be_as_u64() u64[..= 0xFFFF]",
	"io_reader.peek_u16le_as_u64() u64[..= 0xFFFF]",
	"io_reader.peek_u24be_as_u64() u64[..= 0xFFFFFF]",
	"io_reader.peek_u24le_as_u64() u64[..= 0xFFFFFF]",
	"io_reader.peek_u32be_as_u64() u64[..= 0xFFFFFFFF]",
	"io_reader.peek_u32le_as_u64() u64[..= 0xFFFFFFFF]",
	"io_reader.peek_u40be_as_u64() u64[..= 0xFFFFFFFFFF]",
	"io_reader.peek_u40le_as_u64() u64[..= 0xFFFFFFFFFF]",
	"io_reader.peek_u48be_as_u64() u64[..= 0xFFFFFFFFFFFF]",
	"io_reader.peek_u48le_as_u64() u64[..= 0xFFFFFFFFFFFF]",
	"io_reader.peek_u56be_as_u64() u64[..= 0xFFFFFFFFFFFFFF]",
	"io_reader.peek_u56le_as_u64() u64[..= 0xFFFFFFFFFFFFFF]",
	"io_reader.peek_u64be() u64",
	"io_reader.peek_u64le() u64",

	"io_reader.available() u64",
	"io_reader.count_since(mark: u64) u64",
	"io_reader.mark() u64",
	"io_reader.position() u64",
	"io_reader.since(mark: u64) slice u8",
	"io_reader.take!(n: u64) slice u8",

	"io_reader.skip?(n: u32)",

	// TODO: this should have explicit pre-conditions "actual <= worst_case"
	// and "worst_case <= available()". As an implementation restriction, we
	// also require that worst_case has a constant value. For now, that's all
	// implicitly checked (i.e. hard coded).
	"io_reader.skip_fast!(actual: u32, worst_case: u32)",

	// ---- io_writer

	"io_writer.write_u8?(a: u8)",
	"io_writer.write_u16be?(a: u16)",
	"io_writer.write_u16le?(a: u16)",
	"io_writer.write_u24be?(a: u32[..= 0xFFFFFF])",
	"io_writer.write_u24le?(a: u32[..= 0xFFFFFF])",
	"io_writer.write_u32be?(a: u32)",
	"io_writer.write_u32le?(a: u32)",
	"io_writer.write_u40be?(a: u64[..= 0xFFFFFFFFFF])",
	"io_writer.write_u40le?(a: u64[..= 0xFFFFFFFFFF])",
	"io_writer.write_u48be?(a: u64[..= 0xFFFFFFFFFFFF])",
	"io_writer.write_u48le?(a: u64[..= 0xFFFFFFFFFFFF])",
	"io_writer.write_u56be?(a: u64[..= 0xFFFFFFFFFFFFFF])",
	"io_writer.write_u56le?(a: u64[..= 0xFFFFFFFFFFFFFF])",
	"io_writer.write_u64be?(a: u64)",
	"io_writer.write_u64le?(a: u64)",

	// TODO: these should have an explicit pre-condition "available() >= N".
	// For now, that's implicitly checked (i.e. hard coded).
	//
	// The io_writer has write_fast_etc methods, not poke_etc and skip_fast,
	// because skip_fast could leave uninitialized bytes in the io_buffer.
	"io_writer.write_fast_u8!(a: u8)",
	"io_writer.write_fast_u16be!(a: u16)",
	"io_writer.write_fast_u16le!(a: u16)",
	"io_writer.write_fast_u24be!(a: u32[..= 0xFFFFFF])",
	"io_writer.write_fast_u24le!(a: u32[..= 0xFFFFFF])",
	"io_writer.write_fast_u32be!(a: u32)",
	"io_writer.write_fast_u32le!(a: u32)",
	"io_writer.write_fast_u40be!(a: u64[..= 0xFFFFFFFFFF])",
	"io_writer.write_fast_u40le!(a: u64[..= 0xFFFFFFFFFF])",
	"io_writer.write_fast_u48be!(a: u64[..= 0xFFFFFFFFFFFF])",
	"io_writer.write_fast_u48le!(a: u64[..= 0xFFFFFFFFFFFF])",
	"io_writer.write_fast_u56be!(a: u64[..= 0xFFFFFFFFFFFFFF])",
	"io_writer.write_fast_u56le!(a: u64[..= 0xFFFFFFFFFFFFFF])",
	"io_writer.write_fast_u64be!(a: u64)",
	"io_writer.write_fast_u64le!(a: u64)",

	"io_writer.available() u64",
	"io_writer.count_since(mark: u64) u64",
	"io_writer.history_available() u64",
	"io_writer.mark() u64",
	"io_writer.position() u64",
	"io_writer.since(mark: u64) slice u8",

	"io_writer.copy_from_slice!(s: slice u8) u64",
	"io_writer.copy_n_from_history!(n: u32, distance: u32) u32",
	"io_writer.copy_n_from_reader!(n: u32, r: io_reader) u32",
	"io_writer.copy_n_from_slice!(n: u32, s: slice u8) u32",

	// TODO: this should have explicit pre-conditions:
	// - n <= this.available()
	// - distance > 0
	// - distance <= this.since_mark().length()
	// For now, that's all implicitly checked (i.e. hard coded).
	"io_writer.copy_n_from_history_fast!(n: u32, distance: u32) u32",

	// ---- status

	// TODO: should we add is_complete?
	"status.is_error() bool",
	"status.is_note() bool",
	"status.is_ok() bool",
	"status.is_suspension() bool",

	// ---- frame_config
	// Duration's upper bound is the maximum possible i64 value.

	"frame_config.blend() u8",
	"frame_config.disposal() u8",
	"frame_config.duration() u64[..= 0x7FFFFFFFFFFFFFFF]",
	"frame_config.index() u64",
	"frame_config.io_position() u64",

	"frame_config.update!(bounds: rect_ie_u32, duration: u64[..= 0x7FFFFFFFFFFFFFFF], " +
	"index: u64, io_position: u64, blend: u8, disposal: u8, background_color: u32)",

	// ---- image_config

	"image_config.set!(pixfmt: u32, pixsub: u32, width: u32, height: u32, " +
	"first_frame_io_position: u64, first_frame_is_opaque: bool)",

	// ---- pixel_buffer

	"pixel_buffer.palette() slice u8",
	"pixel_buffer.pixel_format() pixel_format",
	"pixel_buffer.plane(p: u32[..= 3]) table u8",

	// ---- pixel_format

	"pixel_format.bits_per_pixel() u32",

	// ---- pixel_swizzler

	"pixel_swizzler.prepare!(" +
	"dst_pixfmt: pixel_format, dst_palette: slice u8, " +
	"src_pixfmt: pixel_format, src_palette: slice u8) status",
	"pixel_swizzler.swizzle_interleaved!(" +
	"dst: slice u8, dst_palette: slice u8, src: slice u8) u64",
	}

	var Interfaces = []string{
	"hasher_u32",
	}

	var InterfaceFuncs = []string{
	// ---- hasher_u32

	"hasher_u32.update_u32!(x: slice u8) u32",
	}

	// The "T1" and "T2" types here are placeholders for generic "slice T" or
	// "table T" types. After tokenizing (but before parsing) these XxxFunc strings
	// (e.g. in the lang/check package), replace "T1" and "T2" with "†" or "‡"
	// daggers, to avoid collision with a user-defined "T1" or "T2" type.

	const (
	GenericOldName1 = t.IDT1
	GenericOldName2 = t.IDT2
	GenericNewName1 = t.IDDagger1
	GenericNewName2 = t.IDDagger2
	)

	var SliceFuncs = []string{
	"T1.copy_from_slice!(s: T1) u64",
	"T1.length() u64",
	"T1.prefix(up_to: u64) T1",
	"T1.suffix(up_to: u64) T1",
	}

	var TableFuncs = []string{
	"T2.height() u64",
	"T2.stride() u64",
	"T2.width() u64",

	"T2.row(y: u32) T1",
	}

	func ParseFuncs(tm t.Map, ss []string, generic bool, callback func(a.Func) error) error {
	buf := []byte(nil)
	for _, s := range ss {
	buf = buf[:0]
	buf = append(buf, "pub func "...)
	buf = append(buf, s...)
	buf = append(buf, "{}\n"...)

	const filename = "builtin.wuffs"
	tokens, _, err := t.Tokenize(tm, filename, buf)
	if err != nil {
	return fmt.Errorf("parsing %q: could not tokenize built-in funcs: %v", s, err)
	}
	if generic {
	for i := range tokens {
	if tokens[i].ID == GenericOldName1 {
	tokens[i].ID = GenericNewName1
	} else if tokens[i].ID == GenericOldName2 {
	tokens[i].ID = GenericNewName2
	}
	}
	}
	file, err := parse.Parse(tm, filename, tokens, &parse.Options{
	AllowBuiltInNames: true,
	})
	if err != nil {
	return fmt.Errorf("parsing %q: could not parse built-in funcs: %v", s, err)
	}

	tlds := file.TopLevelDecls()
	if len(tlds) != 1 \|\| tlds[0].Kind() != a.KFunc {
	return fmt.Errorf("parsing %q: got %d top level decls, want %d", s, len(tlds), 1)
	}
	f := tlds[0].AsFunc()
	f.AsNode().AsRaw().SetPackage(tm, t.IDBase)

	if callback != nil {
	if err := callback(f); err != nil {
	return err
	}
	}
	}
	return nil
	}