diff options
Diffstat (limited to 'src/encoding/base64')
-rw-r--r-- | src/encoding/base64/base64.go | 612 | ||||
-rw-r--r-- | src/encoding/base64/base64_test.go | 530 | ||||
-rw-r--r-- | src/encoding/base64/example_test.go | 83 |
3 files changed, 1225 insertions, 0 deletions
diff --git a/src/encoding/base64/base64.go b/src/encoding/base64/base64.go new file mode 100644 index 0000000..4a3e590 --- /dev/null +++ b/src/encoding/base64/base64.go @@ -0,0 +1,612 @@ +// Copyright 2009 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 base64 implements base64 encoding as specified by RFC 4648. +package base64 + +import ( + "encoding/binary" + "io" + "strconv" +) + +/* + * Encodings + */ + +// An Encoding is a radix 64 encoding/decoding scheme, defined by a +// 64-character alphabet. The most common encoding is the "base64" +// encoding defined in RFC 4648 and used in MIME (RFC 2045) and PEM +// (RFC 1421). RFC 4648 also defines an alternate encoding, which is +// the standard encoding with - and _ substituted for + and /. +type Encoding struct { + encode [64]byte + decodeMap [256]byte + padChar rune + strict bool +} + +const ( + StdPadding rune = '=' // Standard padding character + NoPadding rune = -1 // No padding +) + +const encodeStd = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/" +const encodeURL = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_" + +// NewEncoding returns a new padded Encoding defined by the given alphabet, +// which must be a 64-byte string that does not contain the padding character +// or CR / LF ('\r', '\n'). +// The resulting Encoding uses the default padding character ('='), +// which may be changed or disabled via WithPadding. +func NewEncoding(encoder string) *Encoding { + if len(encoder) != 64 { + panic("encoding alphabet is not 64-bytes long") + } + for i := 0; i < len(encoder); i++ { + if encoder[i] == '\n' || encoder[i] == '\r' { + panic("encoding alphabet contains newline character") + } + } + + e := new(Encoding) + e.padChar = StdPadding + copy(e.encode[:], encoder) + + for i := 0; i < len(e.decodeMap); i++ { + e.decodeMap[i] = 0xFF + } + for i := 0; i < len(encoder); i++ { + e.decodeMap[encoder[i]] = byte(i) + } + return e +} + +// WithPadding creates a new encoding identical to enc except +// with a specified padding character, or NoPadding to disable padding. +// The padding character must not be '\r' or '\n', must not +// be contained in the encoding's alphabet and must be a rune equal or +// below '\xff'. +func (enc Encoding) WithPadding(padding rune) *Encoding { + if padding == '\r' || padding == '\n' || padding > 0xff { + panic("invalid padding") + } + + for i := 0; i < len(enc.encode); i++ { + if rune(enc.encode[i]) == padding { + panic("padding contained in alphabet") + } + } + + enc.padChar = padding + return &enc +} + +// Strict creates a new encoding identical to enc except with +// strict decoding enabled. In this mode, the decoder requires that +// trailing padding bits are zero, as described in RFC 4648 section 3.5. +// +// Note that the input is still malleable, as new line characters +// (CR and LF) are still ignored. +func (enc Encoding) Strict() *Encoding { + enc.strict = true + return &enc +} + +// StdEncoding is the standard base64 encoding, as defined in +// RFC 4648. +var StdEncoding = NewEncoding(encodeStd) + +// URLEncoding is the alternate base64 encoding defined in RFC 4648. +// It is typically used in URLs and file names. +var URLEncoding = NewEncoding(encodeURL) + +// RawStdEncoding is the standard raw, unpadded base64 encoding, +// as defined in RFC 4648 section 3.2. +// This is the same as StdEncoding but omits padding characters. +var RawStdEncoding = StdEncoding.WithPadding(NoPadding) + +// RawURLEncoding is the unpadded alternate base64 encoding defined in RFC 4648. +// It is typically used in URLs and file names. +// This is the same as URLEncoding but omits padding characters. +var RawURLEncoding = URLEncoding.WithPadding(NoPadding) + +/* + * Encoder + */ + +// Encode encodes src using the encoding enc, writing +// EncodedLen(len(src)) bytes to dst. +// +// The encoding pads the output to a multiple of 4 bytes, +// so Encode is not appropriate for use on individual blocks +// of a large data stream. Use NewEncoder() instead. +func (enc *Encoding) Encode(dst, src []byte) { + if len(src) == 0 { + return + } + // enc is a pointer receiver, so the use of enc.encode within the hot + // loop below means a nil check at every operation. Lift that nil check + // outside of the loop to speed up the encoder. + _ = enc.encode + + di, si := 0, 0 + n := (len(src) / 3) * 3 + for si < n { + // Convert 3x 8bit source bytes into 4 bytes + val := uint(src[si+0])<<16 | uint(src[si+1])<<8 | uint(src[si+2]) + + dst[di+0] = enc.encode[val>>18&0x3F] + dst[di+1] = enc.encode[val>>12&0x3F] + dst[di+2] = enc.encode[val>>6&0x3F] + dst[di+3] = enc.encode[val&0x3F] + + si += 3 + di += 4 + } + + remain := len(src) - si + if remain == 0 { + return + } + // Add the remaining small block + val := uint(src[si+0]) << 16 + if remain == 2 { + val |= uint(src[si+1]) << 8 + } + + dst[di+0] = enc.encode[val>>18&0x3F] + dst[di+1] = enc.encode[val>>12&0x3F] + + switch remain { + case 2: + dst[di+2] = enc.encode[val>>6&0x3F] + if enc.padChar != NoPadding { + dst[di+3] = byte(enc.padChar) + } + case 1: + if enc.padChar != NoPadding { + dst[di+2] = byte(enc.padChar) + dst[di+3] = byte(enc.padChar) + } + } +} + +// EncodeToString returns the base64 encoding of src. +func (enc *Encoding) EncodeToString(src []byte) string { + buf := make([]byte, enc.EncodedLen(len(src))) + enc.Encode(buf, src) + return string(buf) +} + +type encoder struct { + err error + enc *Encoding + w io.Writer + buf [3]byte // buffered data waiting to be encoded + nbuf int // number of bytes in buf + out [1024]byte // output buffer +} + +func (e *encoder) Write(p []byte) (n int, err error) { + if e.err != nil { + return 0, e.err + } + + // Leading fringe. + if e.nbuf > 0 { + var i int + for i = 0; i < len(p) && e.nbuf < 3; i++ { + e.buf[e.nbuf] = p[i] + e.nbuf++ + } + n += i + p = p[i:] + if e.nbuf < 3 { + return + } + e.enc.Encode(e.out[:], e.buf[:]) + if _, e.err = e.w.Write(e.out[:4]); e.err != nil { + return n, e.err + } + e.nbuf = 0 + } + + // Large interior chunks. + for len(p) >= 3 { + nn := len(e.out) / 4 * 3 + if nn > len(p) { + nn = len(p) + nn -= nn % 3 + } + e.enc.Encode(e.out[:], p[:nn]) + if _, e.err = e.w.Write(e.out[0 : nn/3*4]); e.err != nil { + return n, e.err + } + n += nn + p = p[nn:] + } + + // Trailing fringe. + copy(e.buf[:], p) + e.nbuf = len(p) + n += len(p) + return +} + +// Close flushes any pending output from the encoder. +// It is an error to call Write after calling Close. +func (e *encoder) Close() error { + // If there's anything left in the buffer, flush it out + if e.err == nil && e.nbuf > 0 { + e.enc.Encode(e.out[:], e.buf[:e.nbuf]) + _, e.err = e.w.Write(e.out[:e.enc.EncodedLen(e.nbuf)]) + e.nbuf = 0 + } + return e.err +} + +// NewEncoder returns a new base64 stream encoder. Data written to +// the returned writer will be encoded using enc and then written to w. +// Base64 encodings operate in 4-byte blocks; when finished +// writing, the caller must Close the returned encoder to flush any +// partially written blocks. +func NewEncoder(enc *Encoding, w io.Writer) io.WriteCloser { + return &encoder{enc: enc, w: w} +} + +// EncodedLen returns the length in bytes of the base64 encoding +// of an input buffer of length n. +func (enc *Encoding) EncodedLen(n int) int { + if enc.padChar == NoPadding { + return (n*8 + 5) / 6 // minimum # chars at 6 bits per char + } + return (n + 2) / 3 * 4 // minimum # 4-char quanta, 3 bytes each +} + +/* + * Decoder + */ + +type CorruptInputError int64 + +func (e CorruptInputError) Error() string { + return "illegal base64 data at input byte " + strconv.FormatInt(int64(e), 10) +} + +// decodeQuantum decodes up to 4 base64 bytes. The received parameters are +// the destination buffer dst, the source buffer src and an index in the +// source buffer si. +// It returns the number of bytes read from src, the number of bytes written +// to dst, and an error, if any. +func (enc *Encoding) decodeQuantum(dst, src []byte, si int) (nsi, n int, err error) { + // Decode quantum using the base64 alphabet + var dbuf [4]byte + dlen := 4 + + // Lift the nil check outside of the loop. + _ = enc.decodeMap + + for j := 0; j < len(dbuf); j++ { + if len(src) == si { + switch { + case j == 0: + return si, 0, nil + case j == 1, enc.padChar != NoPadding: + return si, 0, CorruptInputError(si - j) + } + dlen = j + break + } + in := src[si] + si++ + + out := enc.decodeMap[in] + if out != 0xff { + dbuf[j] = out + continue + } + + if in == '\n' || in == '\r' { + j-- + continue + } + + if rune(in) != enc.padChar { + return si, 0, CorruptInputError(si - 1) + } + + // We've reached the end and there's padding + switch j { + case 0, 1: + // incorrect padding + return si, 0, CorruptInputError(si - 1) + case 2: + // "==" is expected, the first "=" is already consumed. + // skip over newlines + for si < len(src) && (src[si] == '\n' || src[si] == '\r') { + si++ + } + if si == len(src) { + // not enough padding + return si, 0, CorruptInputError(len(src)) + } + if rune(src[si]) != enc.padChar { + // incorrect padding + return si, 0, CorruptInputError(si - 1) + } + + si++ + } + + // skip over newlines + for si < len(src) && (src[si] == '\n' || src[si] == '\r') { + si++ + } + if si < len(src) { + // trailing garbage + err = CorruptInputError(si) + } + dlen = j + break + } + + // Convert 4x 6bit source bytes into 3 bytes + val := uint(dbuf[0])<<18 | uint(dbuf[1])<<12 | uint(dbuf[2])<<6 | uint(dbuf[3]) + dbuf[2], dbuf[1], dbuf[0] = byte(val>>0), byte(val>>8), byte(val>>16) + switch dlen { + case 4: + dst[2] = dbuf[2] + dbuf[2] = 0 + fallthrough + case 3: + dst[1] = dbuf[1] + if enc.strict && dbuf[2] != 0 { + return si, 0, CorruptInputError(si - 1) + } + dbuf[1] = 0 + fallthrough + case 2: + dst[0] = dbuf[0] + if enc.strict && (dbuf[1] != 0 || dbuf[2] != 0) { + return si, 0, CorruptInputError(si - 2) + } + } + + return si, dlen - 1, err +} + +// DecodeString returns the bytes represented by the base64 string s. +func (enc *Encoding) DecodeString(s string) ([]byte, error) { + dbuf := make([]byte, enc.DecodedLen(len(s))) + n, err := enc.Decode(dbuf, []byte(s)) + return dbuf[:n], err +} + +type decoder struct { + err error + readErr error // error from r.Read + enc *Encoding + r io.Reader + buf [1024]byte // leftover input + nbuf int + out []byte // leftover decoded output + outbuf [1024 / 4 * 3]byte +} + +func (d *decoder) Read(p []byte) (n int, err error) { + // Use leftover decoded output from last read. + if len(d.out) > 0 { + n = copy(p, d.out) + d.out = d.out[n:] + return n, nil + } + + if d.err != nil { + return 0, d.err + } + + // This code assumes that d.r strips supported whitespace ('\r' and '\n'). + + // Refill buffer. + for d.nbuf < 4 && d.readErr == nil { + nn := len(p) / 3 * 4 + if nn < 4 { + nn = 4 + } + if nn > len(d.buf) { + nn = len(d.buf) + } + nn, d.readErr = d.r.Read(d.buf[d.nbuf:nn]) + d.nbuf += nn + } + + if d.nbuf < 4 { + if d.enc.padChar == NoPadding && d.nbuf > 0 { + // Decode final fragment, without padding. + var nw int + nw, d.err = d.enc.Decode(d.outbuf[:], d.buf[:d.nbuf]) + d.nbuf = 0 + d.out = d.outbuf[:nw] + n = copy(p, d.out) + d.out = d.out[n:] + if n > 0 || len(p) == 0 && len(d.out) > 0 { + return n, nil + } + if d.err != nil { + return 0, d.err + } + } + d.err = d.readErr + if d.err == io.EOF && d.nbuf > 0 { + d.err = io.ErrUnexpectedEOF + } + return 0, d.err + } + + // Decode chunk into p, or d.out and then p if p is too small. + nr := d.nbuf / 4 * 4 + nw := d.nbuf / 4 * 3 + if nw > len(p) { + nw, d.err = d.enc.Decode(d.outbuf[:], d.buf[:nr]) + d.out = d.outbuf[:nw] + n = copy(p, d.out) + d.out = d.out[n:] + } else { + n, d.err = d.enc.Decode(p, d.buf[:nr]) + } + d.nbuf -= nr + copy(d.buf[:d.nbuf], d.buf[nr:]) + return n, d.err +} + +// Decode decodes src using the encoding enc. It writes at most +// DecodedLen(len(src)) bytes to dst and returns the number of bytes +// written. If src contains invalid base64 data, it will return the +// number of bytes successfully written and CorruptInputError. +// New line characters (\r and \n) are ignored. +func (enc *Encoding) Decode(dst, src []byte) (n int, err error) { + if len(src) == 0 { + return 0, nil + } + + // Lift the nil check outside of the loop. enc.decodeMap is directly + // used later in this function, to let the compiler know that the + // receiver can't be nil. + _ = enc.decodeMap + + si := 0 + for strconv.IntSize >= 64 && len(src)-si >= 8 && len(dst)-n >= 8 { + src2 := src[si : si+8] + if dn, ok := assemble64( + enc.decodeMap[src2[0]], + enc.decodeMap[src2[1]], + enc.decodeMap[src2[2]], + enc.decodeMap[src2[3]], + enc.decodeMap[src2[4]], + enc.decodeMap[src2[5]], + enc.decodeMap[src2[6]], + enc.decodeMap[src2[7]], + ); ok { + binary.BigEndian.PutUint64(dst[n:], dn) + n += 6 + si += 8 + } else { + var ninc int + si, ninc, err = enc.decodeQuantum(dst[n:], src, si) + n += ninc + if err != nil { + return n, err + } + } + } + + for len(src)-si >= 4 && len(dst)-n >= 4 { + src2 := src[si : si+4] + if dn, ok := assemble32( + enc.decodeMap[src2[0]], + enc.decodeMap[src2[1]], + enc.decodeMap[src2[2]], + enc.decodeMap[src2[3]], + ); ok { + binary.BigEndian.PutUint32(dst[n:], dn) + n += 3 + si += 4 + } else { + var ninc int + si, ninc, err = enc.decodeQuantum(dst[n:], src, si) + n += ninc + if err != nil { + return n, err + } + } + } + + for si < len(src) { + var ninc int + si, ninc, err = enc.decodeQuantum(dst[n:], src, si) + n += ninc + if err != nil { + return n, err + } + } + return n, err +} + +// assemble32 assembles 4 base64 digits into 3 bytes. +// Each digit comes from the decode map, and will be 0xff +// if it came from an invalid character. +func assemble32(n1, n2, n3, n4 byte) (dn uint32, ok bool) { + // Check that all the digits are valid. If any of them was 0xff, their + // bitwise OR will be 0xff. + if n1|n2|n3|n4 == 0xff { + return 0, false + } + return uint32(n1)<<26 | + uint32(n2)<<20 | + uint32(n3)<<14 | + uint32(n4)<<8, + true +} + +// assemble64 assembles 8 base64 digits into 6 bytes. +// Each digit comes from the decode map, and will be 0xff +// if it came from an invalid character. +func assemble64(n1, n2, n3, n4, n5, n6, n7, n8 byte) (dn uint64, ok bool) { + // Check that all the digits are valid. If any of them was 0xff, their + // bitwise OR will be 0xff. + if n1|n2|n3|n4|n5|n6|n7|n8 == 0xff { + return 0, false + } + return uint64(n1)<<58 | + uint64(n2)<<52 | + uint64(n3)<<46 | + uint64(n4)<<40 | + uint64(n5)<<34 | + uint64(n6)<<28 | + uint64(n7)<<22 | + uint64(n8)<<16, + true +} + +type newlineFilteringReader struct { + wrapped io.Reader +} + +func (r *newlineFilteringReader) Read(p []byte) (int, error) { + n, err := r.wrapped.Read(p) + for n > 0 { + offset := 0 + for i, b := range p[:n] { + if b != '\r' && b != '\n' { + if i != offset { + p[offset] = b + } + offset++ + } + } + if offset > 0 { + return offset, err + } + // Previous buffer entirely whitespace, read again + n, err = r.wrapped.Read(p) + } + return n, err +} + +// NewDecoder constructs a new base64 stream decoder. +func NewDecoder(enc *Encoding, r io.Reader) io.Reader { + return &decoder{enc: enc, r: &newlineFilteringReader{r}} +} + +// DecodedLen returns the maximum length in bytes of the decoded data +// corresponding to n bytes of base64-encoded data. +func (enc *Encoding) DecodedLen(n int) int { + if enc.padChar == NoPadding { + // Unpadded data may end with partial block of 2-3 characters. + return n * 6 / 8 + } + // Padded base64 should always be a multiple of 4 characters in length. + return n / 4 * 3 +} diff --git a/src/encoding/base64/base64_test.go b/src/encoding/base64/base64_test.go new file mode 100644 index 0000000..57256a3 --- /dev/null +++ b/src/encoding/base64/base64_test.go @@ -0,0 +1,530 @@ +// Copyright 2009 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 base64 + +import ( + "bytes" + "errors" + "fmt" + "io" + "reflect" + "runtime/debug" + "strings" + "testing" + "time" +) + +type testpair struct { + decoded, encoded string +} + +var pairs = []testpair{ + // RFC 3548 examples + {"\x14\xfb\x9c\x03\xd9\x7e", "FPucA9l+"}, + {"\x14\xfb\x9c\x03\xd9", "FPucA9k="}, + {"\x14\xfb\x9c\x03", "FPucAw=="}, + + // RFC 4648 examples + {"", ""}, + {"f", "Zg=="}, + {"fo", "Zm8="}, + {"foo", "Zm9v"}, + {"foob", "Zm9vYg=="}, + {"fooba", "Zm9vYmE="}, + {"foobar", "Zm9vYmFy"}, + + // Wikipedia examples + {"sure.", "c3VyZS4="}, + {"sure", "c3VyZQ=="}, + {"sur", "c3Vy"}, + {"su", "c3U="}, + {"leasure.", "bGVhc3VyZS4="}, + {"easure.", "ZWFzdXJlLg=="}, + {"asure.", "YXN1cmUu"}, + {"sure.", "c3VyZS4="}, +} + +// Do nothing to a reference base64 string (leave in standard format) +func stdRef(ref string) string { + return ref +} + +// Convert a reference string to URL-encoding +func urlRef(ref string) string { + ref = strings.ReplaceAll(ref, "+", "-") + ref = strings.ReplaceAll(ref, "/", "_") + return ref +} + +// Convert a reference string to raw, unpadded format +func rawRef(ref string) string { + return strings.TrimRight(ref, "=") +} + +// Both URL and unpadding conversions +func rawURLRef(ref string) string { + return rawRef(urlRef(ref)) +} + +// A nonstandard encoding with a funny padding character, for testing +var funnyEncoding = NewEncoding(encodeStd).WithPadding(rune('@')) + +func funnyRef(ref string) string { + return strings.ReplaceAll(ref, "=", "@") +} + +type encodingTest struct { + enc *Encoding // Encoding to test + conv func(string) string // Reference string converter +} + +var encodingTests = []encodingTest{ + {StdEncoding, stdRef}, + {URLEncoding, urlRef}, + {RawStdEncoding, rawRef}, + {RawURLEncoding, rawURLRef}, + {funnyEncoding, funnyRef}, + {StdEncoding.Strict(), stdRef}, + {URLEncoding.Strict(), urlRef}, + {RawStdEncoding.Strict(), rawRef}, + {RawURLEncoding.Strict(), rawURLRef}, + {funnyEncoding.Strict(), funnyRef}, +} + +var bigtest = testpair{ + "Twas brillig, and the slithy toves", + "VHdhcyBicmlsbGlnLCBhbmQgdGhlIHNsaXRoeSB0b3Zlcw==", +} + +func testEqual(t *testing.T, msg string, args ...any) bool { + t.Helper() + if args[len(args)-2] != args[len(args)-1] { + t.Errorf(msg, args...) + return false + } + return true +} + +func TestEncode(t *testing.T) { + for _, p := range pairs { + for _, tt := range encodingTests { + got := tt.enc.EncodeToString([]byte(p.decoded)) + testEqual(t, "Encode(%q) = %q, want %q", p.decoded, + got, tt.conv(p.encoded)) + } + } +} + +func TestEncoder(t *testing.T) { + for _, p := range pairs { + bb := &bytes.Buffer{} + encoder := NewEncoder(StdEncoding, bb) + encoder.Write([]byte(p.decoded)) + encoder.Close() + testEqual(t, "Encode(%q) = %q, want %q", p.decoded, bb.String(), p.encoded) + } +} + +func TestEncoderBuffering(t *testing.T) { + input := []byte(bigtest.decoded) + for bs := 1; bs <= 12; bs++ { + bb := &bytes.Buffer{} + encoder := NewEncoder(StdEncoding, bb) + for pos := 0; pos < len(input); pos += bs { + end := pos + bs + if end > len(input) { + end = len(input) + } + n, err := encoder.Write(input[pos:end]) + testEqual(t, "Write(%q) gave error %v, want %v", input[pos:end], err, error(nil)) + testEqual(t, "Write(%q) gave length %v, want %v", input[pos:end], n, end-pos) + } + err := encoder.Close() + testEqual(t, "Close gave error %v, want %v", err, error(nil)) + testEqual(t, "Encoding/%d of %q = %q, want %q", bs, bigtest.decoded, bb.String(), bigtest.encoded) + } +} + +func TestDecode(t *testing.T) { + for _, p := range pairs { + for _, tt := range encodingTests { + encoded := tt.conv(p.encoded) + dbuf := make([]byte, tt.enc.DecodedLen(len(encoded))) + count, err := tt.enc.Decode(dbuf, []byte(encoded)) + testEqual(t, "Decode(%q) = error %v, want %v", encoded, err, error(nil)) + testEqual(t, "Decode(%q) = length %v, want %v", encoded, count, len(p.decoded)) + testEqual(t, "Decode(%q) = %q, want %q", encoded, string(dbuf[0:count]), p.decoded) + + dbuf, err = tt.enc.DecodeString(encoded) + testEqual(t, "DecodeString(%q) = error %v, want %v", encoded, err, error(nil)) + testEqual(t, "DecodeString(%q) = %q, want %q", encoded, string(dbuf), p.decoded) + } + } +} + +func TestDecoder(t *testing.T) { + for _, p := range pairs { + decoder := NewDecoder(StdEncoding, strings.NewReader(p.encoded)) + dbuf := make([]byte, StdEncoding.DecodedLen(len(p.encoded))) + count, err := decoder.Read(dbuf) + if err != nil && err != io.EOF { + t.Fatal("Read failed", err) + } + testEqual(t, "Read from %q = length %v, want %v", p.encoded, count, len(p.decoded)) + testEqual(t, "Decoding of %q = %q, want %q", p.encoded, string(dbuf[0:count]), p.decoded) + if err != io.EOF { + _, err = decoder.Read(dbuf) + } + testEqual(t, "Read from %q = %v, want %v", p.encoded, err, io.EOF) + } +} + +func TestDecoderBuffering(t *testing.T) { + for bs := 1; bs <= 12; bs++ { + decoder := NewDecoder(StdEncoding, strings.NewReader(bigtest.encoded)) + buf := make([]byte, len(bigtest.decoded)+12) + var total int + var n int + var err error + for total = 0; total < len(bigtest.decoded) && err == nil; { + n, err = decoder.Read(buf[total : total+bs]) + total += n + } + if err != nil && err != io.EOF { + t.Errorf("Read from %q at pos %d = %d, unexpected error %v", bigtest.encoded, total, n, err) + } + testEqual(t, "Decoding/%d of %q = %q, want %q", bs, bigtest.encoded, string(buf[0:total]), bigtest.decoded) + } +} + +func TestDecodeCorrupt(t *testing.T) { + testCases := []struct { + input string + offset int // -1 means no corruption. + }{ + {"", -1}, + {"\n", -1}, + {"AAA=\n", -1}, + {"AAAA\n", -1}, + {"!!!!", 0}, + {"====", 0}, + {"x===", 1}, + {"=AAA", 0}, + {"A=AA", 1}, + {"AA=A", 2}, + {"AA==A", 4}, + {"AAA=AAAA", 4}, + {"AAAAA", 4}, + {"AAAAAA", 4}, + {"A=", 1}, + {"A==", 1}, + {"AA=", 3}, + {"AA==", -1}, + {"AAA=", -1}, + {"AAAA", -1}, + {"AAAAAA=", 7}, + {"YWJjZA=====", 8}, + {"A!\n", 1}, + {"A=\n", 1}, + } + for _, tc := range testCases { + dbuf := make([]byte, StdEncoding.DecodedLen(len(tc.input))) + _, err := StdEncoding.Decode(dbuf, []byte(tc.input)) + if tc.offset == -1 { + if err != nil { + t.Error("Decoder wrongly detected corruption in", tc.input) + } + continue + } + switch err := err.(type) { + case CorruptInputError: + testEqual(t, "Corruption in %q at offset %v, want %v", tc.input, int(err), tc.offset) + default: + t.Error("Decoder failed to detect corruption in", tc) + } + } +} + +func TestDecodeBounds(t *testing.T) { + var buf [32]byte + s := StdEncoding.EncodeToString(buf[:]) + defer func() { + if err := recover(); err != nil { + t.Fatalf("Decode panicked unexpectedly: %v\n%s", err, debug.Stack()) + } + }() + n, err := StdEncoding.Decode(buf[:], []byte(s)) + if n != len(buf) || err != nil { + t.Fatalf("StdEncoding.Decode = %d, %v, want %d, nil", n, err, len(buf)) + } +} + +func TestEncodedLen(t *testing.T) { + for _, tt := range []struct { + enc *Encoding + n int + want int + }{ + {RawStdEncoding, 0, 0}, + {RawStdEncoding, 1, 2}, + {RawStdEncoding, 2, 3}, + {RawStdEncoding, 3, 4}, + {RawStdEncoding, 7, 10}, + {StdEncoding, 0, 0}, + {StdEncoding, 1, 4}, + {StdEncoding, 2, 4}, + {StdEncoding, 3, 4}, + {StdEncoding, 4, 8}, + {StdEncoding, 7, 12}, + } { + if got := tt.enc.EncodedLen(tt.n); got != tt.want { + t.Errorf("EncodedLen(%d): got %d, want %d", tt.n, got, tt.want) + } + } +} + +func TestDecodedLen(t *testing.T) { + for _, tt := range []struct { + enc *Encoding + n int + want int + }{ + {RawStdEncoding, 0, 0}, + {RawStdEncoding, 2, 1}, + {RawStdEncoding, 3, 2}, + {RawStdEncoding, 4, 3}, + {RawStdEncoding, 10, 7}, + {StdEncoding, 0, 0}, + {StdEncoding, 4, 3}, + {StdEncoding, 8, 6}, + } { + if got := tt.enc.DecodedLen(tt.n); got != tt.want { + t.Errorf("DecodedLen(%d): got %d, want %d", tt.n, got, tt.want) + } + } +} + +func TestBig(t *testing.T) { + n := 3*1000 + 1 + raw := make([]byte, n) + const alpha = "0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ" + for i := 0; i < n; i++ { + raw[i] = alpha[i%len(alpha)] + } + encoded := new(bytes.Buffer) + w := NewEncoder(StdEncoding, encoded) + nn, err := w.Write(raw) + if nn != n || err != nil { + t.Fatalf("Encoder.Write(raw) = %d, %v want %d, nil", nn, err, n) + } + err = w.Close() + if err != nil { + t.Fatalf("Encoder.Close() = %v want nil", err) + } + decoded, err := io.ReadAll(NewDecoder(StdEncoding, encoded)) + if err != nil { + t.Fatalf("io.ReadAll(NewDecoder(...)): %v", err) + } + + if !bytes.Equal(raw, decoded) { + var i int + for i = 0; i < len(decoded) && i < len(raw); i++ { + if decoded[i] != raw[i] { + break + } + } + t.Errorf("Decode(Encode(%d-byte string)) failed at offset %d", n, i) + } +} + +func TestNewLineCharacters(t *testing.T) { + // Each of these should decode to the string "sure", without errors. + const expected = "sure" + examples := []string{ + "c3VyZQ==", + "c3VyZQ==\r", + "c3VyZQ==\n", + "c3VyZQ==\r\n", + "c3VyZ\r\nQ==", + "c3V\ryZ\nQ==", + "c3V\nyZ\rQ==", + "c3VyZ\nQ==", + "c3VyZQ\n==", + "c3VyZQ=\n=", + "c3VyZQ=\r\n\r\n=", + } + for _, e := range examples { + buf, err := StdEncoding.DecodeString(e) + if err != nil { + t.Errorf("Decode(%q) failed: %v", e, err) + continue + } + if s := string(buf); s != expected { + t.Errorf("Decode(%q) = %q, want %q", e, s, expected) + } + } +} + +type nextRead struct { + n int // bytes to return + err error // error to return +} + +// faultInjectReader returns data from source, rate-limited +// and with the errors as written to nextc. +type faultInjectReader struct { + source string + nextc <-chan nextRead +} + +func (r *faultInjectReader) Read(p []byte) (int, error) { + nr := <-r.nextc + if len(p) > nr.n { + p = p[:nr.n] + } + n := copy(p, r.source) + r.source = r.source[n:] + return n, nr.err +} + +// tests that we don't ignore errors from our underlying reader +func TestDecoderIssue3577(t *testing.T) { + next := make(chan nextRead, 10) + wantErr := errors.New("my error") + next <- nextRead{5, nil} + next <- nextRead{10, wantErr} + next <- nextRead{0, wantErr} + d := NewDecoder(StdEncoding, &faultInjectReader{ + source: "VHdhcyBicmlsbGlnLCBhbmQgdGhlIHNsaXRoeSB0b3Zlcw==", // twas brillig... + nextc: next, + }) + errc := make(chan error, 1) + go func() { + _, err := io.ReadAll(d) + errc <- err + }() + select { + case err := <-errc: + if err != wantErr { + t.Errorf("got error %v; want %v", err, wantErr) + } + case <-time.After(5 * time.Second): + t.Errorf("timeout; Decoder blocked without returning an error") + } +} + +func TestDecoderIssue4779(t *testing.T) { + encoded := `CP/EAT8AAAEF +AQEBAQEBAAAAAAAAAAMAAQIEBQYHCAkKCwEAAQUBAQEBAQEAAAAAAAAAAQACAwQFBgcICQoLEAAB +BAEDAgQCBQcGCAUDDDMBAAIRAwQhEjEFQVFhEyJxgTIGFJGhsUIjJBVSwWIzNHKC0UMHJZJT8OHx +Y3M1FqKygyZEk1RkRcKjdDYX0lXiZfKzhMPTdePzRieUpIW0lcTU5PSltcXV5fVWZnaGlqa2xtbm +9jdHV2d3h5ent8fX5/cRAAICAQIEBAMEBQYHBwYFNQEAAhEDITESBEFRYXEiEwUygZEUobFCI8FS +0fAzJGLhcoKSQ1MVY3M08SUGFqKygwcmNcLSRJNUoxdkRVU2dGXi8rOEw9N14/NGlKSFtJXE1OT0 +pbXF1eX1VmZ2hpamtsbW5vYnN0dXZ3eHl6e3x//aAAwDAQACEQMRAD8A9VSSSSUpJJJJSkkkJ+Tj +1kiy1jCJJDnAcCTykpKkuQ6p/jN6FgmxlNduXawwAzaGH+V6jn/R/wCt71zdn+N/qL3kVYFNYB4N +ji6PDVjWpKp9TSXnvTf8bFNjg3qOEa2n6VlLpj/rT/pf567DpX1i6L1hs9Py67X8mqdtg/rUWbbf ++gkp0kkkklKSSSSUpJJJJT//0PVUkkklKVLq3WMDpGI7KzrNjADtYNXvI/Mqr/Pd/q9W3vaxjnvM +NaCXE9gNSvGPrf8AWS3qmba5jjsJhoB0DAf0NDf6sevf+/lf8Hj0JJATfWT6/dV6oXU1uOLQeKKn +EQP+Hubtfe/+R7Mf/g7f5xcocp++Z11JMCJPgFBxOg7/AOuqDx8I/ikpkXkmSdU8mJIJA/O8EMAy +j+mSARB/17pKVXYWHXjsj7yIex0PadzXMO1zT5KHoNA3HT8ietoGhgjsfA+CSnvvqh/jJtqsrwOv +2b6NGNzXfTYexzJ+nU7/ALkf4P8Awv6P9KvTQQ4AgyDqCF85Pho3CTB7eHwXoH+LT65uZbX9X+o2 +bqbPb06551Y4 +` + encodedShort := strings.ReplaceAll(encoded, "\n", "") + + dec := NewDecoder(StdEncoding, strings.NewReader(encoded)) + res1, err := io.ReadAll(dec) + if err != nil { + t.Errorf("ReadAll failed: %v", err) + } + + dec = NewDecoder(StdEncoding, strings.NewReader(encodedShort)) + var res2 []byte + res2, err = io.ReadAll(dec) + if err != nil { + t.Errorf("ReadAll failed: %v", err) + } + + if !bytes.Equal(res1, res2) { + t.Error("Decoded results not equal") + } +} + +func TestDecoderIssue7733(t *testing.T) { + s, err := StdEncoding.DecodeString("YWJjZA=====") + want := CorruptInputError(8) + if !reflect.DeepEqual(want, err) { + t.Errorf("Error = %v; want CorruptInputError(8)", err) + } + if string(s) != "abcd" { + t.Errorf("DecodeString = %q; want abcd", s) + } +} + +func TestDecoderIssue15656(t *testing.T) { + _, err := StdEncoding.Strict().DecodeString("WvLTlMrX9NpYDQlEIFlnDB==") + want := CorruptInputError(22) + if !reflect.DeepEqual(want, err) { + t.Errorf("Error = %v; want CorruptInputError(22)", err) + } + _, err = StdEncoding.Strict().DecodeString("WvLTlMrX9NpYDQlEIFlnDA==") + if err != nil { + t.Errorf("Error = %v; want nil", err) + } + _, err = StdEncoding.DecodeString("WvLTlMrX9NpYDQlEIFlnDB==") + if err != nil { + t.Errorf("Error = %v; want nil", err) + } +} + +func BenchmarkEncodeToString(b *testing.B) { + data := make([]byte, 8192) + b.SetBytes(int64(len(data))) + for i := 0; i < b.N; i++ { + StdEncoding.EncodeToString(data) + } +} + +func BenchmarkDecodeString(b *testing.B) { + sizes := []int{2, 4, 8, 64, 8192} + benchFunc := func(b *testing.B, benchSize int) { + data := StdEncoding.EncodeToString(make([]byte, benchSize)) + b.SetBytes(int64(len(data))) + b.ResetTimer() + for i := 0; i < b.N; i++ { + StdEncoding.DecodeString(data) + } + } + for _, size := range sizes { + b.Run(fmt.Sprintf("%d", size), func(b *testing.B) { + benchFunc(b, size) + }) + } +} + +func TestDecoderRaw(t *testing.T) { + source := "AAAAAA" + want := []byte{0, 0, 0, 0} + + // Direct. + dec1, err := RawURLEncoding.DecodeString(source) + if err != nil || !bytes.Equal(dec1, want) { + t.Errorf("RawURLEncoding.DecodeString(%q) = %x, %v, want %x, nil", source, dec1, err, want) + } + + // Through reader. Used to fail. + r := NewDecoder(RawURLEncoding, bytes.NewReader([]byte(source))) + dec2, err := io.ReadAll(io.LimitReader(r, 100)) + if err != nil || !bytes.Equal(dec2, want) { + t.Errorf("reading NewDecoder(RawURLEncoding, %q) = %x, %v, want %x, nil", source, dec2, err, want) + } + + // Should work with padding. + r = NewDecoder(URLEncoding, bytes.NewReader([]byte(source+"=="))) + dec3, err := io.ReadAll(r) + if err != nil || !bytes.Equal(dec3, want) { + t.Errorf("reading NewDecoder(URLEncoding, %q) = %x, %v, want %x, nil", source+"==", dec3, err, want) + } +} diff --git a/src/encoding/base64/example_test.go b/src/encoding/base64/example_test.go new file mode 100644 index 0000000..61a3adc --- /dev/null +++ b/src/encoding/base64/example_test.go @@ -0,0 +1,83 @@ +// Copyright 2012 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. + +// Keep in sync with ../base32/example_test.go. + +package base64_test + +import ( + "encoding/base64" + "fmt" + "os" +) + +func Example() { + msg := "Hello, 世界" + encoded := base64.StdEncoding.EncodeToString([]byte(msg)) + fmt.Println(encoded) + decoded, err := base64.StdEncoding.DecodeString(encoded) + if err != nil { + fmt.Println("decode error:", err) + return + } + fmt.Println(string(decoded)) + // Output: + // SGVsbG8sIOS4lueVjA== + // Hello, 世界 +} + +func ExampleEncoding_EncodeToString() { + data := []byte("any + old & data") + str := base64.StdEncoding.EncodeToString(data) + fmt.Println(str) + // Output: + // YW55ICsgb2xkICYgZGF0YQ== +} + +func ExampleEncoding_Encode() { + data := []byte("Hello, world!") + dst := make([]byte, base64.StdEncoding.EncodedLen(len(data))) + base64.StdEncoding.Encode(dst, data) + fmt.Println(string(dst)) + // Output: + // SGVsbG8sIHdvcmxkIQ== +} + +func ExampleEncoding_DecodeString() { + str := "c29tZSBkYXRhIHdpdGggACBhbmQg77u/" + data, err := base64.StdEncoding.DecodeString(str) + if err != nil { + fmt.Println("error:", err) + return + } + fmt.Printf("%q\n", data) + // Output: + // "some data with \x00 and \ufeff" +} + +func ExampleEncoding_Decode() { + str := "SGVsbG8sIHdvcmxkIQ==" + dst := make([]byte, base64.StdEncoding.DecodedLen(len(str))) + n, err := base64.StdEncoding.Decode(dst, []byte(str)) + if err != nil { + fmt.Println("decode error:", err) + return + } + dst = dst[:n] + fmt.Printf("%q\n", dst) + // Output: + // "Hello, world!" +} + +func ExampleNewEncoder() { + input := []byte("foo\x00bar") + encoder := base64.NewEncoder(base64.StdEncoding, os.Stdout) + encoder.Write(input) + // Must close the encoder when finished to flush any partial blocks. + // If you comment out the following line, the last partial block "r" + // won't be encoded. + encoder.Close() + // Output: + // Zm9vAGJhcg== +} |