--- /dev/null
+// 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.
+
+package cookiejar
+
+// This file implements the Punycode algorithm from RFC 3492.
+
+import (
+ "fmt"
+ "strings"
+ "unicode/utf8"
+)
+
+// These parameter values are specified in section 5.
+//
+// All computation is done with int32s, so that overflow behavior is identical
+// regardless of whether int is 32-bit or 64-bit.
+const (
+ base int32 = 36
+ damp int32 = 700
+ initialBias int32 = 72
+ initialN int32 = 128
+ skew int32 = 38
+ tmax int32 = 26
+ tmin int32 = 1
+)
+
+// encode encodes a string as specified in section 6.3 and prepends prefix to
+// the result.
+//
+// The "while h < length(input)" line in the specification becomes "for
+// remaining != 0" in the Go code, because len(s) in Go is in bytes, not runes.
+func encode(prefix, s string) (string, error) {
+ output := make([]byte, len(prefix), len(prefix)+1+2*len(s))
+ copy(output, prefix)
+ delta, n, bias := int32(0), initialN, initialBias
+ b, remaining := int32(0), int32(0)
+ for _, r := range s {
+ if r < 0x80 {
+ b++
+ output = append(output, byte(r))
+ } else {
+ remaining++
+ }
+ }
+ h := b
+ if b > 0 {
+ output = append(output, '-')
+ }
+ for remaining != 0 {
+ m := int32(0x7fffffff)
+ for _, r := range s {
+ if m > r && r >= n {
+ m = r
+ }
+ }
+ delta += (m - n) * (h + 1)
+ if delta < 0 {
+ return "", fmt.Errorf("cookiejar: invalid label %q", s)
+ }
+ n = m
+ for _, r := range s {
+ if r < n {
+ delta++
+ if delta < 0 {
+ return "", fmt.Errorf("cookiejar: invalid label %q", s)
+ }
+ continue
+ }
+ if r > n {
+ continue
+ }
+ q := delta
+ for k := base; ; k += base {
+ t := k - bias
+ if t < tmin {
+ t = tmin
+ } else if t > tmax {
+ t = tmax
+ }
+ if q < t {
+ break
+ }
+ output = append(output, encodeDigit(t+(q-t)%(base-t)))
+ q = (q - t) / (base - t)
+ }
+ output = append(output, encodeDigit(q))
+ bias = adapt(delta, h+1, h == b)
+ delta = 0
+ h++
+ remaining--
+ }
+ delta++
+ n++
+ }
+ return string(output), nil
+}
+
+func encodeDigit(digit int32) byte {
+ switch {
+ case 0 <= digit && digit < 26:
+ return byte(digit + 'a')
+ case 26 <= digit && digit < 36:
+ return byte(digit + ('0' - 26))
+ }
+ panic("cookiejar: internal error in punycode encoding")
+}
+
+// adapt is the bias adaptation function specified in section 6.1.
+func adapt(delta, numPoints int32, firstTime bool) int32 {
+ if firstTime {
+ delta /= damp
+ } else {
+ delta /= 2
+ }
+ delta += delta / numPoints
+ k := int32(0)
+ for delta > ((base-tmin)*tmax)/2 {
+ delta /= base - tmin
+ k += base
+ }
+ return k + (base-tmin+1)*delta/(delta+skew)
+}
+
+// Strictly speaking, the remaining code below deals with IDNA (RFC 5890 and
+// friends) and not Punycode (RFC 3492) per se.
+
+// acePrefix is the ASCII Compatible Encoding prefix.
+const acePrefix = "xn--"
+
+// toASCII converts a domain or domain label to its ASCII form. For example,
+// toASCII("bücher.example.com") is "xn--bcher-kva.example.com", and
+// toASCII("golang") is "golang".
+func toASCII(s string) (string, error) {
+ if ascii(s) {
+ return s, nil
+ }
+ labels := strings.Split(s, ".")
+ for i, label := range labels {
+ if !ascii(label) {
+ a, err := encode(acePrefix, label)
+ if err != nil {
+ return "", err
+ }
+ labels[i] = a
+ }
+ }
+ return strings.Join(labels, "."), nil
+}
+
+func ascii(s string) bool {
+ for i := 0; i < len(s); i++ {
+ if s[i] >= utf8.RuneSelf {
+ return false
+ }
+ }
+ return true
+}
--- /dev/null
+// 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.
+
+package cookiejar
+
+import (
+ "testing"
+)
+
+var punycodeTestCases = [...]struct {
+ s, encoded string
+}{
+ {"", ""},
+ {"-", "--"},
+ {"-a", "-a-"},
+ {"-a-", "-a--"},
+ {"a", "a-"},
+ {"a-", "a--"},
+ {"a-b", "a-b-"},
+ {"books", "books-"},
+ {"bücher", "bcher-kva"},
+ {"Hello世界", "Hello-ck1hg65u"},
+ {"ü", "tda"},
+ {"üý", "tdac"},
+
+ // The test cases below come from RFC 3492 section 7.1 with Errata 3026.
+ {
+ // (A) Arabic (Egyptian).
+ "\u0644\u064A\u0647\u0645\u0627\u0628\u062A\u0643\u0644" +
+ "\u0645\u0648\u0634\u0639\u0631\u0628\u064A\u061F",
+ "egbpdaj6bu4bxfgehfvwxn",
+ },
+ {
+ // (B) Chinese (simplified).
+ "\u4ED6\u4EEC\u4E3A\u4EC0\u4E48\u4E0D\u8BF4\u4E2D\u6587",
+ "ihqwcrb4cv8a8dqg056pqjye",
+ },
+ {
+ // (C) Chinese (traditional).
+ "\u4ED6\u5011\u7232\u4EC0\u9EBD\u4E0D\u8AAA\u4E2D\u6587",
+ "ihqwctvzc91f659drss3x8bo0yb",
+ },
+ {
+ // (D) Czech.
+ "\u0050\u0072\u006F\u010D\u0070\u0072\u006F\u0073\u0074" +
+ "\u011B\u006E\u0065\u006D\u006C\u0075\u0076\u00ED\u010D" +
+ "\u0065\u0073\u006B\u0079",
+ "Proprostnemluvesky-uyb24dma41a",
+ },
+ {
+ // (E) Hebrew.
+ "\u05DC\u05DE\u05D4\u05D4\u05DD\u05E4\u05E9\u05D5\u05D8" +
+ "\u05DC\u05D0\u05DE\u05D3\u05D1\u05E8\u05D9\u05DD\u05E2" +
+ "\u05D1\u05E8\u05D9\u05EA",
+ "4dbcagdahymbxekheh6e0a7fei0b",
+ },
+ {
+ // (F) Hindi (Devanagari).
+ "\u092F\u0939\u0932\u094B\u0917\u0939\u093F\u0928\u094D" +
+ "\u0926\u0940\u0915\u094D\u092F\u094B\u0902\u0928\u0939" +
+ "\u0940\u0902\u092C\u094B\u0932\u0938\u0915\u0924\u0947" +
+ "\u0939\u0948\u0902",
+ "i1baa7eci9glrd9b2ae1bj0hfcgg6iyaf8o0a1dig0cd",
+ },
+ {
+ // (G) Japanese (kanji and hiragana).
+ "\u306A\u305C\u307F\u3093\u306A\u65E5\u672C\u8A9E\u3092" +
+ "\u8A71\u3057\u3066\u304F\u308C\u306A\u3044\u306E\u304B",
+ "n8jok5ay5dzabd5bym9f0cm5685rrjetr6pdxa",
+ },
+ {
+ // (H) Korean (Hangul syllables).
+ "\uC138\uACC4\uC758\uBAA8\uB4E0\uC0AC\uB78C\uB4E4\uC774" +
+ "\uD55C\uAD6D\uC5B4\uB97C\uC774\uD574\uD55C\uB2E4\uBA74" +
+ "\uC5BC\uB9C8\uB098\uC88B\uC744\uAE4C",
+ "989aomsvi5e83db1d2a355cv1e0vak1dwrv93d5xbh15a0dt30a5j" +
+ "psd879ccm6fea98c",
+ },
+ {
+ // (I) Russian (Cyrillic).
+ "\u043F\u043E\u0447\u0435\u043C\u0443\u0436\u0435\u043E" +
+ "\u043D\u0438\u043D\u0435\u0433\u043E\u0432\u043E\u0440" +
+ "\u044F\u0442\u043F\u043E\u0440\u0443\u0441\u0441\u043A" +
+ "\u0438",
+ "b1abfaaepdrnnbgefbadotcwatmq2g4l",
+ },
+ {
+ // (J) Spanish.
+ "\u0050\u006F\u0072\u0071\u0075\u00E9\u006E\u006F\u0070" +
+ "\u0075\u0065\u0064\u0065\u006E\u0073\u0069\u006D\u0070" +
+ "\u006C\u0065\u006D\u0065\u006E\u0074\u0065\u0068\u0061" +
+ "\u0062\u006C\u0061\u0072\u0065\u006E\u0045\u0073\u0070" +
+ "\u0061\u00F1\u006F\u006C",
+ "PorqunopuedensimplementehablarenEspaol-fmd56a",
+ },
+ {
+ // (K) Vietnamese.
+ "\u0054\u1EA1\u0069\u0073\u0061\u006F\u0068\u1ECD\u006B" +
+ "\u0068\u00F4\u006E\u0067\u0074\u0068\u1EC3\u0063\u0068" +
+ "\u1EC9\u006E\u00F3\u0069\u0074\u0069\u1EBF\u006E\u0067" +
+ "\u0056\u0069\u1EC7\u0074",
+ "TisaohkhngthchnitingVit-kjcr8268qyxafd2f1b9g",
+ },
+ {
+ // (L) 3<nen>B<gumi><kinpachi><sensei>.
+ "\u0033\u5E74\u0042\u7D44\u91D1\u516B\u5148\u751F",
+ "3B-ww4c5e180e575a65lsy2b",
+ },
+ {
+ // (M) <amuro><namie>-with-SUPER-MONKEYS.
+ "\u5B89\u5BA4\u5948\u7F8E\u6075\u002D\u0077\u0069\u0074" +
+ "\u0068\u002D\u0053\u0055\u0050\u0045\u0052\u002D\u004D" +
+ "\u004F\u004E\u004B\u0045\u0059\u0053",
+ "-with-SUPER-MONKEYS-pc58ag80a8qai00g7n9n",
+ },
+ {
+ // (N) Hello-Another-Way-<sorezore><no><basho>.
+ "\u0048\u0065\u006C\u006C\u006F\u002D\u0041\u006E\u006F" +
+ "\u0074\u0068\u0065\u0072\u002D\u0057\u0061\u0079\u002D" +
+ "\u305D\u308C\u305E\u308C\u306E\u5834\u6240",
+ "Hello-Another-Way--fc4qua05auwb3674vfr0b",
+ },
+ {
+ // (O) <hitotsu><yane><no><shita>2.
+ "\u3072\u3068\u3064\u5C4B\u6839\u306E\u4E0B\u0032",
+ "2-u9tlzr9756bt3uc0v",
+ },
+ {
+ // (P) Maji<de>Koi<suru>5<byou><mae>
+ "\u004D\u0061\u006A\u0069\u3067\u004B\u006F\u0069\u3059" +
+ "\u308B\u0035\u79D2\u524D",
+ "MajiKoi5-783gue6qz075azm5e",
+ },
+ {
+ // (Q) <pafii>de<runba>
+ "\u30D1\u30D5\u30A3\u30FC\u0064\u0065\u30EB\u30F3\u30D0",
+ "de-jg4avhby1noc0d",
+ },
+ {
+ // (R) <sono><supiido><de>
+ "\u305D\u306E\u30B9\u30D4\u30FC\u30C9\u3067",
+ "d9juau41awczczp",
+ },
+ {
+ // (S) -> $1.00 <-
+ "\u002D\u003E\u0020\u0024\u0031\u002E\u0030\u0030\u0020" +
+ "\u003C\u002D",
+ "-> $1.00 <--",
+ },
+}
+
+func TestPunycode(t *testing.T) {
+ for _, tc := range punycodeTestCases {
+ if got, err := encode("", tc.s); err != nil {
+ t.Errorf(`encode("", %q): %v`, tc.s, err)
+ } else if got != tc.encoded {
+ t.Errorf(`encode("", %q): got %q, want %q`, tc.s, got, tc.encoded)
+ }
+ }
+}