intl/lwbrk/tests/TestLineBreak.cpp

   1 /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
   2 /* ***** BEGIN LICENSE BLOCK *****
   3  * Version: MPL 1.1/GPL 2.0/LGPL 2.1
   4  *
   5  * The contents of this file are subject to the Mozilla Public License Version
   6  * 1.1 (the "License"); you may not use this file except in compliance with
   7  * the License. You may obtain a copy of the License at
   8  * http://www.mozilla.org/MPL/
   9  *
  10  * Software distributed under the License is distributed on an "AS IS" basis,
  11  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  12  * for the specific language governing rights and limitations under the
  13  * License.
  14  *
  15  * The Original Code is mozilla.org code.
  16  *
  17  * The Initial Developer of the Original Code is
  18  * Netscape Communications Corporation.
  19  * Portions created by the Initial Developer are Copyright (C) 1998
  20  * the Initial Developer. All Rights Reserved.
  21  *
  22  * Contributor(s):
  23  *
  24  * Alternatively, the contents of this file may be used under the terms of
  25  * either of the GNU General Public License Version 2 or later (the "GPL"),
  26  * or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  27  * in which case the provisions of the GPL or the LGPL are applicable instead
  28  * of those above. If you wish to allow use of your version of this file only
  29  * under the terms of either the GPL or the LGPL, and not to allow others to
  30  * use your version of this file under the terms of the MPL, indicate your
  31  * decision by deleting the provisions above and replace them with the notice
  32  * and other provisions required by the GPL or the LGPL. If you do not delete
  33  * the provisions above, a recipient may use your version of this file under
  34  * the terms of any one of the MPL, the GPL or the LGPL.
  35  *
  36  * ***** END LICENSE BLOCK ***** */
  37 #include <stdio.h>
  38 #include "nsXPCOM.h"
  39 #include "nsIComponentManager.h"
  40 #include "nsISupports.h"
  41 #include "nsServiceManagerUtils.h"
  42 #include "nsILineBreaker.h"
  43 #include "nsIWordBreaker.h"
  44 #include "nsLWBrkCIID.h"
  45 #include "nsStringAPI.h"
  46 #include "nsEmbedString.h"
  47
  48 #define WORK_AROUND_SERVICE_MANAGER_ASSERT
  49
  50 NS_DEFINE_CID(kLBrkCID, NS_LBRK_CID);
  51 NS_DEFINE_CID(kWBrkCID, NS_WBRK_CID);
  52
  53
  54 static char teng1[] =
  55 //          1         2         3         4         5         6         7
  56 //01234567890123456789012345678901234567890123456789012345678901234567890123456789
  57  "This is a test to test(reasonable) line    break. This 0.01123 = 45 x 48.";
  58
  59 static PRUint32 exp1[] = {
  60   4,5,7,8,9,10,14,15,17,18,22,34,35,39,40,41,42,43,49,50,54,55,62,63,64,65,
  61   67,68,69,70
  62 };
  63
  64 static PRUint32 wexp1[] = {
  65
  66   4,5,7,8,9,10,14,15,17,18,22,23,33,34,35,39,43,48,49,50,54,55,56,57,62,63,
  67   64,65,67,68,69,70,72
  68 };
  69 //          1         2         3         4         5         6         7
  70 //01234567890123456789012345678901234567890123456789012345678901234567890123456789
  71 static char teng2[] =
  72  "()((reasonab(l)e) line  break. .01123=45x48.";
  73
  74 static PRUint32 lexp2[] = {
  75   2,12,15,17,18,22,23,24,30,31,37,38,
  76 };
  77 static PRUint32 wexp2[] = {
  78   4,12,13,14,15,16,17,18,22,24,29,30,31,32,37,38,43
  79 };
  80
  81 //          1         2         3         4         5         6         7
  82 //01234567890123456789012345678901234567890123456789012345678901234567890123456789
  83 static char teng3[] =
  84  "It's a test to test(ronae ) line break....";
  85 static PRUint32 exp3[] = {
  86   4, 5, 6,7,11,12,14,15,19,25,27,28,32,33
  87 };
  88 static PRUint32 wexp3[] = {
  89   4,5,6,7,11,12,14,15,19,20,25,26,27,28,32,33,38
  90 };
  91
  92 static char ruler1[] =
  93 "          1         2         3         4         5         6         7  ";
  94 static char ruler2[] =
  95 "0123456789012345678901234567890123456789012345678901234567890123456789012";
  96
  97
  98 PRBool TestASCIILB(nsILineBreaker *lb,
  99                  const char* in, const PRUint32 len,
 100                  const PRUint32* out, PRUint32 outlen)
 101 {
 102          NS_ConvertASCIItoUTF16 eng1(in);
 103          PRUint32 i,j;
 104          PRUint32 res[256];
 105          PRBool ok = PR_TRUE;
 106          PRInt32 curr;
 107          for(i = 0, curr = 0; (curr != NS_LINEBREAKER_NEED_MORE_TEXT) &&
 108              (i < 256); i++)
 109          {
 110             curr = lb->Next(eng1.get(), eng1.Length(), curr);
 111             res [i] = curr != NS_LINEBREAKER_NEED_MORE_TEXT ? curr : eng1.Length();
 112
 113          }
 114          if (i != outlen)
 115          {
 116             ok = PR_FALSE;
 117             printf("WARNING!!! return size wrong, expect %d but got %d \n",
 118                    outlen, i);
 119          }
 120          printf("string  = \n%s\n", in);
 121          printf("%s\n", ruler1);
 122          printf("%s\n", ruler2);
 123          printf("Expect = \n");
 124          for(j=0;j<outlen;j++)
 125          {
 126             printf("%d,", out[j]);
 127          }
 128          printf("\nResult = \n");
 129          for(j=0;j<i;j++)
 130          {
 131             printf("%d,", res[j]);
 132          }
 133          printf("\n");
 134          for(j=0;j<i;j++)
 135          {
 136             if(j < outlen)
 137             {
 138                 if (res[j] != out[j])
 139                 {
 140                    ok = PR_FALSE;
 141                    printf("[%d] expect %d but got %d\n", j, out[j], res[j]);
 142                 }
 143             } else {
 144                    ok = PR_FALSE;
 145                    printf("[%d] additional %d\n", j, res[j]);
 146             }
 147          }
 148          return ok;
 149 }
 150
 151 PRBool TestASCIIWB(nsIWordBreaker *lb,
 152                  const char* in, const PRUint32 len,
 153                  const PRUint32* out, PRUint32 outlen)
 154 {
 155          NS_ConvertASCIItoUTF16 eng1(in);
 156
 157          PRUint32 i,j;
 158          PRUint32 res[256];
 159          PRBool ok = PR_TRUE;
 160          PRInt32 curr = 0;
 161
 162          for(i = 0, curr = lb->NextWord(eng1.get(), eng1.Length(), curr);
 163                     (curr != NS_WORDBREAKER_NEED_MORE_TEXT) && (i < 256);
 164                     curr = lb->NextWord(eng1.get(), eng1.Length(), curr), i++)
 165          {
 166             res [i] = curr != NS_WORDBREAKER_NEED_MORE_TEXT ? curr : eng1.Length();
 167          }
 168          if (i != outlen)
 169          {
 170             ok = PR_FALSE;
 171             printf("WARNING!!! return size wrong, expect %d but got %d\n",
 172                    outlen, i);
 173          }
 174          printf("string  = \n%s\n", in);
 175          printf("%s\n", ruler1);
 176          printf("%s\n", ruler2);
 177          printf("Expect = \n");
 178          for(j=0;j<outlen;j++)
 179          {
 180             printf("%d,", out[j]);
 181          }
 182          printf("\nResult = \n");
 183          for(j=0;j<i;j++)
 184          {
 185             printf("%d,", res[j]);
 186          }
 187          printf("\n");
 188          for(j=0;j<i;j++)
 189          {
 190             if(j < outlen)
 191             {
 192                 if (res[j] != out[j])
 193                 {
 194                    ok = PR_FALSE;
 195                    printf("[%d] expect %d but got %d\n", j, out[j], res[j]);
 196                 }
 197             } else {
 198                    ok = PR_FALSE;
 199                    printf("[%d] additional %d\n", j, res[j]);
 200             }
 201          }
 202          return ok;
 203 }
 204
 205
 206 PRBool TestLineBreaker()
 207 {
 208    printf("===========================\n");
 209    printf("Finish nsILineBreaker Test \n");
 210    printf("===========================\n");
 211    nsILineBreaker *t = NULL;
 212    nsresult res;
 213    PRBool ok = PR_TRUE;
 214    res = CallGetService(kLBrkCID, &t);
 215
 216    printf("Test 1 - GetService():\n");
 217    if(NS_FAILED(res) || ( t == NULL ) ) {
 218      printf("\t1st GetService failed\n");
 219      ok = PR_FALSE;
 220    }
 221
 222    NS_IF_RELEASE(t);
 223
 224    res = CallGetService(kLBrkCID, &t);
 225
 226    if(NS_FAILED(res) || ( t == NULL ) ) {
 227      printf("\t2nd GetService failed\n");
 228      ok = PR_FALSE;
 229    } else {
 230      printf("Test 4 - {First,Next}ForwardBreak():\n");
 231      if( TestASCIILB(t, teng1, sizeof(teng1)/sizeof(char),
 232               exp1, sizeof(exp1)/sizeof(PRUint32)) )
 233      {
 234        printf("Test 4 Passed\n\n");
 235      } else {
 236        ok = PR_FALSE;
 237        printf("Test 4 Failed\n\n");
 238      }
 239
 240      printf("Test 5 - {First,Next}ForwardBreak():\n");
 241      if(TestASCIILB(t, teng2, sizeof(teng2)/sizeof(char),
 242                lexp2, sizeof(lexp2)/sizeof(PRUint32)) )
 243      {
 244        printf("Test 5 Passed\n\n");
 245      } else {
 246        ok = PR_FALSE;
 247        printf("Test 5 Failed\n\n");
 248      }
 249
 250      printf("Test 6 - {First,Next}ForwardBreak():\n");
 251      if(TestASCIILB(t, teng3, sizeof(teng3)/sizeof(char),
 252                exp3, sizeof(exp3)/sizeof(PRUint32)) )
 253      {
 254        printf("Test 6 Passed\n\n");
 255      } else {
 256        ok = PR_FALSE;
 257        printf("Test 6 Failed\n\n");
 258      }
 259
 260
 261      NS_RELEASE(t);
 262
 263    }
 264
 265    printf("===========================\n");
 266    printf("Finish nsILineBreaker Test \n");
 267    printf("===========================\n");
 268
 269  return ok;
 270 }
 271
 272 PRBool TestWordBreaker()
 273 {
 274    printf("===========================\n");
 275    printf("Finish nsIWordBreaker Test \n");
 276    printf("===========================\n");
 277    nsIWordBreaker *t = NULL;
 278    nsresult res;
 279    PRBool ok = PR_TRUE;
 280    res = CallGetService(kWBrkCID, &t);
 281
 282    printf("Test 1 - GetService():\n");
 283    if(NS_FAILED(res) || ( t == NULL ) ) {
 284      printf("\t1st GetService failed\n");
 285      ok = PR_FALSE;
 286    } else {
 287      NS_RELEASE(t);
 288    }
 289
 290    res = CallGetService(kWBrkCID, &t);
 291
 292    if(NS_FAILED(res) || ( t == NULL ) ) {
 293      printf("\t2nd GetService failed\n");
 294      ok = PR_FALSE;
 295    } else {
 296
 297      printf("Test 4 - {First,Next}ForwardBreak():\n");
 298      if( TestASCIIWB(t, teng1, sizeof(teng1)/sizeof(char),
 299                wexp1, sizeof(wexp1)/sizeof(PRUint32)) )
 300      {
 301         printf("Test 4 Passed\n\n");
 302      } else {
 303        ok = PR_FALSE;
 304        printf("Test 4 Failed\n\n");
 305      }
 306
 307      printf("Test 5 - {First,Next}ForwardBreak():\n");
 308      if(TestASCIIWB(t, teng2, sizeof(teng2)/sizeof(char),
 309                wexp2, sizeof(wexp2)/sizeof(PRUint32)) )
 310      {
 311        printf("Test 5 Passed\n\n");
 312      } else {
 313        ok = PR_FALSE;
 314        printf("Test 5 Failed\n\n");
 315      }
 316
 317      printf("Test 6 - {First,Next}ForwardBreak():\n");
 318      if(TestASCIIWB(t, teng3, sizeof(teng3)/sizeof(char),
 319                wexp3, sizeof(wexp3)/sizeof(PRUint32)) )
 320      {
 321        printf("Test 6 Passed\n\n");
 322      } else {
 323        ok = PR_FALSE;
 324        printf("Test 6 Failed\n\n");
 325      }
 326
 327
 328      NS_RELEASE(t);
 329    }
 330
 331    printf("===========================\n");
 332    printf("Finish nsIWordBreaker Test \n");
 333    printf("===========================\n");
 334
 335    return ok;
 336 }
 337
 338 void   SamplePrintWordWithBreak();
 339 void   SampleFindWordBreakFromPosition(PRUint32 fragN, PRUint32 offset);
 340 // Sample Code
 341
 342 //                          012345678901234
 343 static const char wb0[] =  "T";
 344 static const char wb1[] =  "h";
 345 static const char wb2[] =  "is   is a int";
 346 static const char wb3[] =  "ernationali";
 347 static const char wb4[] =  "zation work.";
 348
 349 static const char* wb[] = {wb0,wb1,wb2,wb3,wb4};
 350 void SampleWordBreakUsage()
 351 {
 352    SamplePrintWordWithBreak();
 353    SampleFindWordBreakFromPosition(0,0); // This
 354    SampleFindWordBreakFromPosition(1,0); // This
 355    SampleFindWordBreakFromPosition(2,0); // This
 356    SampleFindWordBreakFromPosition(2,1); // This
 357    SampleFindWordBreakFromPosition(2,9); // [space]
 358    SampleFindWordBreakFromPosition(2,10); // internationalization
 359    SampleFindWordBreakFromPosition(3,4);  // internationalization
 360    SampleFindWordBreakFromPosition(3,8);  // internationalization
 361    SampleFindWordBreakFromPosition(4,6);  // [space]
 362    SampleFindWordBreakFromPosition(4,7);  // work
 363 }
 364
 365
 366 void SamplePrintWordWithBreak()
 367 {
 368    PRUint32 numOfFragment = sizeof(wb) / sizeof(char*);
 369    nsIWordBreaker *wbk = NULL;
 370
 371    CallGetService(kWBrkCID, &wbk);
 372
 373    nsAutoString result;
 374
 375    for(PRUint32 i = 0; i < numOfFragment; i++)
 376    {
 377       NS_ConvertASCIItoUTF16 fragText(wb[i]);
 378
 379       PRInt32 cur = 0;
 380       cur = wbk->NextWord(fragText.get(), fragText.Length(), cur);
 381       PRUint32 start = 0;
 382       for(PRUint32 j = 0; cur != NS_WORDBREAKER_NEED_MORE_TEXT ; j++)
 383       {
 384             result.Append(Substring(fragText, start, cur - start));
 385             result.Append('^');
 386             start = (cur >= 0 ? cur : cur - start);
 387             cur = wbk->NextWord(fragText.get(), fragText.Length(), cur);
 388       }
 389
 390       result.Append(Substring(fragText, fragText.Length() - start));
 391
 392       if( i != (numOfFragment -1 ))
 393       {
 394         NS_ConvertASCIItoUTF16 nextFragText(wb[i+1]);
 395
 396         PRBool canBreak = PR_TRUE;
 397         canBreak = wbk->BreakInBetween( fragText.get(),
 398                                         fragText.Length(),
 399                                         nextFragText.get(),
 400                                         nextFragText.Length());
 401         if(canBreak)
 402             result.Append('^');
 403
 404         fragText.Assign(nextFragText);
 405       }
 406    }
 407    printf("Output From  SamplePrintWordWithBreak() \n\n");
 408    printf("[%s]\n", NS_ConvertUTF16toUTF8(result).get());
 409
 410    NS_IF_RELEASE(wbk);
 411 }
 412
 413 void SampleFindWordBreakFromPosition(PRUint32 fragN, PRUint32 offset)
 414 {
 415    PRUint32 numOfFragment = sizeof(wb) / sizeof(char*);
 416    nsIWordBreaker *wbk = NULL;
 417
 418    CallGetService(kWBrkCID, &wbk);
 419
 420    NS_ConvertASCIItoUTF16 fragText(wb[fragN]);
 421
 422    nsWordRange res = wbk->FindWord(fragText.get(), fragText.Length(), offset);
 423
 424    PRBool canBreak;
 425    nsAutoString result(Substring(fragText, res.mBegin, res.mEnd-res.mBegin));
 426
 427    if((PRUint32)fragText.Length() == res.mEnd) // if we hit the end of the fragment
 428    {
 429      nsAutoString curFragText = fragText;
 430      for(PRUint32  p = fragN +1; p < numOfFragment ;p++)
 431      {
 432         NS_ConvertASCIItoUTF16 nextFragText(wb[p]);
 433         canBreak = wbk->BreakInBetween(curFragText.get(),
 434                                        curFragText.Length(),
 435                                        nextFragText.get(),
 436                                        nextFragText.Length());
 437         if(canBreak)
 438            break;
 439
 440         nsWordRange r = wbk->FindWord(nextFragText.get(), nextFragText.Length(),
 441                                       0);
 442
 443         result.Append(Substring(nextFragText, r.mBegin, r.mEnd - r.mBegin));
 444
 445         if((PRUint32)nextFragText.Length() != r.mEnd)
 446           break;
 447
 448         nextFragText.Assign(curFragText);
 449      }
 450    }
 451
 452    if(0 == res.mBegin) // if we hit the beginning of the fragment
 453    {
 454      nsAutoString curFragText = fragText;
 455      for(PRUint32  p = fragN ; p > 0 ;p--)
 456      {
 457         NS_ConvertASCIItoUTF16 prevFragText(wb[p-1]);
 458         canBreak = wbk->BreakInBetween(prevFragText.get(),
 459                                        prevFragText.Length(),
 460                                        curFragText.get(),
 461                                        curFragText.Length());
 462         if(canBreak)
 463            break;
 464
 465         nsWordRange r = wbk->FindWord(prevFragText.get(), prevFragText.Length(),
 466                                       prevFragText.Length());
 467
 468         result.Insert(Substring(prevFragText, r.mBegin, r.mEnd - r.mBegin), 0);
 469
 470         if(0 != r.mBegin)
 471           break;
 472
 473         prevFragText.Assign(curFragText);
 474      }
 475    }
 476
 477    printf("Output From  SamplePrintWordWithBreak() \n\n");
 478    printf("[%s]\n", NS_ConvertUTF16toUTF8(result).get());
 479
 480    NS_IF_RELEASE(wbk);
 481 }
 482
 483 // Main
 484
 485 int main(int argc, char** argv) {
 486
 487    NS_InitXPCOM2(nsnull, nsnull, nsnull);
 488
 489    // --------------------------------------------
 490    printf("Test Line Break\n");
 491
 492    PRBool lbok ;
 493    PRBool wbok ;
 494    lbok =TestWordBreaker();
 495    if(lbok)
 496       printf("Line Break Test\nOK\n");
 497    else
 498       printf("Line Break Test\nFailed\n");
 499
 500    wbok = TestLineBreaker();
 501    if(wbok)
 502       printf("Word Break Test\nOK\n");
 503    else
 504       printf("Word Break Test\nFailed\n");
 505
 506    SampleWordBreakUsage();
 507
 508
 509    // --------------------------------------------
 510    printf("Finish All The Test Cases\n");
 511
 512    if(lbok && wbok)
 513       printf("Line/Word Break Test\nOK\n");
 514    else
 515       printf("Line/Word Break Test\nFailed\n");
 516    return 0;
 517 }