diff options
Diffstat (limited to 'vendor/golang.org/x/text/encoding/unicode')
5 files changed, 1559 insertions, 0 deletions
diff --git a/vendor/golang.org/x/text/encoding/unicode/override.go b/vendor/golang.org/x/text/encoding/unicode/override.go new file mode 100644 index 000000000..35d62fcc9 --- /dev/null +++ b/vendor/golang.org/x/text/encoding/unicode/override.go @@ -0,0 +1,82 @@ +// Copyright 2015 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package unicode + +import ( + "golang.org/x/text/transform" +) + +// BOMOverride returns a new decoder transformer that is identical to fallback, +// except that the presence of a Byte Order Mark at the start of the input +// causes it to switch to the corresponding Unicode decoding. It will only +// consider BOMs for UTF-8, UTF-16BE, and UTF-16LE. +// +// This differs from using ExpectBOM by allowing a BOM to switch to UTF-8, not +// just UTF-16 variants, and allowing falling back to any encoding scheme. +// +// This technique is recommended by the W3C for use in HTML 5: "For +// compatibility with deployed content, the byte order mark (also known as BOM) +// is considered more authoritative than anything else." +// http://www.w3.org/TR/encoding/#specification-hooks +// +// Using BOMOverride is mostly intended for use cases where the first characters +// of a fallback encoding are known to not be a BOM, for example, for valid HTML +// and most encodings. +func BOMOverride(fallback transform.Transformer) transform.Transformer { + // TODO: possibly allow a variadic argument of unicode encodings to allow + // specifying details of which fallbacks are supported as well as + // specifying the details of the implementations. This would also allow for + // support for UTF-32, which should not be supported by default. + return &bomOverride{fallback: fallback} +} + +type bomOverride struct { + fallback transform.Transformer + current transform.Transformer +} + +func (d *bomOverride) Reset() { + d.current = nil + d.fallback.Reset() +} + +var ( + // TODO: we could use decode functions here, instead of allocating a new + // decoder on every NewDecoder as IgnoreBOM decoders can be stateless. + utf16le = UTF16(LittleEndian, IgnoreBOM) + utf16be = UTF16(BigEndian, IgnoreBOM) +) + +const utf8BOM = "\ufeff" + +func (d *bomOverride) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) { + if d.current != nil { + return d.current.Transform(dst, src, atEOF) + } + if len(src) < 3 && !atEOF { + return 0, 0, transform.ErrShortSrc + } + d.current = d.fallback + bomSize := 0 + if len(src) >= 2 { + if src[0] == 0xFF && src[1] == 0xFE { + d.current = utf16le.NewDecoder() + bomSize = 2 + } else if src[0] == 0xFE && src[1] == 0xFF { + d.current = utf16be.NewDecoder() + bomSize = 2 + } else if len(src) >= 3 && + src[0] == utf8BOM[0] && + src[1] == utf8BOM[1] && + src[2] == utf8BOM[2] { + d.current = transform.Nop + bomSize = 3 + } + } + if bomSize < len(src) { + nDst, nSrc, err = d.current.Transform(dst, src[bomSize:], atEOF) + } + return nDst, nSrc + bomSize, err +} diff --git a/vendor/golang.org/x/text/encoding/unicode/unicode.go b/vendor/golang.org/x/text/encoding/unicode/unicode.go new file mode 100644 index 000000000..579cadfb1 --- /dev/null +++ b/vendor/golang.org/x/text/encoding/unicode/unicode.go @@ -0,0 +1,434 @@ +// Copyright 2013 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package unicode provides Unicode encodings such as UTF-16. +package unicode // import "golang.org/x/text/encoding/unicode" + +import ( + "errors" + "unicode/utf16" + "unicode/utf8" + + "golang.org/x/text/encoding" + "golang.org/x/text/encoding/internal" + "golang.org/x/text/encoding/internal/identifier" + "golang.org/x/text/internal/utf8internal" + "golang.org/x/text/runes" + "golang.org/x/text/transform" +) + +// TODO: I think the Transformers really should return errors on unmatched +// surrogate pairs and odd numbers of bytes. This is not required by RFC 2781, +// which leaves it open, but is suggested by WhatWG. It will allow for all error +// modes as defined by WhatWG: fatal, HTML and Replacement. This would require +// the introduction of some kind of error type for conveying the erroneous code +// point. + +// UTF8 is the UTF-8 encoding. +var UTF8 encoding.Encoding = utf8enc + +var utf8enc = &internal.Encoding{ + &internal.SimpleEncoding{utf8Decoder{}, runes.ReplaceIllFormed()}, + "UTF-8", + identifier.UTF8, +} + +type utf8Decoder struct{ transform.NopResetter } + +func (utf8Decoder) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) { + var pSrc int // point from which to start copy in src + var accept utf8internal.AcceptRange + + // The decoder can only make the input larger, not smaller. + n := len(src) + if len(dst) < n { + err = transform.ErrShortDst + n = len(dst) + atEOF = false + } + for nSrc < n { + c := src[nSrc] + if c < utf8.RuneSelf { + nSrc++ + continue + } + first := utf8internal.First[c] + size := int(first & utf8internal.SizeMask) + if first == utf8internal.FirstInvalid { + goto handleInvalid // invalid starter byte + } + accept = utf8internal.AcceptRanges[first>>utf8internal.AcceptShift] + if nSrc+size > n { + if !atEOF { + // We may stop earlier than necessary here if the short sequence + // has invalid bytes. Not checking for this simplifies the code + // and may avoid duplicate computations in certain conditions. + if err == nil { + err = transform.ErrShortSrc + } + break + } + // Determine the maximal subpart of an ill-formed subsequence. + switch { + case nSrc+1 >= n || src[nSrc+1] < accept.Lo || accept.Hi < src[nSrc+1]: + size = 1 + case nSrc+2 >= n || src[nSrc+2] < utf8internal.LoCB || utf8internal.HiCB < src[nSrc+2]: + size = 2 + default: + size = 3 // As we are short, the maximum is 3. + } + goto handleInvalid + } + if c = src[nSrc+1]; c < accept.Lo || accept.Hi < c { + size = 1 + goto handleInvalid // invalid continuation byte + } else if size == 2 { + } else if c = src[nSrc+2]; c < utf8internal.LoCB || utf8internal.HiCB < c { + size = 2 + goto handleInvalid // invalid continuation byte + } else if size == 3 { + } else if c = src[nSrc+3]; c < utf8internal.LoCB || utf8internal.HiCB < c { + size = 3 + goto handleInvalid // invalid continuation byte + } + nSrc += size + continue + + handleInvalid: + // Copy the scanned input so far. + nDst += copy(dst[nDst:], src[pSrc:nSrc]) + + // Append RuneError to the destination. + const runeError = "\ufffd" + if nDst+len(runeError) > len(dst) { + return nDst, nSrc, transform.ErrShortDst + } + nDst += copy(dst[nDst:], runeError) + + // Skip the maximal subpart of an ill-formed subsequence according to + // the W3C standard way instead of the Go way. This Transform is + // probably the only place in the text repo where it is warranted. + nSrc += size + pSrc = nSrc + + // Recompute the maximum source length. + if sz := len(dst) - nDst; sz < len(src)-nSrc { + err = transform.ErrShortDst + n = nSrc + sz + atEOF = false + } + } + return nDst + copy(dst[nDst:], src[pSrc:nSrc]), nSrc, err +} + +// UTF16 returns a UTF-16 Encoding for the given default endianness and byte +// order mark (BOM) policy. +// +// When decoding from UTF-16 to UTF-8, if the BOMPolicy is IgnoreBOM then +// neither BOMs U+FEFF nor noncharacters U+FFFE in the input stream will affect +// the endianness used for decoding, and will instead be output as their +// standard UTF-8 encodings: "\xef\xbb\xbf" and "\xef\xbf\xbe". If the BOMPolicy +// is UseBOM or ExpectBOM a staring BOM is not written to the UTF-8 output. +// Instead, it overrides the default endianness e for the remainder of the +// transformation. Any subsequent BOMs U+FEFF or noncharacters U+FFFE will not +// affect the endianness used, and will instead be output as their standard +// UTF-8 encodings. For UseBOM, if there is no starting BOM, it will proceed +// with the default Endianness. For ExpectBOM, in that case, the transformation +// will return early with an ErrMissingBOM error. +// +// When encoding from UTF-8 to UTF-16, a BOM will be inserted at the start of +// the output if the BOMPolicy is UseBOM or ExpectBOM. Otherwise, a BOM will not +// be inserted. The UTF-8 input does not need to contain a BOM. +// +// There is no concept of a 'native' endianness. If the UTF-16 data is produced +// and consumed in a greater context that implies a certain endianness, use +// IgnoreBOM. Otherwise, use ExpectBOM and always produce and consume a BOM. +// +// In the language of http://www.unicode.org/faq/utf_bom.html#bom10, IgnoreBOM +// corresponds to "Where the precise type of the data stream is known... the +// BOM should not be used" and ExpectBOM corresponds to "A particular +// protocol... may require use of the BOM". +func UTF16(e Endianness, b BOMPolicy) encoding.Encoding { + return utf16Encoding{config{e, b}, mibValue[e][b&bomMask]} +} + +// mibValue maps Endianness and BOMPolicy settings to MIB constants. Note that +// some configurations map to the same MIB identifier. RFC 2781 has requirements +// and recommendations. Some of the "configurations" are merely recommendations, +// so multiple configurations could match. +var mibValue = map[Endianness][numBOMValues]identifier.MIB{ + BigEndian: [numBOMValues]identifier.MIB{ + IgnoreBOM: identifier.UTF16BE, + UseBOM: identifier.UTF16, // BigEnding default is preferred by RFC 2781. + // TODO: acceptBOM | strictBOM would map to UTF16BE as well. + }, + LittleEndian: [numBOMValues]identifier.MIB{ + IgnoreBOM: identifier.UTF16LE, + UseBOM: identifier.UTF16, // LittleEndian default is allowed and preferred on Windows. + // TODO: acceptBOM | strictBOM would map to UTF16LE as well. + }, + // ExpectBOM is not widely used and has no valid MIB identifier. +} + +// All lists a configuration for each IANA-defined UTF-16 variant. +var All = []encoding.Encoding{ + UTF8, + UTF16(BigEndian, UseBOM), + UTF16(BigEndian, IgnoreBOM), + UTF16(LittleEndian, IgnoreBOM), +} + +// BOMPolicy is a UTF-16 encoding's byte order mark policy. +type BOMPolicy uint8 + +const ( + writeBOM BOMPolicy = 0x01 + acceptBOM BOMPolicy = 0x02 + requireBOM BOMPolicy = 0x04 + bomMask BOMPolicy = 0x07 + + // HACK: numBOMValues == 8 triggers a bug in the 1.4 compiler (cannot have a + // map of an array of length 8 of a type that is also used as a key or value + // in another map). See golang.org/issue/11354. + // TODO: consider changing this value back to 8 if the use of 1.4.* has + // been minimized. + numBOMValues = 8 + 1 + + // IgnoreBOM means to ignore any byte order marks. + IgnoreBOM BOMPolicy = 0 + // Common and RFC 2781-compliant interpretation for UTF-16BE/LE. + + // UseBOM means that the UTF-16 form may start with a byte order mark, which + // will be used to override the default encoding. + UseBOM BOMPolicy = writeBOM | acceptBOM + // Common and RFC 2781-compliant interpretation for UTF-16. + + // ExpectBOM means that the UTF-16 form must start with a byte order mark, + // which will be used to override the default encoding. + ExpectBOM BOMPolicy = writeBOM | acceptBOM | requireBOM + // Used in Java as Unicode (not to be confused with Java's UTF-16) and + // ICU's UTF-16,version=1. Not compliant with RFC 2781. + + // TODO (maybe): strictBOM: BOM must match Endianness. This would allow: + // - UTF-16(B|L)E,version=1: writeBOM | acceptBOM | requireBOM | strictBOM + // (UnicodeBig and UnicodeLittle in Java) + // - RFC 2781-compliant, but less common interpretation for UTF-16(B|L)E: + // acceptBOM | strictBOM (e.g. assigned to CheckBOM). + // This addition would be consistent with supporting ExpectBOM. +) + +// Endianness is a UTF-16 encoding's default endianness. +type Endianness bool + +const ( + // BigEndian is UTF-16BE. + BigEndian Endianness = false + // LittleEndian is UTF-16LE. + LittleEndian Endianness = true +) + +// ErrMissingBOM means that decoding UTF-16 input with ExpectBOM did not find a +// starting byte order mark. +var ErrMissingBOM = errors.New("encoding: missing byte order mark") + +type utf16Encoding struct { + config + mib identifier.MIB +} + +type config struct { + endianness Endianness + bomPolicy BOMPolicy +} + +func (u utf16Encoding) NewDecoder() *encoding.Decoder { + return &encoding.Decoder{Transformer: &utf16Decoder{ + initial: u.config, + current: u.config, + }} +} + +func (u utf16Encoding) NewEncoder() *encoding.Encoder { + return &encoding.Encoder{Transformer: &utf16Encoder{ + endianness: u.endianness, + initialBOMPolicy: u.bomPolicy, + currentBOMPolicy: u.bomPolicy, + }} +} + +func (u utf16Encoding) ID() (mib identifier.MIB, other string) { + return u.mib, "" +} + +func (u utf16Encoding) String() string { + e, b := "B", "" + if u.endianness == LittleEndian { + e = "L" + } + switch u.bomPolicy { + case ExpectBOM: + b = "Expect" + case UseBOM: + b = "Use" + case IgnoreBOM: + b = "Ignore" + } + return "UTF-16" + e + "E (" + b + " BOM)" +} + +type utf16Decoder struct { + initial config + current config +} + +func (u *utf16Decoder) Reset() { + u.current = u.initial +} + +func (u *utf16Decoder) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) { + if len(src) == 0 { + if atEOF && u.current.bomPolicy&requireBOM != 0 { + return 0, 0, ErrMissingBOM + } + return 0, 0, nil + } + if u.current.bomPolicy&acceptBOM != 0 { + if len(src) < 2 { + return 0, 0, transform.ErrShortSrc + } + switch { + case src[0] == 0xfe && src[1] == 0xff: + u.current.endianness = BigEndian + nSrc = 2 + case src[0] == 0xff && src[1] == 0xfe: + u.current.endianness = LittleEndian + nSrc = 2 + default: + if u.current.bomPolicy&requireBOM != 0 { + return 0, 0, ErrMissingBOM + } + } + u.current.bomPolicy = IgnoreBOM + } + + var r rune + var dSize, sSize int + for nSrc < len(src) { + if nSrc+1 < len(src) { + x := uint16(src[nSrc+0])<<8 | uint16(src[nSrc+1]) + if u.current.endianness == LittleEndian { + x = x>>8 | x<<8 + } + r, sSize = rune(x), 2 + if utf16.IsSurrogate(r) { + if nSrc+3 < len(src) { + x = uint16(src[nSrc+2])<<8 | uint16(src[nSrc+3]) + if u.current.endianness == LittleEndian { + x = x>>8 | x<<8 + } + // Save for next iteration if it is not a high surrogate. + if isHighSurrogate(rune(x)) { + r, sSize = utf16.DecodeRune(r, rune(x)), 4 + } + } else if !atEOF { + err = transform.ErrShortSrc + break + } + } + if dSize = utf8.RuneLen(r); dSize < 0 { + r, dSize = utf8.RuneError, 3 + } + } else if atEOF { + // Single trailing byte. + r, dSize, sSize = utf8.RuneError, 3, 1 + } else { + err = transform.ErrShortSrc + break + } + if nDst+dSize > len(dst) { + err = transform.ErrShortDst + break + } + nDst += utf8.EncodeRune(dst[nDst:], r) + nSrc += sSize + } + return nDst, nSrc, err +} + +func isHighSurrogate(r rune) bool { + return 0xDC00 <= r && r <= 0xDFFF +} + +type utf16Encoder struct { + endianness Endianness + initialBOMPolicy BOMPolicy + currentBOMPolicy BOMPolicy +} + +func (u *utf16Encoder) Reset() { + u.currentBOMPolicy = u.initialBOMPolicy +} + +func (u *utf16Encoder) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) { + if u.currentBOMPolicy&writeBOM != 0 { + if len(dst) < 2 { + return 0, 0, transform.ErrShortDst + } + dst[0], dst[1] = 0xfe, 0xff + u.currentBOMPolicy = IgnoreBOM + nDst = 2 + } + + r, size := rune(0), 0 + for nSrc < len(src) { + r = rune(src[nSrc]) + + // Decode a 1-byte rune. + if r < utf8.RuneSelf { + size = 1 + + } else { + // Decode a multi-byte rune. + r, size = utf8.DecodeRune(src[nSrc:]) + if size == 1 { + // All valid runes of size 1 (those below utf8.RuneSelf) were + // handled above. We have invalid UTF-8 or we haven't seen the + // full character yet. + if !atEOF && !utf8.FullRune(src[nSrc:]) { + err = transform.ErrShortSrc + break + } + } + } + + if r <= 0xffff { + if nDst+2 > len(dst) { + err = transform.ErrShortDst + break + } + dst[nDst+0] = uint8(r >> 8) + dst[nDst+1] = uint8(r) + nDst += 2 + } else { + if nDst+4 > len(dst) { + err = transform.ErrShortDst + break + } + r1, r2 := utf16.EncodeRune(r) + dst[nDst+0] = uint8(r1 >> 8) + dst[nDst+1] = uint8(r1) + dst[nDst+2] = uint8(r2 >> 8) + dst[nDst+3] = uint8(r2) + nDst += 4 + } + nSrc += size + } + + if u.endianness == LittleEndian { + for i := 0; i < nDst; i += 2 { + dst[i], dst[i+1] = dst[i+1], dst[i] + } + } + return nDst, nSrc, err +} diff --git a/vendor/golang.org/x/text/encoding/unicode/unicode_test.go b/vendor/golang.org/x/text/encoding/unicode/unicode_test.go new file mode 100644 index 000000000..51b4712d8 --- /dev/null +++ b/vendor/golang.org/x/text/encoding/unicode/unicode_test.go @@ -0,0 +1,499 @@ +// Copyright 2015 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package unicode + +import ( + "testing" + + "golang.org/x/text/encoding" + "golang.org/x/text/encoding/charmap" + "golang.org/x/text/encoding/internal/enctest" + "golang.org/x/text/transform" +) + +func TestBasics(t *testing.T) { + testCases := []struct { + e encoding.Encoding + encPrefix string + encSuffix string + encoded string + utf8 string + }{{ + e: utf16BEIB, + encoded: "\x00\x57\x00\xe4\xd8\x35\xdd\x65", + utf8: "\x57\u00e4\U0001d565", + }, { + e: utf16BEEB, + encPrefix: "\xfe\xff", + encoded: "\x00\x57\x00\xe4\xd8\x35\xdd\x65", + utf8: "\x57\u00e4\U0001d565", + }, { + e: utf16LEIB, + encoded: "\x57\x00\xe4\x00\x35\xd8\x65\xdd", + utf8: "\x57\u00e4\U0001d565", + }, { + e: utf16LEEB, + encPrefix: "\xff\xfe", + encoded: "\x57\x00\xe4\x00\x35\xd8\x65\xdd", + utf8: "\x57\u00e4\U0001d565", + }} + + for _, tc := range testCases { + enctest.TestEncoding(t, tc.e, tc.encoded, tc.utf8, tc.encPrefix, tc.encSuffix) + } +} + +func TestFiles(t *testing.T) { + enctest.TestFile(t, UTF8) + enctest.TestFile(t, utf16LEIB) +} + +func BenchmarkEncoding(b *testing.B) { + enctest.Benchmark(b, UTF8) + enctest.Benchmark(b, utf16LEIB) +} + +var ( + utf16LEIB = UTF16(LittleEndian, IgnoreBOM) // UTF-16LE (atypical interpretation) + utf16LEUB = UTF16(LittleEndian, UseBOM) // UTF-16, LE + utf16LEEB = UTF16(LittleEndian, ExpectBOM) // UTF-16, LE, Expect + utf16BEIB = UTF16(BigEndian, IgnoreBOM) // UTF-16BE (atypical interpretation) + utf16BEUB = UTF16(BigEndian, UseBOM) // UTF-16 default + utf16BEEB = UTF16(BigEndian, ExpectBOM) // UTF-16 Expect +) + +func TestUTF16(t *testing.T) { + testCases := []struct { + desc string + src string + notEOF bool // the inverse of atEOF + sizeDst int + want string + nSrc int + err error + t transform.Transformer + }{{ + desc: "utf-16 IgnoreBOM dec: empty string", + t: utf16BEIB.NewDecoder(), + }, { + desc: "utf-16 UseBOM dec: empty string", + t: utf16BEUB.NewDecoder(), + }, { + desc: "utf-16 ExpectBOM dec: empty string", + err: ErrMissingBOM, + t: utf16BEEB.NewDecoder(), + }, { + desc: "utf-16 dec: BOM determines encoding BE (RFC 2781:3.3)", + src: "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 12, + t: utf16BEUB.NewDecoder(), + }, { + desc: "utf-16 dec: BOM determines encoding LE (RFC 2781:3.3)", + src: "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 12, + t: utf16LEUB.NewDecoder(), + }, { + desc: "utf-16 dec: BOM determines encoding LE, change default (RFC 2781:3.3)", + src: "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 12, + t: utf16BEUB.NewDecoder(), + }, { + desc: "utf-16 dec: Fail on missing BOM when required", + src: "\x08\xD8\x45\xDF\x3D\x00\xFF\xFE\xFE\xFF\x00\x52\x00\x61", + sizeDst: 100, + want: "", + nSrc: 0, + err: ErrMissingBOM, + t: utf16BEEB.NewDecoder(), + }, { + desc: "utf-16 dec: SHOULD interpret text as big-endian when BOM not present (RFC 2781:4.3)", + src: "\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 10, + t: utf16BEUB.NewDecoder(), + }, { + // This is an error according to RFC 2781. But errors in RFC 2781 are + // open to interpretations, so I guess this is fine. + desc: "utf-16le dec: incorrect BOM is an error (RFC 2781:4.1)", + src: "\xFE\xFF\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00", + sizeDst: 100, + want: "\uFFFE\U00012345=Ra", + nSrc: 12, + t: utf16LEIB.NewDecoder(), + }, { + desc: "utf-16 enc: SHOULD write BOM (RFC 2781:3.3)", + src: "\U00012345=Ra", + sizeDst: 100, + want: "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00", + nSrc: 7, + t: utf16LEUB.NewEncoder(), + }, { + desc: "utf-16 enc: SHOULD write BOM (RFC 2781:3.3)", + src: "\U00012345=Ra", + sizeDst: 100, + want: "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61", + nSrc: 7, + t: utf16BEUB.NewEncoder(), + }, { + desc: "utf-16le enc: MUST NOT write BOM (RFC 2781:3.3)", + src: "\U00012345=Ra", + sizeDst: 100, + want: "\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00", + nSrc: 7, + t: utf16LEIB.NewEncoder(), + }, { + desc: "utf-16be dec: incorrect UTF-16: odd bytes", + src: "\x00", + sizeDst: 100, + want: "\uFFFD", + nSrc: 1, + t: utf16BEIB.NewDecoder(), + }, { + desc: "utf-16be dec: unpaired surrogate, odd bytes", + src: "\xD8\x45\x00", + sizeDst: 100, + want: "\uFFFD\uFFFD", + nSrc: 3, + t: utf16BEIB.NewDecoder(), + }, { + desc: "utf-16be dec: unpaired low surrogate + valid text", + src: "\xD8\x45\x00a", + sizeDst: 100, + want: "\uFFFDa", + nSrc: 4, + t: utf16BEIB.NewDecoder(), + }, { + desc: "utf-16be dec: unpaired low surrogate + valid text + single byte", + src: "\xD8\x45\x00ab", + sizeDst: 100, + want: "\uFFFDa\uFFFD", + nSrc: 5, + t: utf16BEIB.NewDecoder(), + }, { + desc: "utf-16le dec: unpaired high surrogate", + src: "\x00\x00\x00\xDC\x12\xD8", + sizeDst: 100, + want: "\x00\uFFFD\uFFFD", + nSrc: 6, + t: utf16LEIB.NewDecoder(), + }, { + desc: "utf-16be dec: two unpaired low surrogates", + src: "\xD8\x45\xD8\x12", + sizeDst: 100, + want: "\uFFFD\uFFFD", + nSrc: 4, + t: utf16BEIB.NewDecoder(), + }, { + desc: "utf-16be dec: short dst", + src: "\x00a", + sizeDst: 0, + want: "", + nSrc: 0, + t: utf16BEIB.NewDecoder(), + err: transform.ErrShortDst, + }, { + desc: "utf-16be dec: short dst surrogate", + src: "\xD8\xF5\xDC\x12", + sizeDst: 3, + want: "", + nSrc: 0, + t: utf16BEIB.NewDecoder(), + err: transform.ErrShortDst, + }, { + desc: "utf-16be dec: short dst trailing byte", + src: "\x00", + sizeDst: 2, + want: "", + nSrc: 0, + t: utf16BEIB.NewDecoder(), + err: transform.ErrShortDst, + }, { + desc: "utf-16be dec: short src", + src: "\x00", + notEOF: true, + sizeDst: 3, + want: "", + nSrc: 0, + t: utf16BEIB.NewDecoder(), + err: transform.ErrShortSrc, + }, { + desc: "utf-16 enc", + src: "\U00012345=Ra", + sizeDst: 100, + want: "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61", + nSrc: 7, + t: utf16BEUB.NewEncoder(), + }, { + desc: "utf-16 enc: short dst normal", + src: "\U00012345=Ra", + sizeDst: 9, + want: "\xD8\x08\xDF\x45\x00\x3D\x00\x52", + nSrc: 6, + t: utf16BEIB.NewEncoder(), + err: transform.ErrShortDst, + }, { + desc: "utf-16 enc: short dst surrogate", + src: "\U00012345=Ra", + sizeDst: 3, + want: "", + nSrc: 0, + t: utf16BEIB.NewEncoder(), + err: transform.ErrShortDst, + }, { + desc: "utf-16 enc: short src", + src: "\U00012345=Ra\xC2", + notEOF: true, + sizeDst: 100, + want: "\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61", + nSrc: 7, + t: utf16BEIB.NewEncoder(), + err: transform.ErrShortSrc, + }, { + desc: "utf-16be dec: don't change byte order mid-stream", + src: "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\xFF\xFE\x00\x52\x00\x61", + sizeDst: 100, + want: "\U00012345=\ufffeRa", + nSrc: 14, + t: utf16BEUB.NewDecoder(), + }, { + desc: "utf-16le dec: don't change byte order mid-stream", + src: "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\xFF\xFE\xFE\xFF\x52\x00\x61\x00", + sizeDst: 100, + want: "\U00012345=\ufeff\ufffeRa", + nSrc: 16, + t: utf16LEUB.NewDecoder(), + }} + for i, tc := range testCases { + b := make([]byte, tc.sizeDst) + nDst, nSrc, err := tc.t.Transform(b, []byte(tc.src), !tc.notEOF) + if err != tc.err { + t.Errorf("%d:%s: error was %v; want %v", i, tc.desc, err, tc.err) + } + if got := string(b[:nDst]); got != tc.want { + t.Errorf("%d:%s: result was %q: want %q", i, tc.desc, got, tc.want) + } + if nSrc != tc.nSrc { + t.Errorf("%d:%s: nSrc was %d; want %d", i, tc.desc, nSrc, tc.nSrc) + } + } +} + +func TestUTF8Decoder(t *testing.T) { + testCases := []struct { + desc string + src string + notEOF bool // the inverse of atEOF + sizeDst int + want string + nSrc int + err error + }{{ + desc: "empty string, empty dest buffer", + }, { + desc: "empty string", + sizeDst: 8, + }, { + desc: "empty string, streaming", + notEOF: true, + sizeDst: 8, + }, { + desc: "ascii", + src: "abcde", + sizeDst: 8, + want: "abcde", + nSrc: 5, + }, { + desc: "ascii and error", + src: "ab\x80de", + sizeDst: 7, + want: "ab\ufffdde", + nSrc: 5, + }, { + desc: "valid two-byte sequence", + src: "a\u0300bc", + sizeDst: 7, + want: "a\u0300bc", + nSrc: 5, + }, { + desc: "valid three-byte sequence", + src: "a\u0300中", + sizeDst: 7, + want: "a\u0300中", + nSrc: 6, + }, { + desc: "valid four-byte sequence", + src: "a中\U00016F50", + sizeDst: 8, + want: "a中\U00016F50", + nSrc: 8, + }, { + desc: "short source buffer", + src: "abc\xf0\x90", + notEOF: true, + sizeDst: 10, + want: "abc", + nSrc: 3, + err: transform.ErrShortSrc, + }, { + // We don't check for the maximal subpart of an ill-formed subsequence + // at the end of an open segment. + desc: "complete invalid that looks like short at end", + src: "abc\xf0\x80", + notEOF: true, + sizeDst: 10, + want: "abc", // instead of "abc\ufffd\ufffd", + nSrc: 3, + err: transform.ErrShortSrc, + }, { + desc: "incomplete sequence at end", + src: "a\x80bc\xf0\x90", + sizeDst: 9, + want: "a\ufffdbc\ufffd", + nSrc: 6, + }, { + desc: "invalid second byte", + src: "abc\xf0dddd", + sizeDst: 10, + want: "abc\ufffddddd", + nSrc: 8, + }, { + desc: "invalid second byte at end", + src: "abc\xf0d", + sizeDst: 10, + want: "abc\ufffdd", + nSrc: 5, + }, { + desc: "invalid third byte", + src: "a\u0300bc\xf0\x90dddd", + sizeDst: 12, + want: "a\u0300bc\ufffddddd", + nSrc: 11, + }, { + desc: "invalid third byte at end", + src: "a\u0300bc\xf0\x90d", + sizeDst: 12, + want: "a\u0300bc\ufffdd", + nSrc: 8, + }, { + desc: "invalid fourth byte, tight buffer", + src: "a\u0300bc\xf0\x90\x80d", + sizeDst: 9, + want: "a\u0300bc\ufffdd", + nSrc: 9, + }, { + desc: "invalid fourth byte at end", + src: "a\u0300bc\xf0\x90\x80", + sizeDst: 8, + want: "a\u0300bc\ufffd", + nSrc: 8, + }, { + desc: "invalid fourth byte and short four byte sequence", + src: "a\u0300bc\xf0\x90\x80\xf0\x90\x80", + notEOF: true, + sizeDst: 20, + want: "a\u0300bc\ufffd", + nSrc: 8, + err: transform.ErrShortSrc, + }, { + desc: "valid four-byte sequence overflowing short buffer", + src: "a\u0300bc\xf0\x90\x80\x80", + notEOF: true, + sizeDst: 8, + want: "a\u0300bc", + nSrc: 5, + err: transform.ErrShortDst, + }, { + desc: "invalid fourth byte at end short, but short dst", + src: "a\u0300bc\xf0\x90\x80\xf0\x90\x80", + notEOF: true, + sizeDst: 8, + // More bytes would fit in the buffer, but this seems to require a more + // complicated and slower algorithm. + want: "a\u0300bc", // instead of "a\u0300bc" + nSrc: 5, + err: transform.ErrShortDst, + }, { + desc: "short dst for error", + src: "abc\x80", + notEOF: true, + sizeDst: 5, + want: "abc", + nSrc: 3, + err: transform.ErrShortDst, + }, { + desc: "adjusting short dst buffer", + src: "abc\x80ef", + notEOF: true, + sizeDst: 6, + want: "abc\ufffd", + nSrc: 4, + err: transform.ErrShortDst, + }} + tr := UTF8.NewDecoder() + for i, tc := range testCases { + b := make([]byte, tc.sizeDst) + nDst, nSrc, err := tr.Transform(b, []byte(tc.src), !tc.notEOF) + if err != tc.err { + t.Errorf("%d:%s: error was %v; want %v", i, tc.desc, err, tc.err) + } + if got := string(b[:nDst]); got != tc.want { + t.Errorf("%d:%s: result was %q: want %q", i, tc.desc, got, tc.want) + } + if nSrc != tc.nSrc { + t.Errorf("%d:%s: nSrc was %d; want %d", i, tc.desc, nSrc, tc.nSrc) + } + } +} + +func TestBOMOverride(t *testing.T) { + dec := BOMOverride(charmap.CodePage437.NewDecoder()) + dst := make([]byte, 100) + for i, tc := range []struct { + src string + atEOF bool + dst string + nSrc int + err error + }{ + 0: {"H\x82ll\x93", true, "Héllô", 5, nil}, + 1: {"\uFEFFHéllö", true, "Héllö", 10, nil}, + 2: {"\xFE\xFF\x00H\x00e\x00l\x00l\x00o", true, "Hello", 12, nil}, + 3: {"\xFF\xFEH\x00e\x00l\x00l\x00o\x00", true, "Hello", 12, nil}, + 4: {"\uFEFF", true, "", 3, nil}, + 5: {"\xFE\xFF", true, "", 2, nil}, + 6: {"\xFF\xFE", true, "", 2, nil}, + 7: {"\xEF\xBB", true, "\u2229\u2557", 2, nil}, + 8: {"\xEF", true, "\u2229", 1, nil}, + 9: {"", true, "", 0, nil}, + 10: {"\xFE", true, "\u25a0", 1, nil}, + 11: {"\xFF", true, "\u00a0", 1, nil}, + 12: {"\xEF\xBB", false, "", 0, transform.ErrShortSrc}, + 13: {"\xEF", false, "", 0, transform.ErrShortSrc}, + 14: {"", false, "", 0, transform.ErrShortSrc}, + 15: {"\xFE", false, "", 0, transform.ErrShortSrc}, + 16: {"\xFF", false, "", 0, transform.ErrShortSrc}, + 17: {"\xFF\xFE", false, "", 0, transform.ErrShortSrc}, + } { + dec.Reset() + nDst, nSrc, err := dec.Transform(dst, []byte(tc.src), tc.atEOF) + got := string(dst[:nDst]) + if nSrc != tc.nSrc { + t.Errorf("%d: nSrc: got %d; want %d", i, nSrc, tc.nSrc) + } + if got != tc.dst { + t.Errorf("%d: got %+q; want %+q", i, got, tc.dst) + } + if err != tc.err { + t.Errorf("%d: error: got %v; want %v", i, err, tc.err) + } + } +} diff --git a/vendor/golang.org/x/text/encoding/unicode/utf32/utf32.go b/vendor/golang.org/x/text/encoding/unicode/utf32/utf32.go new file mode 100644 index 000000000..48b215217 --- /dev/null +++ b/vendor/golang.org/x/text/encoding/unicode/utf32/utf32.go @@ -0,0 +1,296 @@ +// Copyright 2016 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package utf32 provides the UTF-32 Unicode encoding. +// +// Please note that support for UTF-32 is discouraged as it is a rare and +// inefficient encoding, unfit for use as an interchange format. For use +// on the web, the W3C strongly discourages its use +// (https://www.w3.org/TR/html5/document-metadata.html#charset) +// while WHATWG directly prohibits supporting it +// (https://html.spec.whatwg.org/multipage/syntax.html#character-encodings). +package utf32 // import "golang.org/x/text/encoding/unicode/utf32" + +import ( + "errors" + "unicode/utf8" + + "golang.org/x/text/encoding" + "golang.org/x/text/encoding/internal/identifier" + "golang.org/x/text/transform" +) + +// All lists a configuration for each IANA-defined UTF-32 variant. +var All = []encoding.Encoding{ + UTF32(BigEndian, UseBOM), + UTF32(BigEndian, IgnoreBOM), + UTF32(LittleEndian, IgnoreBOM), +} + +// ErrMissingBOM means that decoding UTF-32 input with ExpectBOM did not +// find a starting byte order mark. +var ErrMissingBOM = errors.New("encoding: missing byte order mark") + +// UTF32 returns a UTF-32 Encoding for the given default endianness and +// byte order mark (BOM) policy. +// +// When decoding from UTF-32 to UTF-8, if the BOMPolicy is IgnoreBOM then +// neither BOMs U+FEFF nor ill-formed code units 0xFFFE0000 in the input +// stream will affect the endianness used for decoding. Instead BOMs will +// be output as their standard UTF-8 encoding "\xef\xbb\xbf" while +// 0xFFFE0000 code units will be output as "\xef\xbf\xbd", the standard +// UTF-8 encoding for the Unicode replacement character. If the BOMPolicy +// is UseBOM or ExpectBOM a starting BOM is not written to the UTF-8 +// output. Instead, it overrides the default endianness e for the remainder +// of the transformation. Any subsequent BOMs U+FEFF or ill-formed code +// units 0xFFFE0000 will not affect the endianness used, and will instead +// be output as their standard UTF-8 (replacement) encodings. For UseBOM, +// if there is no starting BOM, it will proceed with the default +// Endianness. For ExpectBOM, in that case, the transformation will return +// early with an ErrMissingBOM error. +// +// When encoding from UTF-8 to UTF-32, a BOM will be inserted at the start +// of the output if the BOMPolicy is UseBOM or ExpectBOM. Otherwise, a BOM +// will not be inserted. The UTF-8 input does not need to contain a BOM. +// +// There is no concept of a 'native' endianness. If the UTF-32 data is +// produced and consumed in a greater context that implies a certain +// endianness, use IgnoreBOM. Otherwise, use ExpectBOM and always produce +// and consume a BOM. +// +// In the language of http://www.unicode.org/faq/utf_bom.html#bom10, +// IgnoreBOM corresponds to "Where the precise type of the data stream is +// known... the BOM should not be used" and ExpectBOM corresponds to "A +// particular protocol... may require use of the BOM". +func UTF32(e Endianness, b BOMPolicy) encoding.Encoding { + return utf32Encoding{config{e, b}, mibValue[e][b&bomMask]} +} + +// mibValue maps Endianness and BOMPolicy settings to MIB constants for UTF-32. +// Note that some configurations map to the same MIB identifier. +var mibValue = map[Endianness][numBOMValues]identifier.MIB{ + BigEndian: [numBOMValues]identifier.MIB{ + IgnoreBOM: identifier.UTF32BE, + UseBOM: identifier.UTF32, + }, + LittleEndian: [numBOMValues]identifier.MIB{ + IgnoreBOM: identifier.UTF32LE, + UseBOM: identifier.UTF32, + }, + // ExpectBOM is not widely used and has no valid MIB identifier. +} + +// BOMPolicy is a UTF-32 encodings's byte order mark policy. +type BOMPolicy uint8 + +const ( + writeBOM BOMPolicy = 0x01 + acceptBOM BOMPolicy = 0x02 + requireBOM BOMPolicy = 0x04 + bomMask BOMPolicy = 0x07 + + // HACK: numBOMValues == 8 triggers a bug in the 1.4 compiler (cannot have a + // map of an array of length 8 of a type that is also used as a key or value + // in another map). See golang.org/issue/11354. + // TODO: consider changing this value back to 8 if the use of 1.4.* has + // been minimized. + numBOMValues = 8 + 1 + + // IgnoreBOM means to ignore any byte order marks. + IgnoreBOM BOMPolicy = 0 + // Unicode-compliant interpretation for UTF-32BE/LE. + + // UseBOM means that the UTF-32 form may start with a byte order mark, + // which will be used to override the default encoding. + UseBOM BOMPolicy = writeBOM | acceptBOM + // Unicode-compliant interpretation for UTF-32. + + // ExpectBOM means that the UTF-32 form must start with a byte order mark, + // which will be used to override the default encoding. + ExpectBOM BOMPolicy = writeBOM | acceptBOM | requireBOM + // Consistent with BOMPolicy definition in golang.org/x/text/encoding/unicode +) + +// Endianness is a UTF-32 encoding's default endianness. +type Endianness bool + +const ( + // BigEndian is UTF-32BE. + BigEndian Endianness = false + // LittleEndian is UTF-32LE. + LittleEndian Endianness = true +) + +type config struct { + endianness Endianness + bomPolicy BOMPolicy +} + +type utf32Encoding struct { + config + mib identifier.MIB +} + +func (u utf32Encoding) NewDecoder() *encoding.Decoder { + return &encoding.Decoder{Transformer: &utf32Decoder{ + initial: u.config, + current: u.config, + }} +} + +func (u utf32Encoding) NewEncoder() *encoding.Encoder { + return &encoding.Encoder{Transformer: &utf32Encoder{ + endianness: u.endianness, + initialBOMPolicy: u.bomPolicy, + currentBOMPolicy: u.bomPolicy, + }} +} + +func (u utf32Encoding) ID() (mib identifier.MIB, other string) { + return u.mib, "" +} + +func (u utf32Encoding) String() string { + e, b := "B", "" + if u.endianness == LittleEndian { + e = "L" + } + switch u.bomPolicy { + case ExpectBOM: + b = "Expect" + case UseBOM: + b = "Use" + case IgnoreBOM: + b = "Ignore" + } + return "UTF-32" + e + "E (" + b + " BOM)" +} + +type utf32Decoder struct { + initial config + current config +} + +func (u *utf32Decoder) Reset() { + u.current = u.initial +} + +func (u *utf32Decoder) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) { + if len(src) == 0 { + if atEOF && u.current.bomPolicy&requireBOM != 0 { + return 0, 0, ErrMissingBOM + } + return 0, 0, nil + } + if u.current.bomPolicy&acceptBOM != 0 { + if len(src) < 4 { + return 0, 0, transform.ErrShortSrc + } + switch { + case src[0] == 0x00 && src[1] == 0x00 && src[2] == 0xfe && src[3] == 0xff: + u.current.endianness = BigEndian + nSrc = 4 + case src[0] == 0xff && src[1] == 0xfe && src[2] == 0x00 && src[3] == 0x00: + u.current.endianness = LittleEndian + nSrc = 4 + default: + if u.current.bomPolicy&requireBOM != 0 { + return 0, 0, ErrMissingBOM + } + } + u.current.bomPolicy = IgnoreBOM + } + + var r rune + var dSize, sSize int + for nSrc < len(src) { + if nSrc+3 < len(src) { + x := uint32(src[nSrc+0])<<24 | uint32(src[nSrc+1])<<16 | + uint32(src[nSrc+2])<<8 | uint32(src[nSrc+3]) + if u.current.endianness == LittleEndian { + x = x>>24 | (x >> 8 & 0x0000FF00) | (x << 8 & 0x00FF0000) | x<<24 + } + r, sSize = rune(x), 4 + if dSize = utf8.RuneLen(r); dSize < 0 { + r, dSize = utf8.RuneError, 3 + } + } else if atEOF { + // 1..3 trailing bytes. + r, dSize, sSize = utf8.RuneError, 3, len(src)-nSrc + } else { + err = transform.ErrShortSrc + break + } + if nDst+dSize > len(dst) { + err = transform.ErrShortDst + break + } + nDst += utf8.EncodeRune(dst[nDst:], r) + nSrc += sSize + } + return nDst, nSrc, err +} + +type utf32Encoder struct { + endianness Endianness + initialBOMPolicy BOMPolicy + currentBOMPolicy BOMPolicy +} + +func (u *utf32Encoder) Reset() { + u.currentBOMPolicy = u.initialBOMPolicy +} + +func (u *utf32Encoder) Transform(dst, src []byte, atEOF bool) (nDst, nSrc int, err error) { + if u.currentBOMPolicy&writeBOM != 0 { + if len(dst) < 4 { + return 0, 0, transform.ErrShortDst + } + dst[0], dst[1], dst[2], dst[3] = 0x00, 0x00, 0xfe, 0xff + u.currentBOMPolicy = IgnoreBOM + nDst = 4 + } + + r, size := rune(0), 0 + for nSrc < len(src) { + r = rune(src[nSrc]) + + // Decode a 1-byte rune. + if r < utf8.RuneSelf { + size = 1 + + } else { + // Decode a multi-byte rune. + r, size = utf8.DecodeRune(src[nSrc:]) + if size == 1 { + // All valid runes of size 1 (those below utf8.RuneSelf) were + // handled above. We have invalid UTF-8 or we haven't seen the + // full character yet. + if !atEOF && !utf8.FullRune(src[nSrc:]) { + err = transform.ErrShortSrc + break + } + } + } + + if nDst+4 > len(dst) { + err = transform.ErrShortDst + break + } + + dst[nDst+0] = uint8(r >> 24) + dst[nDst+1] = uint8(r >> 16) + dst[nDst+2] = uint8(r >> 8) + dst[nDst+3] = uint8(r) + nDst += 4 + nSrc += size + } + + if u.endianness == LittleEndian { + for i := 0; i < nDst; i += 4 { + dst[i], dst[i+1], dst[i+2], dst[i+3] = dst[i+3], dst[i+2], dst[i+1], dst[i] + } + } + return nDst, nSrc, err +} diff --git a/vendor/golang.org/x/text/encoding/unicode/utf32/utf32_test.go b/vendor/golang.org/x/text/encoding/unicode/utf32/utf32_test.go new file mode 100644 index 000000000..cd6158e5e --- /dev/null +++ b/vendor/golang.org/x/text/encoding/unicode/utf32/utf32_test.go @@ -0,0 +1,248 @@ +// Copyright 2016 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package utf32 + +import ( + "testing" + + "golang.org/x/text/encoding" + "golang.org/x/text/encoding/internal/enctest" + "golang.org/x/text/transform" +) + +var ( + utf32LEIB = UTF32(LittleEndian, IgnoreBOM) // UTF-32LE (atypical interpretation) + utf32LEUB = UTF32(LittleEndian, UseBOM) // UTF-32, LE + // utf32LEEB = UTF32(LittleEndian, ExpectBOM) // UTF-32, LE, Expect - covered in encoding_test.go + utf32BEIB = UTF32(BigEndian, IgnoreBOM) // UTF-32BE (atypical interpretation) + utf32BEUB = UTF32(BigEndian, UseBOM) // UTF-32 default + utf32BEEB = UTF32(BigEndian, ExpectBOM) // UTF-32 Expect +) + +func TestBasics(t *testing.T) { + testCases := []struct { + e encoding.Encoding + encPrefix string + encSuffix string + encoded string + utf8 string + }{{ + e: utf32BEIB, + encoded: "\x00\x00\x00\x57\x00\x00\x00\xe4\x00\x01\xd5\x65", + utf8: "\x57\u00e4\U0001d565", + }, { + e: UTF32(BigEndian, ExpectBOM), + encPrefix: "\x00\x00\xfe\xff", + encoded: "\x00\x00\x00\x57\x00\x00\x00\xe4\x00\x01\xd5\x65", + utf8: "\x57\u00e4\U0001d565", + }, { + e: UTF32(LittleEndian, IgnoreBOM), + encoded: "\x57\x00\x00\x00\xe4\x00\x00\x00\x65\xd5\x01\x00", + utf8: "\x57\u00e4\U0001d565", + }, { + e: UTF32(LittleEndian, ExpectBOM), + encPrefix: "\xff\xfe\x00\x00", + encoded: "\x57\x00\x00\x00\xe4\x00\x00\x00\x65\xd5\x01\x00", + utf8: "\x57\u00e4\U0001d565", + }} + + for _, tc := range testCases { + enctest.TestEncoding(t, tc.e, tc.encoded, tc.utf8, tc.encPrefix, tc.encSuffix) + } +} + +func TestFiles(t *testing.T) { enctest.TestFile(t, utf32BEIB) } + +func BenchmarkEncoding(b *testing.B) { enctest.Benchmark(b, utf32BEIB) } + +func TestUTF32(t *testing.T) { + testCases := []struct { + desc string + src string + notEOF bool // the inverse of atEOF + sizeDst int + want string + nSrc int + err error + t transform.Transformer + }{{ + desc: "utf-32 IgnoreBOM dec: empty string", + t: utf32BEIB.NewDecoder(), + }, { + desc: "utf-32 UseBOM dec: empty string", + t: utf32BEUB.NewDecoder(), + }, { + desc: "utf-32 ExpectBOM dec: empty string", + err: ErrMissingBOM, + t: utf32BEEB.NewDecoder(), + }, { + desc: "utf-32be dec: Doesn't interpret U+FEFF as BOM", + src: "\x00\x00\xFE\xFF\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + sizeDst: 100, + want: "\uFEFF\U00012345=Ra", + nSrc: 20, + t: utf32BEIB.NewDecoder(), + }, { + desc: "utf-32be dec: Interprets little endian U+FEFF as invalid", + src: "\xFF\xFE\x00\x00\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + sizeDst: 100, + want: "\uFFFD\U00012345=Ra", + nSrc: 20, + t: utf32BEIB.NewDecoder(), + }, { + desc: "utf-32le dec: Doesn't interpret U+FEFF as BOM", + src: "\xFF\xFE\x00\x00\x45\x23\x01\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61\x00\x00\x00", + sizeDst: 100, + want: "\uFEFF\U00012345=Ra", + nSrc: 20, + t: utf32LEIB.NewDecoder(), + }, { + desc: "utf-32le dec: Interprets big endian U+FEFF as invalid", + src: "\x00\x00\xFE\xFF\x45\x23\x01\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61\x00\x00\x00", + sizeDst: 100, + want: "\uFFFD\U00012345=Ra", + nSrc: 20, + t: utf32LEIB.NewDecoder(), + }, { + desc: "utf-32 enc: Writes big-endian BOM", + src: "\U00012345=Ra", + sizeDst: 100, + want: "\x00\x00\xFE\xFF\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + nSrc: 7, + t: utf32BEUB.NewEncoder(), + }, { + desc: "utf-32 enc: Writes little-endian BOM", + src: "\U00012345=Ra", + sizeDst: 100, + want: "\xFF\xFE\x00\x00\x45\x23\x01\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61\x00\x00\x00", + nSrc: 7, + t: utf32LEUB.NewEncoder(), + }, { + desc: "utf-32 dec: Interprets text using big-endian default when BOM not present", + src: "\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 16, + t: utf32BEUB.NewDecoder(), + }, { + desc: "utf-32 dec: Interprets text using little-endian default when BOM not present", + src: "\x45\x23\x01\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61\x00\x00\x00", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 16, + t: utf32LEUB.NewDecoder(), + }, { + desc: "utf-32 dec: BOM determines encoding BE", + src: "\x00\x00\xFE\xFF\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 20, + t: utf32BEUB.NewDecoder(), + }, { + desc: "utf-32 dec: BOM determines encoding LE", + src: "\xFF\xFE\x00\x00\x45\x23\x01\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61\x00\x00\x00", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 20, + t: utf32LEUB.NewDecoder(), + }, { + desc: "utf-32 dec: BOM determines encoding LE, change default", + src: "\xFF\xFE\x00\x00\x45\x23\x01\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61\x00\x00\x00", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 20, + t: utf32BEUB.NewDecoder(), + }, { + desc: "utf-32 dec: BOM determines encoding BE, change default", + src: "\x00\x00\xFE\xFF\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + sizeDst: 100, + want: "\U00012345=Ra", + nSrc: 20, + t: utf32LEUB.NewDecoder(), + }, { + desc: "utf-32 dec: Don't change big-endian byte order mid-stream", + src: "\x00\x01\x23\x45\x00\x00\x00\x3D\xFF\xFE\x00\x00\x00\x00\xFE\xFF\x00\x00\x00\x52\x00\x00\x00\x61", + sizeDst: 100, + want: "\U00012345=\uFFFD\uFEFFRa", + nSrc: 24, + t: utf32BEUB.NewDecoder(), + }, { + desc: "utf-32 dec: Don't change little-endian byte order mid-stream", + src: "\x45\x23\x01\x00\x3D\x00\x00\x00\x00\x00\xFE\xFF\xFF\xFE\x00\x00\x52\x00\x00\x00\x61\x00\x00\x00", + sizeDst: 100, + want: "\U00012345=\uFFFD\uFEFFRa", + nSrc: 24, + t: utf32LEUB.NewDecoder(), + }, { + desc: "utf-32 dec: Fail on missing BOM when required", + src: "\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + sizeDst: 100, + want: "", + nSrc: 0, + err: ErrMissingBOM, + t: utf32BEEB.NewDecoder(), + }, { + desc: "utf-32 enc: Short dst", + src: "\U00012345=Ra", + sizeDst: 15, + want: "\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52", + nSrc: 6, + err: transform.ErrShortDst, + t: utf32BEIB.NewEncoder(), + }, { + desc: "utf-32 enc: Short src", + src: "\U00012345=Ra\xC2", + notEOF: true, + sizeDst: 100, + want: "\x00\x01\x23\x45\x00\x00\x00\x3D\x00\x00\x00\x52\x00\x00\x00\x61", + nSrc: 7, + err: transform.ErrShortSrc, + t: utf32BEIB.NewEncoder(), + }, { + desc: "utf-32 enc: Invalid input", + src: "\x80\xC1\xC2\x7F\xC2", + sizeDst: 100, + want: "\x00\x00\xFF\xFD\x00\x00\xFF\xFD\x00\x00\xFF\xFD\x00\x00\x00\x7F\x00\x00\xFF\xFD", + nSrc: 5, + t: utf32BEIB.NewEncoder(), + }, { + desc: "utf-32 dec: Short dst", + src: "\x00\x00\x00\x41", + sizeDst: 0, + want: "", + nSrc: 0, + err: transform.ErrShortDst, + t: utf32BEIB.NewDecoder(), + }, { + desc: "utf-32 dec: Short src", + src: "\x00\x00\x00", + notEOF: true, + sizeDst: 4, + want: "", + nSrc: 0, + err: transform.ErrShortSrc, + t: utf32BEIB.NewDecoder(), + }, { + desc: "utf-32 dec: Invalid input", + src: "\x00\x00\xD8\x00\x00\x00\xDF\xFF\x00\x11\x00\x00\x00\x00\x00", + sizeDst: 100, + want: "\uFFFD\uFFFD\uFFFD\uFFFD", + nSrc: 15, + t: utf32BEIB.NewDecoder(), + }} + for i, tc := range testCases { + b := make([]byte, tc.sizeDst) + nDst, nSrc, err := tc.t.Transform(b, []byte(tc.src), !tc.notEOF) + if err != tc.err { + t.Errorf("%d:%s: error was %v; want %v", i, tc.desc, err, tc.err) + } + if got := string(b[:nDst]); got != tc.want { + t.Errorf("%d:%s: result was %q: want %q", i, tc.desc, got, tc.want) + } + if nSrc != tc.nSrc { + t.Errorf("%d:%s: nSrc was %d; want %d", i, tc.desc, nSrc, tc.nSrc) + } + } +} |