Merge pull request #36 from multiformats/feat/base36

Base36 implementation

.codecov.yml

@@ -0,0 +1 @@
+comment: off

.gitignore

@@ -1,2 +1,3 @@
 *.swp
 
+multibase-conv/multibase-conv

.gitmodules

@@ -0,0 +1,3 @@
+[submodule "spec"]
+	path = spec
+	url = https://github.com/multiformats/multibase.git

.gx/lastpubver

@@ -1 +0,0 @@
-0.1.0: QmRdt1nxGDL6UWo2ziX9no4BDmMd8YZoDP88FgXBpKDZZi

.travis.yml

@@ -0,0 +1,30 @@
+os:
+  - linux
+
+language: go
+
+go:
+  - 1.11.x
+
+env:
+  global:
+    - GOTFLAGS="-race"
+  matrix:
+    - BUILD_DEPTYPE=gomod
+
+
+# disable travis install
+install:
+  - true
+
+script:
+  - bash <(curl -s https://raw.githubusercontent.com/ipfs/ci-helpers/master/travis-ci/run-standard-tests.sh)
+
+
+cache:
+  directories:
+    - $GOPATH/pkg/mod
+    - /home/travis/.cache/go-build
+
+notifications:
+  email: false

LICENSE

@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2016 Protocol Labs Inc.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.

Makefile

@@ -0,0 +1,7 @@
+test: deps
+	go test -count=1 -race -v ./...
+
+export IPFS_API ?= v04x.ipfs.io
+
+deps:
+	go get -t ./...

README.md

@@ -0,0 +1,31 @@
+# go-multibase
+
+[![](https://img.shields.io/badge/made%20by-Protocol%20Labs-blue.svg?style=flat-square)](http://ipn.io)
+[![](https://img.shields.io/badge/project-multiformats-blue.svg?style=flat-square)](https://github.com/multiformats/multiformats)
+[![](https://img.shields.io/badge/freenode-%23ipfs-blue.svg?style=flat-square)](https://webchat.freenode.net/?channels=%23ipfs)
+[![](https://img.shields.io/badge/readme%20style-standard-brightgreen.svg?style=flat-square)](https://github.com/RichardLitt/standard-readme)
+[![Travis CI](https://img.shields.io/travis/multiformats/go-multibase.svg?style=flat-square&branch=master)](https://travis-ci.org/multiformats/go-multibase)
+[![codecov.io](https://img.shields.io/codecov/c/github/multiformats/go-multibase.svg?style=flat-square&branch=master)](https://codecov.io/github/multiformats/go-multibase?branch=master)
+
+> Implementation of [multibase](https://github.com/multiformats/multibase) -self identifying base encodings- in Go.
+
+
+## Install
+
+`go-multibase` is a standard Go module which can be installed with:
+
+```sh
+go get github.com/multiformats/go-multibase
+```
+
+## Contribute
+
+Contributions welcome. Please check out [the issues](https://github.com/multiformats/go-multibase/issues).
+
+Check out our [contributing document](https://github.com/multiformats/multiformats/blob/master/contributing.md) for more information on how we work, and about contributing in general. Please be aware that all interactions related to multiformats are subject to the IPFS [Code of Conduct](https://github.com/ipfs/community/blob/master/code-of-conduct.md).
+
+Small note: If editing the README, please conform to the [standard-readme](https://github.com/RichardLitt/standard-readme) specification.
+
+## License
+
+[MIT](LICENSE) © 2016 Protocol Labs Inc.

base16.go

@@ -0,0 +1,21 @@
+package multibase
+
+func hexEncodeToStringUpper(src []byte) string {
+	dst := make([]byte, len(src)*2)
+	hexEncodeUpper(dst, src)
+	return string(dst)
+}
+
+var hexTableUppers = [16]byte{
+	'0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
+	'A', 'B', 'C', 'D', 'E', 'F',
+}
+
+func hexEncodeUpper(dst, src []byte) int {
+	for i, v := range src {
+		dst[i*2] = hexTableUppers[v>>4]
+		dst[i*2+1] = hexTableUppers[v&0x0f]
+	}
+
+	return len(src) * 2
+}

base2.go

@@ -0,0 +1,52 @@
+package multibase
+
+import (
+	"fmt"
+	"strconv"
+	"strings"
+)
+
+// binaryEncodeToString takes an array of bytes and returns
+// multibase binary representation
+func binaryEncodeToString(src []byte) string {
+	dst := make([]byte, len(src)*8)
+	encodeBinary(dst, src)
+	return string(dst)
+}
+
+// encodeBinary takes the src and dst bytes and converts each
+// byte to their binary rep using power reduction method
+func encodeBinary(dst []byte, src []byte) {
+	for i, b := range src {
+		for j := 0; j < 8; j++ {
+			if b&(1<<uint(7-j)) == 0 {
+				dst[i*8+j] = '0'
+			} else {
+				dst[i*8+j] = '1'
+			}
+		}
+	}
+}
+
+// decodeBinaryString takes multibase binary representation
+// and returns a byte array
+func decodeBinaryString(s string) ([]byte, error) {
+	if len(s)&7 != 0 {
+		// prepend the padding
+		s = strings.Repeat("0", 8-len(s)&7) + s
+	}
+
+	data := make([]byte, len(s)>>3)
+
+	for i, dstIndex := 0, 0; i < len(s); i = i + 8 {
+		value, err := strconv.ParseInt(s[i:i+8], 2, 0)
+		if err != nil {
+			return nil, fmt.Errorf("error while conversion: %s", err)
+		}
+
+		data[dstIndex] = byte(value)
+		dstIndex++
+	}
+
+	return data, nil
+}

base32.go

@@ -0,0 +1,17 @@
+package multibase
+
+import (
+	b32 "github.com/multiformats/go-base32"
+)
+
+var base32StdLowerPad = b32.NewEncodingCI("abcdefghijklmnopqrstuvwxyz234567")
+var base32StdLowerNoPad = base32StdLowerPad.WithPadding(b32.NoPadding)
+
+var base32StdUpperPad = b32.NewEncodingCI("ABCDEFGHIJKLMNOPQRSTUVWXYZ234567")
+var base32StdUpperNoPad = base32StdUpperPad.WithPadding(b32.NoPadding)
+
+var base32HexLowerPad = b32.NewEncodingCI("0123456789abcdefghijklmnopqrstuv")
+var base32HexLowerNoPad = base32HexLowerPad.WithPadding(b32.NoPadding)
+
+var base32HexUpperPad = b32.NewEncodingCI("0123456789ABCDEFGHIJKLMNOPQRSTUV")
+var base32HexUpperNoPad = base32HexUpperPad.WithPadding(b32.NoPadding)

encoder.go

@@ -0,0 +1,63 @@
+package multibase
+
+import (
+	"fmt"
+)
+
+// Encoder is a multibase encoding that is verified to be supported and
+// supports an Encode method that does not return an error
+type Encoder struct {
+	enc Encoding
+}
+
+// NewEncoder create a new Encoder from an Encoding
+func NewEncoder(base Encoding) (Encoder, error) {
+	_, ok := EncodingToStr[base]
+	if !ok {
+		return Encoder{-1}, fmt.Errorf("Unsupported multibase encoding: %d", base)
+	}
+	return Encoder{base}, nil
+}
+
+// MustNewEncoder is like NewEncoder but will panic if the encoding is
+// invalid.
+func MustNewEncoder(base Encoding) Encoder {
+	_, ok := EncodingToStr[base]
+	if !ok {
+		panic("Unsupported multibase encoding")
+	}
+	return Encoder{base}
+}
+
+// EncoderByName creates an encoder from a string, the string can
+// either be the multibase name or single character multibase prefix
+func EncoderByName(str string) (Encoder, error) {
+	var base Encoding
+	ok := true
+	if len(str) == 0 {
+		return Encoder{-1}, fmt.Errorf("Empty multibase encoding")
+	} else if len(str) == 1 {
+		base = Encoding(str[0])
+		_, ok = EncodingToStr[base]
+	} else {
+		base, ok = Encodings[str]
+	}
+	if !ok {
+		return Encoder{-1}, fmt.Errorf("Unsupported multibase encoding: %s", str)
+	}
+	return Encoder{base}, nil
+}
+
+func (p Encoder) Encoding() Encoding {
+	return p.enc
+}
+
+// Encode encodes the multibase using the given Encoder.
+func (p Encoder) Encode(data []byte) string {
+	str, err := Encode(p.enc, data)
+	if err != nil {
+		// should not happen
+		panic(err)
+	}
+	return str
+}

encoder_test.go

@@ -0,0 +1,51 @@
+package multibase
+
+import (
+	"testing"
+)
+
+func TestInvalidCode(t *testing.T) {
+	_, err := NewEncoder('q')
+	if err == nil {
+		t.Error("expected failure")
+	}
+}
+
+func TestInvalidName(t *testing.T) {
+	values := []string{"invalid", "", "q"}
+	for _, val := range values {
+		_, err := EncoderByName(val)
+		if err == nil {
+			t.Errorf("EncoderByName(%v) expected failure", val)
+		}
+	}
+}
+
+func TestEncoder(t *testing.T) {
+	for name, code := range Encodings {
+		encoder, err := NewEncoder(code)
+		if err != nil {
+			t.Fatal(err)
+		}
+		// Make sure the MustNewEncoder doesn't panic
+		MustNewEncoder(code)
+		str, err := Encode(code, sampleBytes)
+		if err != nil {
+			t.Fatal(err)
+		}
+		str2 := encoder.Encode(sampleBytes)
+		if str != str2 {
+			t.Errorf("encoded string mismatch: %s != %s", str, str2)
+		}
+		_, err = EncoderByName(name)
+		if err != nil {
+			t.Fatalf("EncoderByName(%s) failed: %v", name, err)
+		}
+		// Test that an encoder can be created from the single letter
+		// prefix
+		_, err = EncoderByName(str[0:1])
+		if err != nil {
+			t.Fatalf("EncoderByName(%s) failed: %v", str[0:1], err)
+		}
+	}
+}

go.mod

@@ -0,0 +1,9 @@
+module github.com/multiformats/go-multibase
+
+go 1.11
+
+require (
+	github.com/mr-tron/base58 v1.1.0
+	github.com/multiformats/go-base32 v0.0.3
+	github.com/multiformats/go-base36 v0.1.0
+)

go.sum

@@ -0,0 +1,6 @@
+github.com/mr-tron/base58 v1.1.0 h1:Y51FGVJ91WBqCEabAi5OPUz38eAx8DakuAm5svLcsfQ=
+github.com/mr-tron/base58 v1.1.0/go.mod h1:xcD2VGqlgYjBdcBLw+TuYLr8afG+Hj8g2eTVqeSzSU8=
+github.com/multiformats/go-base32 v0.0.3 h1:tw5+NhuwaOjJCC5Pp82QuXbrmLzWg7uxlMFp8Nq/kkI=
+github.com/multiformats/go-base32 v0.0.3/go.mod h1:pLiuGC8y0QR3Ue4Zug5UzK9LjgbkL8NSQj0zQ5Nz/AA=
+github.com/multiformats/go-base36 v0.1.0 h1:JR6TyF7JjGd3m6FbLU2cOxhC0Li8z8dLNGQ89tUg4F4=
+github.com/multiformats/go-base36 v0.1.0/go.mod h1:kFGE83c6s80PklsHO9sRn2NCoffoRdUUOENyW/Vv6sM=

multibase-conv/main.go

@@ -0,0 +1,41 @@
+package main
+
+import (
+	"fmt"
+	"os"
+
+	multibase "github.com/multiformats/go-multibase"
+)
+
+func main() {
+	if len(os.Args) < 3 {
+		fmt.Printf("usage: %s <new-base> <multibase-str>...\n", os.Args[0])
+		os.Exit(1)
+	}
+
+	var newBase multibase.Encoding
+	if baseParam := os.Args[1]; len(baseParam) != 0 {
+		newBase = multibase.Encoding(baseParam[0])
+	} else {
+		fmt.Fprintln(os.Stderr, "<new-base> is empty")
+		os.Exit(1)
+	}
+
+	input := os.Args[2:]
+
+	for _, strmbase := range input {
+		_, data, err := multibase.Decode(strmbase)
+		if err != nil {
+			fmt.Fprintf(os.Stderr, "error while decoding: %s\n", err)
+			os.Exit(1)
+		}
+
+		newCid, err := multibase.Encode(newBase, data)
+		if err != nil {
+			fmt.Fprintf(os.Stderr, "error while encoding: %s\n", err)
+			os.Exit(1)
+		}
+		fmt.Println(newCid)
+	}
+
+}

multibase.go

@@ -1,39 +1,184 @@
 package multibase
 
 import (
+	"encoding/base64"
 	"encoding/hex"
 	"fmt"
 
-	b58 "github.com/jbenet/go-base58"
+	b58 "github.com/mr-tron/base58/base58"
+	b32 "github.com/multiformats/go-base32"
+	b36 "github.com/multiformats/go-base36"
 )
 
+// Encoding identifies the type of base-encoding that a multibase is carrying.
+type Encoding int
+
+// These are the encodings specified in the standard, not are all
+// supported yet
 const (
-	Base1        = '1'
-	Base2        = '0'
-	Base8        = '7'
-	Base10       = '9'
-	Base16       = 'f'
-	Base58Flickr = 'Z'
-	Base58BTC    = 'z'
+	Identity          = 0x00
+	Base2             = '0'
+	Base8             = '7'
+	Base10            = '9'
+	Base16            = 'f'
+	Base16Upper       = 'F'
+	Base32            = 'b'
+	Base32Upper       = 'B'
+	Base32pad         = 'c'
+	Base32padUpper    = 'C'
+	Base32hex         = 'v'
+	Base32hexUpper    = 'V'
+	Base32hexPad      = 't'
+	Base32hexPadUpper = 'T'
+	Base36            = 'k'
+	Base36Upper       = 'K'
+	Base58BTC         = 'z'
+	Base58Flickr      = 'Z'
+	Base64            = 'm'
+	Base64url         = 'u'
+	Base64pad         = 'M'
+	Base64urlPad      = 'U'
 )
 
+// EncodingToStr is a map of the supported encoding, unsupported encoding
+// specified in standard are left out
+var EncodingToStr = map[Encoding]string{
+	0x00: "identity",
+	'0':  "base2",
+	'f':  "base16",
+	'F':  "base16upper",
+	'b':  "base32",
+	'B':  "base32upper",
+	'c':  "base32pad",
+	'C':  "base32padupper",
+	'v':  "base32hex",
+	'V':  "base32hexupper",
+	't':  "base32hexpad",
+	'T':  "base32hexpadupper",
+	'k':  "base36",
+	'K':  "base36upper",
+	'z':  "base58btc",
+	'Z':  "base58flickr",
+	'm':  "base64",
+	'u':  "base64url",
+	'M':  "base64pad",
+	'U':  "base64urlpad",
+}
+
+var Encodings = map[string]Encoding{}
+
+func init() {
+	for e, n := range EncodingToStr {
+		Encodings[n] = e
+	}
+}
+
+// ErrUnsupportedEncoding is returned when the selected encoding is not known or
+// implemented.
 var ErrUnsupportedEncoding = fmt.Errorf("selected encoding not supported")
 
-func Encode(base int, data []byte) (string, error) {
+// Encode encodes a given byte slice with the selected encoding and returns a
+// multibase string (<encoding><base-encoded-string>). It will return
+// an error if the selected base is not known.
+func Encode(base Encoding, data []byte) (string, error) {
 	switch base {
-	case Base58BTC:
-		return string(Base58BTC) + b58.EncodeAlphabet(data, b58.BTCAlphabet), nil
+	case Identity:
+		// 0x00 inside a string is OK in golang and causes no problems with the length calculation.
+		return string(Identity) + string(data), nil
+	case Base2:
+		return string(Base2) + binaryEncodeToString(data), nil
 	case Base16:
 		return string(Base16) + hex.EncodeToString(data), nil
+	case Base16Upper:
+		return string(Base16Upper) + hexEncodeToStringUpper(data), nil
+	case Base32:
+		return string(Base32) + base32StdLowerNoPad.EncodeToString(data), nil
+	case Base32Upper:
+		return string(Base32Upper) + base32StdUpperNoPad.EncodeToString(data), nil
+	case Base32hex:
+		return string(Base32hex) + base32HexLowerNoPad.EncodeToString(data), nil
+	case Base32hexUpper:
+		return string(Base32hexUpper) + base32HexUpperNoPad.EncodeToString(data), nil
+	case Base32pad:
+		return string(Base32pad) + base32StdLowerPad.EncodeToString(data), nil
+	case Base32padUpper:
+		return string(Base32padUpper) + base32StdUpperPad.EncodeToString(data), nil
+	case Base32hexPad:
+		return string(Base32hexPad) + base32HexLowerPad.EncodeToString(data), nil
+	case Base32hexPadUpper:
+		return string(Base32hexPadUpper) + base32HexUpperPad.EncodeToString(data), nil
+	case Base36:
+		return string(Base36) + b36.EncodeToStringLc(data), nil
+	case Base36Upper:
+		return string(Base36Upper) + b36.EncodeToStringUc(data), nil
+	case Base58BTC:
+		return string(Base58BTC) + b58.EncodeAlphabet(data, b58.BTCAlphabet), nil
+	case Base58Flickr:
+		return string(Base58Flickr) + b58.EncodeAlphabet(data, b58.FlickrAlphabet), nil
+	case Base64pad:
+		return string(Base64pad) + base64.StdEncoding.EncodeToString(data), nil
+	case Base64urlPad:
+		return string(Base64urlPad) + base64.URLEncoding.EncodeToString(data), nil
+	case Base64url:
+		return string(Base64url) + base64.RawURLEncoding.EncodeToString(data), nil
+	case Base64:
+		return string(Base64) + base64.RawStdEncoding.EncodeToString(data), nil
 	default:
 		return "", ErrUnsupportedEncoding
 	}
 }
 
-func Decode(data string) (int, []byte, error) {
-	switch data[0] {
+// Decode takes a multibase string and decodes into a bytes buffer.
+// It will return an error if the selected base is not known.
+func Decode(data string) (Encoding, []byte, error) {
+	if len(data) == 0 {
+		return 0, nil, fmt.Errorf("cannot decode multibase for zero length string")
+	}
+
+	enc := Encoding(data[0])
+
+	switch enc {
+	case Identity:
+		return Identity, []byte(data[1:]), nil
+	case Base2:
+		bytes, err := decodeBinaryString(data[1:])
+		return enc, bytes, err
+	case Base16, Base16Upper:
+		bytes, err := hex.DecodeString(data[1:])
+		return enc, bytes, err
+	case Base32, Base32Upper:
+		bytes, err := b32.RawStdEncoding.DecodeString(data[1:])
+		return enc, bytes, err
+	case Base32hex, Base32hexUpper:
+		bytes, err := b32.RawHexEncoding.DecodeString(data[1:])
+		return enc, bytes, err
+	case Base32pad, Base32padUpper:
+		bytes, err := b32.StdEncoding.DecodeString(data[1:])
+		return enc, bytes, err
+	case Base32hexPad, Base32hexPadUpper:
+		bytes, err := b32.HexEncoding.DecodeString(data[1:])
+		return enc, bytes, err
+	case Base36, Base36Upper:
+		bytes, err := b36.DecodeString(data[1:])
+		return enc, bytes, err
 	case Base58BTC:
-		return Base58BTC, b58.DecodeAlphabet(data[1:], b58.BTCAlphabet), nil
+		bytes, err := b58.DecodeAlphabet(data[1:], b58.BTCAlphabet)
+		return Base58BTC, bytes, err
+	case Base58Flickr:
+		bytes, err := b58.DecodeAlphabet(data[1:], b58.FlickrAlphabet)
+		return Base58Flickr, bytes, err
+	case Base64pad:
+		bytes, err := base64.StdEncoding.DecodeString(data[1:])
+		return Base64pad, bytes, err
+	case Base64urlPad:
+		bytes, err := base64.URLEncoding.DecodeString(data[1:])
+		return Base64urlPad, bytes, err
+	case Base64:
+		bytes, err := base64.RawStdEncoding.DecodeString(data[1:])
+		return Base64, bytes, err
+	case Base64url:
+		bytes, err := base64.RawURLEncoding.DecodeString(data[1:])
+		return Base64url, bytes, err
 	default:
 		return -1, nil, ErrUnsupportedEncoding
 	}

multibase_test.go

@@ -3,28 +3,296 @@ package multibase
 import (
 	"bytes"
 	"math/rand"
+	"sort"
 	"testing"
 )
 
-func TestBase58RoundTrip(t *testing.T) {
-	buf := make([]byte, 16)
-	rand.Read(buf)
-
-	enc, err := Encode(Base58BTC, buf)
-	if err != nil {
-		t.Fatal(err)
+func TestMap(t *testing.T) {
+	for s, e := range Encodings {
+		s2 := EncodingToStr[e]
+		if s != s2 {
+			t.Errorf("round trip failed on encoding map: %s != %s", s, s2)
+		}
 	}
-
-	e, out, err := Decode(enc)
-	if err != nil {
-		t.Fatal(err)
-	}
-
-	if e != Base58BTC {
-		t.Fatal("got wrong encoding out")
-	}
-
-	if !bytes.Equal(buf, out) {
-		t.Fatal("input wasnt the same as output", buf, out)
+	for e, s := range EncodingToStr {
+		e2 := Encodings[s]
+		if e != e2 {
+			t.Errorf("round trip failed on encoding map: '%c' != '%c'", e, e2)
+		}
+	}
+}
+
+var sampleBytes = []byte("Decentralize everything!!!")
+var encodedSamples = map[Encoding]string{
+	Identity:          string(0x00) + "Decentralize everything!!!",
+	Base2:             "00100010001100101011000110110010101101110011101000111001001100001011011000110100101111010011001010010000001100101011101100110010101110010011110010111010001101000011010010110111001100111001000010010000100100001",
+	Base16:            "f446563656e7472616c697a652065766572797468696e67212121",
+	Base16Upper:       "F446563656E7472616C697A652065766572797468696E67212121",
+	Base32:            "birswgzloorzgc3djpjssazlwmvzhs5dinfxgoijbee",
+	Base32Upper:       "BIRSWGZLOORZGC3DJPJSSAZLWMVZHS5DINFXGOIJBEE",
+	Base32pad:         "cirswgzloorzgc3djpjssazlwmvzhs5dinfxgoijbee======",
+	Base32padUpper:    "CIRSWGZLOORZGC3DJPJSSAZLWMVZHS5DINFXGOIJBEE======",
+	Base32hex:         "v8him6pbeehp62r39f9ii0pbmclp7it38d5n6e89144",
+	Base32hexUpper:    "V8HIM6PBEEHP62R39F9II0PBMCLP7IT38D5N6E89144",
+	Base32hexPad:      "t8him6pbeehp62r39f9ii0pbmclp7it38d5n6e89144======",
+	Base32hexPadUpper: "T8HIM6PBEEHP62R39F9II0PBMCLP7IT38D5N6E89144======",
+	Base36:            "km552ng4dabi4neu1oo8l4i5mndwmpc3mkukwtxy9",
+	Base36Upper:       "KM552NG4DABI4NEU1OO8L4I5MNDWMPC3MKUKWTXY9",
+	Base58BTC:         "z36UQrhJq9fNDS7DiAHM9YXqDHMPfr4EMArvt",
+	Base58Flickr:      "Z36tpRGiQ9Endr7dHahm9xwQdhmoER4emaRVT",
+	Base64:            "mRGVjZW50cmFsaXplIGV2ZXJ5dGhpbmchISE",
+	Base64url:         "uRGVjZW50cmFsaXplIGV2ZXJ5dGhpbmchISE",
+	Base64pad:         "MRGVjZW50cmFsaXplIGV2ZXJ5dGhpbmchISE=",
+	Base64urlPad:      "URGVjZW50cmFsaXplIGV2ZXJ5dGhpbmchISE=",
+}
+
+func testEncode(t *testing.T, encoding Encoding, bytes []byte, expected string) {
+	actual, err := Encode(encoding, bytes)
+	if err != nil {
+		t.Error(err)
+		return
+	}
+	if actual != expected {
+		t.Errorf("encoding failed for %c (%d / %s), expected: %s, got: %s", encoding, encoding, EncodingToStr[encoding], expected, actual)
+	}
+}
+
+func testDecode(t *testing.T, expectedEncoding Encoding, expectedBytes []byte, data string) {
+	actualEncoding, actualBytes, err := Decode(data)
+	if err != nil {
+		t.Error(err)
+		return
+	}
+	if actualEncoding != expectedEncoding {
+		t.Errorf("wrong encoding code, expected: %c (%d), got %c (%d)", expectedEncoding, expectedEncoding, actualEncoding, actualEncoding)
+	}
+	if !bytes.Equal(actualBytes, expectedBytes) {
+		t.Errorf("decoding failed for %c (%d), expected: %v, got %v", actualEncoding, actualEncoding, expectedBytes, actualBytes)
+	}
+}
+
+func TestEncode(t *testing.T) {
+	for encoding := range EncodingToStr {
+		testEncode(t, encoding, sampleBytes, encodedSamples[encoding])
+	}
+}
+
+func TestDecode(t *testing.T) {
+	for encoding := range EncodingToStr {
+		testDecode(t, encoding, sampleBytes, encodedSamples[encoding])
+	}
+}
+
+func TestRoundTrip(t *testing.T) {
+
+	for base := range EncodingToStr {
+		if int(base) == 0 {
+			// skip identity: any byte goes there
+			continue
+		}
+
+		_, _, err := Decode(string(base) + "\u00A0")
+		if err == nil {
+			t.Fatal(EncodingToStr[base] + " decode should fail on low-unicode")
+		}
+
+		_, _, err = Decode(string(base) + "\u1F4A8")
+		if err == nil {
+			t.Fatal(EncodingToStr[base] + " decode should fail on emoji")
+		}
+
+		_, _, err = Decode(string(base) + "!")
+		if err == nil {
+			t.Fatal(EncodingToStr[base] + " decode should fail on punctuation")
+		}
+
+		_, _, err = Decode(string(base) + "\xA0")
+		if err == nil {
+			t.Fatal(EncodingToStr[base] + " decode should fail on high-latin1")
+		}
+	}
+
+	buf := make([]byte, 137+16) // sufficiently large prime number of bytes + another 16 to test leading 0s
+	rand.Read(buf[16:])
+
+	for base := range EncodingToStr {
+
+		// test roundtrip from the full zero-prefixed buffer down to a single byte
+		for i := 0; i < len(buf); i++ {
+
+			// use a copy to verify we are not overwriting the supplied buffer
+			newBuf := make([]byte, len(buf)-i)
+			copy(newBuf, buf[i:])
+
+			enc, err := Encode(base, newBuf)
+			if err != nil {
+				t.Fatal(err)
+			}
+
+			e, out, err := Decode(enc)
+			if err != nil {
+				t.Fatal(err)
+			}
+
+			if e != base {
+				t.Fatal("got wrong encoding out")
+			}
+
+			if !bytes.Equal(newBuf, buf[i:]) {
+				t.Fatal("the provided buffer was modified", buf[i:], out)
+			}
+
+			if !bytes.Equal(buf[i:], out) {
+				t.Fatal("input wasnt the same as output", buf[i:], out)
+			}
+
+			// When we have 3 leading zeroes, do a few extra tests
+			// ( choice of leading zeroes is arbitrary - just cutting down on test permutations )
+
+			if i == 13 {
+
+				// if this is a case-insensitive codec semi-randomly swap case in enc and try again
+				name := EncodingToStr[base]
+				if name[len(name)-5:] == "upper" || Encodings[name+"upper"] > 0 {
+					caseTamperedEnc := []byte(enc)
+
+					for _, j := range []int{3, 5, 8, 13, 21, 23, 29, 47, 52} {
+						if caseTamperedEnc[j] >= 65 && caseTamperedEnc[j] <= 90 {
+							caseTamperedEnc[j] += 32
+						} else if caseTamperedEnc[j] >= 97 && caseTamperedEnc[j] <= 122 {
+							caseTamperedEnc[j] -= 32
+						}
+					}
+
+					e, out, err := Decode(string(caseTamperedEnc))
+					if err != nil {
+						t.Fatal(err)
+					}
+
+					if e != base {
+						t.Fatal("got wrong encoding out")
+					}
+					if !bytes.Equal(buf[i:], out) {
+						t.Fatal("input wasn't the same as output", buf[i:], out)
+					}
+				}
+			}
+		}
+	}
+
+	// Test that nothing overflows
+	maxValueBuf := make([]byte, 131)
+	for i := 0; i < len(maxValueBuf); i++ {
+		maxValueBuf[i] = 0xFF
+	}
+
+	for base := range EncodingToStr {
+
+		// test roundtrip from the complete buffer down to a single byte
+		for i := 0; i < len(maxValueBuf); i++ {
+
+			enc, err := Encode(base, maxValueBuf[i:])
+			if err != nil {
+				t.Fatal(err)
+			}
+
+			e, out, err := Decode(enc)
+			if err != nil {
+				t.Fatal(err)
+			}
+
+			if e != base {
+				t.Fatal("got wrong encoding out")
+			}
+
+			if !bytes.Equal(maxValueBuf[i:], out) {
+				t.Fatal("input wasn't the same as output", maxValueBuf[i:], out)
+			}
+		}
+	}
+
+	_, _, err := Decode("")
+	if err == nil {
+		t.Fatal("shouldn't be able to decode empty string")
+	}
+}
+
+var benchmarkBuf [36]byte // typical CID size
+var benchmarkCodecs []string
+
+func init() {
+	rand.Read(benchmarkBuf[:])
+
+	benchmarkCodecs = make([]string, 0, len(Encodings))
+	for n := range Encodings {
+
+		// // Only bench b36 and b58
+		// if len(n) < 6 || (n[4:6] != "36" && n[4:6] != "58") {
+		// 	continue
+		// }
+
+		benchmarkCodecs = append(benchmarkCodecs, n)
+	}
+	sort.Strings(benchmarkCodecs)
+}
+
+func BenchmarkRoundTrip(b *testing.B) {
+	b.ResetTimer()
+
+	for _, name := range benchmarkCodecs {
+		b.Run(name, func(b *testing.B) {
+			base := Encodings[name]
+			for i := 0; i < b.N; i++ {
+				enc, err := Encode(base, benchmarkBuf[:])
+				if err != nil {
+					b.Fatal(err)
+				}
+
+				e, out, err := Decode(enc)
+				if err != nil {
+					b.Fatal(err)
+				}
+
+				if e != base {
+					b.Fatal("got wrong encoding out")
+				}
+
+				if !bytes.Equal(benchmarkBuf[:], out) {
+					b.Fatal("input wasnt the same as output", benchmarkBuf, out)
+				}
+			}
+		})
+	}
+}
+
+func BenchmarkEncode(b *testing.B) {
+	b.ResetTimer()
+
+	for _, name := range benchmarkCodecs {
+		b.Run(name, func(b *testing.B) {
+			base := Encodings[name]
+			for i := 0; i < b.N; i++ {
+				_, err := Encode(base, benchmarkBuf[:])
+				if err != nil {
+					b.Fatal(err)
+				}
+			}
+		})
+	}
+}
+
+func BenchmarkDecode(b *testing.B) {
+	b.ResetTimer()
+
+	for _, name := range benchmarkCodecs {
+		b.Run(name, func(b *testing.B) {
+			enc, _ := Encode(Encodings[name], benchmarkBuf[:])
+			for i := 0; i < b.N; i++ {
+				_, _, err := Decode(enc)
+				if err != nil {
+					b.Fatal(err)
+				}
+			}
+		})
 	}
 }

package.json

@@ -3,12 +3,8 @@
   "bugs": {
     "url": "https://github.com/multiformats/go-multibase"
   },
-  "gx": {
-    "dvcsimport": "github.com/multiformats/go-multibase"
-  },
-  "gxVersion": "0.8.0",
   "language": "go",
   "license": "",
   "name": "go-multibase",
-  "version": "0.1.0"
+  "version": "0.3.0"
 }

spec_test.go

@@ -0,0 +1,147 @@
+package multibase
+
+import (
+	"encoding/csv"
+	"os"
+	"path/filepath"
+	"strconv"
+	"strings"
+	"testing"
+	"unicode/utf8"
+)
+
+func TestSpec(t *testing.T) {
+	file, err := os.Open("spec/multibase.csv")
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer file.Close()
+
+	reader := csv.NewReader(file)
+	reader.LazyQuotes = false
+	reader.FieldsPerRecord = 4
+	reader.TrimLeadingSpace = true
+
+	values, err := reader.ReadAll()
+	if err != nil {
+		t.Error(err)
+	}
+	expectedEncodings := make(map[Encoding]string, len(values)-1)
+	for _, v := range values[1:] {
+		encoding := v[0]
+		codeStr := v[1]
+
+		var code Encoding
+		if strings.HasPrefix(codeStr, "0x") {
+			i, err := strconv.ParseUint(codeStr[2:], 16, 64)
+			if err != nil {
+				t.Errorf("invalid multibase byte %q", codeStr)
+				continue
+			}
+			code = Encoding(i)
+		} else {
+			codeRune, length := utf8.DecodeRuneInString(codeStr)
+			if code == utf8.RuneError {
+				t.Errorf("multibase %q wasn't valid utf8", codeStr)
+				continue
+			}
+			if length != len(codeStr) {
+				t.Errorf("multibase %q wasn't a single character", codeStr)
+				continue
+			}
+			code = Encoding(codeRune)
+		}
+		expectedEncodings[code] = encoding
+	}
+
+	for name, enc := range Encodings {
+		expectedName, ok := expectedEncodings[enc]
+		if !ok {
+			t.Errorf("encoding %q (%c) not defined in the spec", name, enc)
+			continue
+		}
+		if expectedName != name {
+			t.Errorf("encoding %q (%c) has unexpected name %q", expectedName, enc, name)
+		}
+	}
+}
+func TestSpecVectors(t *testing.T) {
+	files, err := filepath.Glob("spec/tests/test[0-9]*.csv")
+	if err != nil {
+		t.Fatal(err)
+	}
+	for _, fname := range files {
+		t.Run(fname, func(t *testing.T) {
+			file, err := os.Open(fname)
+			if err != nil {
+				t.Error(err)
+				return
+			}
+			defer file.Close()
+			reader := csv.NewReader(file)
+			reader.LazyQuotes = false
+			reader.FieldsPerRecord = 2
+			reader.TrimLeadingSpace = true
+
+			values, err := reader.ReadAll()
+			if err != nil {
+				t.Error(err)
+			}
+			if len(values) == 0 {
+				t.Error("no test values")
+				return
+			}
+			header := values[0]
+
+			var decodeOnly bool
+			switch header[0] {
+			case "encoding":
+			case "non-canonical encoding":
+				decodeOnly = true
+			default:
+				t.Errorf("invalid test spec %q", fname)
+				return
+			}
+
+			testValue, err := strconv.Unquote("\"" + header[1] + "\"")
+			if err != nil {
+				t.Error("failed to unquote testcase:", err)
+				return
+			}
+
+			for _, testCase := range values[1:] {
+				encodingName := testCase[0]
+				expected := testCase[1]
+
+				t.Run(encodingName, func(t *testing.T) {
+					encoder, err := EncoderByName(encodingName)
+					if err != nil {
+						t.Skipf("skipping %s: not supported", encodingName)
+						return
+					}
+					if !decodeOnly {
+						t.Logf("encoding %q with %s", testValue, encodingName)
+						actual := encoder.Encode([]byte(testValue))
+						if expected != actual {
+							t.Errorf("expected %q, got %q", expected, actual)
+						}
+					}
+					t.Logf("decoding %q", expected)
+					encoding, decoded, err := Decode(expected)
+					if err != nil {
+						t.Error("failed to decode:", err)
+						return
+					}
+					expectedEncoding := Encodings[encodingName]
+					if encoding != expectedEncoding {
+						t.Errorf("expected encoding to be %c, got %c", expectedEncoding, encoding)
+					}
+					if string(decoded) != testValue {
+						t.Errorf("failed to decode %q to %q, got %q", expected, testValue, string(decoded))
+					}
+				})
+
+			}
+		})
+	}
+}