vendor/golang.org/x/text/encoding/unicode/unicode_test.go

   1 // Copyright 2015 The Go Authors. All rights reserved.
   2 // Use of this source code is governed by a BSD-style
   3 // license that can be found in the LICENSE file.
   4
   5 package unicode
   6
   7 import (
   8         "testing"
   9
  10         "golang.org/x/text/encoding"
  11         "golang.org/x/text/encoding/charmap"
  12         "golang.org/x/text/encoding/internal/enctest"
  13         "golang.org/x/text/transform"
  14 )
  15
  16 func TestBasics(t *testing.T) {
  17         testCases := []struct {
  18                 e         encoding.Encoding
  19                 encPrefix string
  20                 encSuffix string
  21                 encoded   string
  22                 utf8      string
  23         }{{
  24                 e:       utf16BEIB,
  25                 encoded: "\x00\x57\x00\xe4\xd8\x35\xdd\x65",
  26                 utf8:    "\x57\u00e4\U0001d565",
  27         }, {
  28                 e:         utf16BEEB,
  29                 encPrefix: "\xfe\xff",
  30                 encoded:   "\x00\x57\x00\xe4\xd8\x35\xdd\x65",
  31                 utf8:      "\x57\u00e4\U0001d565",
  32         }, {
  33                 e:       utf16LEIB,
  34                 encoded: "\x57\x00\xe4\x00\x35\xd8\x65\xdd",
  35                 utf8:    "\x57\u00e4\U0001d565",
  36         }, {
  37                 e:         utf16LEEB,
  38                 encPrefix: "\xff\xfe",
  39                 encoded:   "\x57\x00\xe4\x00\x35\xd8\x65\xdd",
  40                 utf8:      "\x57\u00e4\U0001d565",
  41         }}
  42
  43         for _, tc := range testCases {
  44                 enctest.TestEncoding(t, tc.e, tc.encoded, tc.utf8, tc.encPrefix, tc.encSuffix)
  45         }
  46 }
  47
  48 func TestFiles(t *testing.T) {
  49         enctest.TestFile(t, UTF8)
  50         enctest.TestFile(t, utf16LEIB)
  51 }
  52
  53 func BenchmarkEncoding(b *testing.B) {
  54         enctest.Benchmark(b, UTF8)
  55         enctest.Benchmark(b, utf16LEIB)
  56 }
  57
  58 var (
  59         utf16LEIB = UTF16(LittleEndian, IgnoreBOM) // UTF-16LE (atypical interpretation)
  60         utf16LEUB = UTF16(LittleEndian, UseBOM)    // UTF-16, LE
  61         utf16LEEB = UTF16(LittleEndian, ExpectBOM) // UTF-16, LE, Expect
  62         utf16BEIB = UTF16(BigEndian, IgnoreBOM)    // UTF-16BE (atypical interpretation)
  63         utf16BEUB = UTF16(BigEndian, UseBOM)       // UTF-16 default
  64         utf16BEEB = UTF16(BigEndian, ExpectBOM)    // UTF-16 Expect
  65 )
  66
  67 func TestUTF16(t *testing.T) {
  68         testCases := []struct {
  69                 desc    string
  70                 src     string
  71                 notEOF  bool // the inverse of atEOF
  72                 sizeDst int
  73                 want    string
  74                 nSrc    int
  75                 err     error
  76                 t       transform.Transformer
  77         }{{
  78                 desc: "utf-16 IgnoreBOM dec: empty string",
  79                 t:    utf16BEIB.NewDecoder(),
  80         }, {
  81                 desc: "utf-16 UseBOM dec: empty string",
  82                 t:    utf16BEUB.NewDecoder(),
  83         }, {
  84                 desc: "utf-16 ExpectBOM dec: empty string",
  85                 err:  ErrMissingBOM,
  86                 t:    utf16BEEB.NewDecoder(),
  87         }, {
  88                 desc:    "utf-16 dec: BOM determines encoding BE (RFC 2781:3.3)",
  89                 src:     "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61",
  90                 sizeDst: 100,
  91                 want:    "\U00012345=Ra",
  92                 nSrc:    12,
  93                 t:       utf16BEUB.NewDecoder(),
  94         }, {
  95                 desc:    "utf-16 dec: BOM determines encoding LE (RFC 2781:3.3)",
  96                 src:     "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00",
  97                 sizeDst: 100,
  98                 want:    "\U00012345=Ra",
  99                 nSrc:    12,
 100                 t:       utf16LEUB.NewDecoder(),
 101         }, {
 102                 desc:    "utf-16 dec: BOM determines encoding LE, change default (RFC 2781:3.3)",
 103                 src:     "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00",
 104                 sizeDst: 100,
 105                 want:    "\U00012345=Ra",
 106                 nSrc:    12,
 107                 t:       utf16BEUB.NewDecoder(),
 108         }, {
 109                 desc:    "utf-16 dec: Fail on missing BOM when required",
 110                 src:     "\x08\xD8\x45\xDF\x3D\x00\xFF\xFE\xFE\xFF\x00\x52\x00\x61",
 111                 sizeDst: 100,
 112                 want:    "",
 113                 nSrc:    0,
 114                 err:     ErrMissingBOM,
 115                 t:       utf16BEEB.NewDecoder(),
 116         }, {
 117                 desc:    "utf-16 dec: SHOULD interpret text as big-endian when BOM not present (RFC 2781:4.3)",
 118                 src:     "\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61",
 119                 sizeDst: 100,
 120                 want:    "\U00012345=Ra",
 121                 nSrc:    10,
 122                 t:       utf16BEUB.NewDecoder(),
 123         }, {
 124                 // This is an error according to RFC 2781. But errors in RFC 2781 are
 125                 // open to interpretations, so I guess this is fine.
 126                 desc:    "utf-16le dec: incorrect BOM is an error (RFC 2781:4.1)",
 127                 src:     "\xFE\xFF\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00",
 128                 sizeDst: 100,
 129                 want:    "\uFFFE\U00012345=Ra",
 130                 nSrc:    12,
 131                 t:       utf16LEIB.NewDecoder(),
 132         }, {
 133                 desc:    "utf-16 enc: SHOULD write BOM (RFC 2781:3.3)",
 134                 src:     "\U00012345=Ra",
 135                 sizeDst: 100,
 136                 want:    "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00",
 137                 nSrc:    7,
 138                 t:       utf16LEUB.NewEncoder(),
 139         }, {
 140                 desc:    "utf-16 enc: SHOULD write BOM (RFC 2781:3.3)",
 141                 src:     "\U00012345=Ra",
 142                 sizeDst: 100,
 143                 want:    "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61",
 144                 nSrc:    7,
 145                 t:       utf16BEUB.NewEncoder(),
 146         }, {
 147                 desc:    "utf-16le enc: MUST NOT write BOM (RFC 2781:3.3)",
 148                 src:     "\U00012345=Ra",
 149                 sizeDst: 100,
 150                 want:    "\x08\xD8\x45\xDF\x3D\x00\x52\x00\x61\x00",
 151                 nSrc:    7,
 152                 t:       utf16LEIB.NewEncoder(),
 153         }, {
 154                 desc:    "utf-16be dec: incorrect UTF-16: odd bytes",
 155                 src:     "\x00",
 156                 sizeDst: 100,
 157                 want:    "\uFFFD",
 158                 nSrc:    1,
 159                 t:       utf16BEIB.NewDecoder(),
 160         }, {
 161                 desc:    "utf-16be dec: unpaired surrogate, odd bytes",
 162                 src:     "\xD8\x45\x00",
 163                 sizeDst: 100,
 164                 want:    "\uFFFD\uFFFD",
 165                 nSrc:    3,
 166                 t:       utf16BEIB.NewDecoder(),
 167         }, {
 168                 desc:    "utf-16be dec: unpaired low surrogate + valid text",
 169                 src:     "\xD8\x45\x00a",
 170                 sizeDst: 100,
 171                 want:    "\uFFFDa",
 172                 nSrc:    4,
 173                 t:       utf16BEIB.NewDecoder(),
 174         }, {
 175                 desc:    "utf-16be dec: unpaired low surrogate + valid text + single byte",
 176                 src:     "\xD8\x45\x00ab",
 177                 sizeDst: 100,
 178                 want:    "\uFFFDa\uFFFD",
 179                 nSrc:    5,
 180                 t:       utf16BEIB.NewDecoder(),
 181         }, {
 182                 desc:    "utf-16le dec: unpaired high surrogate",
 183                 src:     "\x00\x00\x00\xDC\x12\xD8",
 184                 sizeDst: 100,
 185                 want:    "\x00\uFFFD\uFFFD",
 186                 nSrc:    6,
 187                 t:       utf16LEIB.NewDecoder(),
 188         }, {
 189                 desc:    "utf-16be dec: two unpaired low surrogates",
 190                 src:     "\xD8\x45\xD8\x12",
 191                 sizeDst: 100,
 192                 want:    "\uFFFD\uFFFD",
 193                 nSrc:    4,
 194                 t:       utf16BEIB.NewDecoder(),
 195         }, {
 196                 desc:    "utf-16be dec: short dst",
 197                 src:     "\x00a",
 198                 sizeDst: 0,
 199                 want:    "",
 200                 nSrc:    0,
 201                 t:       utf16BEIB.NewDecoder(),
 202                 err:     transform.ErrShortDst,
 203         }, {
 204                 desc:    "utf-16be dec: short dst surrogate",
 205                 src:     "\xD8\xF5\xDC\x12",
 206                 sizeDst: 3,
 207                 want:    "",
 208                 nSrc:    0,
 209                 t:       utf16BEIB.NewDecoder(),
 210                 err:     transform.ErrShortDst,
 211         }, {
 212                 desc:    "utf-16be dec: short dst trailing byte",
 213                 src:     "\x00",
 214                 sizeDst: 2,
 215                 want:    "",
 216                 nSrc:    0,
 217                 t:       utf16BEIB.NewDecoder(),
 218                 err:     transform.ErrShortDst,
 219         }, {
 220                 desc:    "utf-16be dec: short src",
 221                 src:     "\x00",
 222                 notEOF:  true,
 223                 sizeDst: 3,
 224                 want:    "",
 225                 nSrc:    0,
 226                 t:       utf16BEIB.NewDecoder(),
 227                 err:     transform.ErrShortSrc,
 228         }, {
 229                 desc:    "utf-16 enc",
 230                 src:     "\U00012345=Ra",
 231                 sizeDst: 100,
 232                 want:    "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61",
 233                 nSrc:    7,
 234                 t:       utf16BEUB.NewEncoder(),
 235         }, {
 236                 desc:    "utf-16 enc: short dst normal",
 237                 src:     "\U00012345=Ra",
 238                 sizeDst: 9,
 239                 want:    "\xD8\x08\xDF\x45\x00\x3D\x00\x52",
 240                 nSrc:    6,
 241                 t:       utf16BEIB.NewEncoder(),
 242                 err:     transform.ErrShortDst,
 243         }, {
 244                 desc:    "utf-16 enc: short dst surrogate",
 245                 src:     "\U00012345=Ra",
 246                 sizeDst: 3,
 247                 want:    "",
 248                 nSrc:    0,
 249                 t:       utf16BEIB.NewEncoder(),
 250                 err:     transform.ErrShortDst,
 251         }, {
 252                 desc:    "utf-16 enc: short src",
 253                 src:     "\U00012345=Ra\xC2",
 254                 notEOF:  true,
 255                 sizeDst: 100,
 256                 want:    "\xD8\x08\xDF\x45\x00\x3D\x00\x52\x00\x61",
 257                 nSrc:    7,
 258                 t:       utf16BEIB.NewEncoder(),
 259                 err:     transform.ErrShortSrc,
 260         }, {
 261                 desc:    "utf-16be dec: don't change byte order mid-stream",
 262                 src:     "\xFE\xFF\xD8\x08\xDF\x45\x00\x3D\xFF\xFE\x00\x52\x00\x61",
 263                 sizeDst: 100,
 264                 want:    "\U00012345=\ufffeRa",
 265                 nSrc:    14,
 266                 t:       utf16BEUB.NewDecoder(),
 267         }, {
 268                 desc:    "utf-16le dec: don't change byte order mid-stream",
 269                 src:     "\xFF\xFE\x08\xD8\x45\xDF\x3D\x00\xFF\xFE\xFE\xFF\x52\x00\x61\x00",
 270                 sizeDst: 100,
 271                 want:    "\U00012345=\ufeff\ufffeRa",
 272                 nSrc:    16,
 273                 t:       utf16LEUB.NewDecoder(),
 274         }}
 275         for i, tc := range testCases {
 276                 b := make([]byte, tc.sizeDst)
 277                 nDst, nSrc, err := tc.t.Transform(b, []byte(tc.src), !tc.notEOF)
 278                 if err != tc.err {
 279                         t.Errorf("%d:%s: error was %v; want %v", i, tc.desc, err, tc.err)
 280                 }
 281                 if got := string(b[:nDst]); got != tc.want {
 282                         t.Errorf("%d:%s: result was %q: want %q", i, tc.desc, got, tc.want)
 283                 }
 284                 if nSrc != tc.nSrc {
 285                         t.Errorf("%d:%s: nSrc was %d; want %d", i, tc.desc, nSrc, tc.nSrc)
 286                 }
 287         }
 288 }
 289
 290 func TestUTF8Decoder(t *testing.T) {
 291         testCases := []struct {
 292                 desc    string
 293                 src     string
 294                 notEOF  bool // the inverse of atEOF
 295                 sizeDst int
 296                 want    string
 297                 nSrc    int
 298                 err     error
 299         }{{
 300                 desc: "empty string, empty dest buffer",
 301         }, {
 302                 desc:    "empty string",
 303                 sizeDst: 8,
 304         }, {
 305                 desc:    "empty string, streaming",
 306                 notEOF:  true,
 307                 sizeDst: 8,
 308         }, {
 309                 desc:    "ascii",
 310                 src:     "abcde",
 311                 sizeDst: 8,
 312                 want:    "abcde",
 313                 nSrc:    5,
 314         }, {
 315                 desc:    "ascii and error",
 316                 src:     "ab\x80de",
 317                 sizeDst: 7,
 318                 want:    "ab\ufffdde",
 319                 nSrc:    5,
 320         }, {
 321                 desc:    "valid two-byte sequence",
 322                 src:     "a\u0300bc",
 323                 sizeDst: 7,
 324                 want:    "a\u0300bc",
 325                 nSrc:    5,
 326         }, {
 327                 desc:    "valid three-byte sequence",
 328                 src:     "a\u0300中",
 329                 sizeDst: 7,
 330                 want:    "a\u0300中",
 331                 nSrc:    6,
 332         }, {
 333                 desc:    "valid four-byte sequence",
 334                 src:     "a中\U00016F50",
 335                 sizeDst: 8,
 336                 want:    "a中\U00016F50",
 337                 nSrc:    8,
 338         }, {
 339                 desc:    "short source buffer",
 340                 src:     "abc\xf0\x90",
 341                 notEOF:  true,
 342                 sizeDst: 10,
 343                 want:    "abc",
 344                 nSrc:    3,
 345                 err:     transform.ErrShortSrc,
 346         }, {
 347                 // We don't check for the maximal subpart of an ill-formed subsequence
 348                 // at the end of an open segment.
 349                 desc:    "complete invalid that looks like short at end",
 350                 src:     "abc\xf0\x80",
 351                 notEOF:  true,
 352                 sizeDst: 10,
 353                 want:    "abc", // instead of "abc\ufffd\ufffd",
 354                 nSrc:    3,
 355                 err:     transform.ErrShortSrc,
 356         }, {
 357                 desc:    "incomplete sequence at end",
 358                 src:     "a\x80bc\xf0\x90",
 359                 sizeDst: 9,
 360                 want:    "a\ufffdbc\ufffd",
 361                 nSrc:    6,
 362         }, {
 363                 desc:    "invalid second byte",
 364                 src:     "abc\xf0dddd",
 365                 sizeDst: 10,
 366                 want:    "abc\ufffddddd",
 367                 nSrc:    8,
 368         }, {
 369                 desc:    "invalid second byte at end",
 370                 src:     "abc\xf0d",
 371                 sizeDst: 10,
 372                 want:    "abc\ufffdd",
 373                 nSrc:    5,
 374         }, {
 375                 desc:    "invalid third byte",
 376                 src:     "a\u0300bc\xf0\x90dddd",
 377                 sizeDst: 12,
 378                 want:    "a\u0300bc\ufffddddd",
 379                 nSrc:    11,
 380         }, {
 381                 desc:    "invalid third byte at end",
 382                 src:     "a\u0300bc\xf0\x90d",
 383                 sizeDst: 12,
 384                 want:    "a\u0300bc\ufffdd",
 385                 nSrc:    8,
 386         }, {
 387                 desc:    "invalid fourth byte, tight buffer",
 388                 src:     "a\u0300bc\xf0\x90\x80d",
 389                 sizeDst: 9,
 390                 want:    "a\u0300bc\ufffdd",
 391                 nSrc:    9,
 392         }, {
 393                 desc:    "invalid fourth byte at end",
 394                 src:     "a\u0300bc\xf0\x90\x80",
 395                 sizeDst: 8,
 396                 want:    "a\u0300bc\ufffd",
 397                 nSrc:    8,
 398         }, {
 399                 desc:    "invalid fourth byte and short four byte sequence",
 400                 src:     "a\u0300bc\xf0\x90\x80\xf0\x90\x80",
 401                 notEOF:  true,
 402                 sizeDst: 20,
 403                 want:    "a\u0300bc\ufffd",
 404                 nSrc:    8,
 405                 err:     transform.ErrShortSrc,
 406         }, {
 407                 desc:    "valid four-byte sequence overflowing short buffer",
 408                 src:     "a\u0300bc\xf0\x90\x80\x80",
 409                 notEOF:  true,
 410                 sizeDst: 8,
 411                 want:    "a\u0300bc",
 412                 nSrc:    5,
 413                 err:     transform.ErrShortDst,
 414         }, {
 415                 desc:    "invalid fourth byte at end short, but short dst",
 416                 src:     "a\u0300bc\xf0\x90\x80\xf0\x90\x80",
 417                 notEOF:  true,
 418                 sizeDst: 8,
 419                 // More bytes would fit in the buffer, but this seems to require a more
 420                 // complicated and slower algorithm.
 421                 want: "a\u0300bc", // instead of "a\u0300bc"
 422                 nSrc: 5,
 423                 err:  transform.ErrShortDst,
 424         }, {
 425                 desc:    "short dst for error",
 426                 src:     "abc\x80",
 427                 notEOF:  true,
 428                 sizeDst: 5,
 429                 want:    "abc",
 430                 nSrc:    3,
 431                 err:     transform.ErrShortDst,
 432         }, {
 433                 desc:    "adjusting short dst buffer",
 434                 src:     "abc\x80ef",
 435                 notEOF:  true,
 436                 sizeDst: 6,
 437                 want:    "abc\ufffd",
 438                 nSrc:    4,
 439                 err:     transform.ErrShortDst,
 440         }}
 441         tr := UTF8.NewDecoder()
 442         for i, tc := range testCases {
 443                 b := make([]byte, tc.sizeDst)
 444                 nDst, nSrc, err := tr.Transform(b, []byte(tc.src), !tc.notEOF)
 445                 if err != tc.err {
 446                         t.Errorf("%d:%s: error was %v; want %v", i, tc.desc, err, tc.err)
 447                 }
 448                 if got := string(b[:nDst]); got != tc.want {
 449                         t.Errorf("%d:%s: result was %q: want %q", i, tc.desc, got, tc.want)
 450                 }
 451                 if nSrc != tc.nSrc {
 452                         t.Errorf("%d:%s: nSrc was %d; want %d", i, tc.desc, nSrc, tc.nSrc)
 453                 }
 454         }
 455 }
 456
 457 func TestBOMOverride(t *testing.T) {
 458         dec := BOMOverride(charmap.CodePage437.NewDecoder())
 459         dst := make([]byte, 100)
 460         for i, tc := range []struct {
 461                 src   string
 462                 atEOF bool
 463                 dst   string
 464                 nSrc  int
 465                 err   error
 466         }{
 467                 0:  {"H\x82ll\x93", true, "Héllô", 5, nil},
 468                 1:  {"\uFEFFHéllö", true, "Héllö", 10, nil},
 469                 2:  {"\xFE\xFF\x00H\x00e\x00l\x00l\x00o", true, "Hello", 12, nil},
 470                 3:  {"\xFF\xFEH\x00e\x00l\x00l\x00o\x00", true, "Hello", 12, nil},
 471                 4:  {"\uFEFF", true, "", 3, nil},
 472                 5:  {"\xFE\xFF", true, "", 2, nil},
 473                 6:  {"\xFF\xFE", true, "", 2, nil},
 474                 7:  {"\xEF\xBB", true, "\u2229\u2557", 2, nil},
 475                 8:  {"\xEF", true, "\u2229", 1, nil},
 476                 9:  {"", true, "", 0, nil},
 477                 10: {"\xFE", true, "\u25a0", 1, nil},
 478                 11: {"\xFF", true, "\u00a0", 1, nil},
 479                 12: {"\xEF\xBB", false, "", 0, transform.ErrShortSrc},
 480                 13: {"\xEF", false, "", 0, transform.ErrShortSrc},
 481                 14: {"", false, "", 0, transform.ErrShortSrc},
 482                 15: {"\xFE", false, "", 0, transform.ErrShortSrc},
 483                 16: {"\xFF", false, "", 0, transform.ErrShortSrc},
 484                 17: {"\xFF\xFE", false, "", 0, transform.ErrShortSrc},
 485         } {
 486                 dec.Reset()
 487                 nDst, nSrc, err := dec.Transform(dst, []byte(tc.src), tc.atEOF)
 488                 got := string(dst[:nDst])
 489                 if nSrc != tc.nSrc {
 490                         t.Errorf("%d: nSrc: got %d; want %d", i, nSrc, tc.nSrc)
 491                 }
 492                 if got != tc.dst {
 493                         t.Errorf("%d: got %+q; want %+q", i, got, tc.dst)
 494                 }
 495                 if err != tc.err {
 496                         t.Errorf("%d: error: got %v; want %v", i, err, tc.err)
 497                 }
 498         }
 499 }