C++ UnicodeSet::toPattern方法代码示例

//---------------------------------------------------------------------
//
//   dump    Output the compiled form of the pattern.
//           Debugging function only.
//
//---------------------------------------------------------------------
void   RegexPattern::dumpOp(int32_t index) const {
    (void)index;  // Suppress warnings in non-debug build.
#if defined(REGEX_DEBUG)
    static const char * const opNames[] = {URX_OPCODE_NAMES};
    int32_t op          = fCompiledPat->elementAti(index);
    int32_t val         = URX_VAL(op);
    int32_t type        = URX_TYPE(op);
    int32_t pinnedType  = type;
    if ((uint32_t)pinnedType >= sizeof(opNames)/sizeof(char *)) {
        pinnedType = 0;
    }

    printf("%4d   %08x    %-15s  ", index, op, opNames[pinnedType]);
    switch (type) {
    case URX_NOP:
    case URX_DOTANY:
    case URX_DOTANY_ALL:
    case URX_FAIL:
    case URX_CARET:
    case URX_DOLLAR:
    case URX_BACKSLASH_G:
    case URX_BACKSLASH_X:
    case URX_END:
    case URX_DOLLAR_M:
    case URX_CARET_M:
        // Types with no operand field of interest.
        break;

    case URX_RESERVED_OP:
    case URX_START_CAPTURE:
    case URX_END_CAPTURE:
    case URX_STATE_SAVE:
    case URX_JMP:
    case URX_JMP_SAV:
    case URX_JMP_SAV_X:
    case URX_BACKSLASH_B:
    case URX_BACKSLASH_BU:
    case URX_BACKSLASH_D:
    case URX_BACKSLASH_Z:
    case URX_STRING_LEN:
    case URX_CTR_INIT:
    case URX_CTR_INIT_NG:
    case URX_CTR_LOOP:
    case URX_CTR_LOOP_NG:
    case URX_RELOC_OPRND:
    case URX_STO_SP:
    case URX_LD_SP:
    case URX_BACKREF:
    case URX_STO_INP_LOC:
    case URX_JMPX:
    case URX_LA_START:
    case URX_LA_END:
    case URX_BACKREF_I:
    case URX_LB_START:
    case URX_LB_CONT:
    case URX_LB_END:
    case URX_LBN_CONT:
    case URX_LBN_END:
    case URX_LOOP_C:
    case URX_LOOP_DOT_I:
    case URX_BACKSLASH_H:
    case URX_BACKSLASH_R:
    case URX_BACKSLASH_V:
        // types with an integer operand field.
        printf("%d", val);
        break;

    case URX_ONECHAR:
    case URX_ONECHAR_I:
        printf("%c", val<256?val:'?');
        break;

    case URX_STRING:
    case URX_STRING_I:
        {
            int32_t lengthOp       = fCompiledPat->elementAti(index+1);
            U_ASSERT(URX_TYPE(lengthOp) == URX_STRING_LEN);
            int32_t length = URX_VAL(lengthOp);
            int32_t i;
            for (i=val; i<val+length; i++) {
                UChar c = fLiteralText[i];
                if (c < 32 || c >= 256) {c = '.';}
                printf("%c", c);
            }
        }
        break;

    case URX_SETREF:
    case URX_LOOP_SR_I:
        {
            UnicodeString s;
            UnicodeSet *set = (UnicodeSet *)fSets->elementAt(val);
            set->toPattern(s, TRUE);
            for (int32_t i=0; i<s.length(); i++) {
//.........这里部分代码省略.........

void CanonicalIteratorTest::TestBasic() {

    UErrorCode status = U_ZERO_ERROR;

    static const char * const testArray[][2] = {
        {"\\u00C5d\\u0307\\u0327", "A\\u030Ad\\u0307\\u0327, A\\u030Ad\\u0327\\u0307, A\\u030A\\u1E0B\\u0327, "
            "A\\u030A\\u1E11\\u0307, \\u00C5d\\u0307\\u0327, \\u00C5d\\u0327\\u0307, "
            "\\u00C5\\u1E0B\\u0327, \\u00C5\\u1E11\\u0307, \\u212Bd\\u0307\\u0327, "
            "\\u212Bd\\u0327\\u0307, \\u212B\\u1E0B\\u0327, \\u212B\\u1E11\\u0307"},
        {"\\u010d\\u017E", "c\\u030Cz\\u030C, c\\u030C\\u017E, \\u010Dz\\u030C, \\u010D\\u017E"},
        {"x\\u0307\\u0327", "x\\u0307\\u0327, x\\u0327\\u0307, \\u1E8B\\u0327"},
    };
    
#if 0
    // This is not interesting for C/C++ as the data is already built beforehand
    // check build
    UnicodeSet ss = CanonicalIterator.getSafeStart();
    logln("Safe Start: " + ss.toPattern(true));
    ss = CanonicalIterator.getStarts('a');
    expectEqual("Characters with 'a' at the start of their decomposition: ", "", CanonicalIterator.getStarts('a'),
        new UnicodeSet("[\u00E0-\u00E5\u0101\u0103\u0105\u01CE\u01DF\u01E1\u01FB"
        + "\u0201\u0203\u0227\u1E01\u1EA1\u1EA3\u1EA5\u1EA7\u1EA9\u1EAB\u1EAD\u1EAF\u1EB1\u1EB3\u1EB5\u1EB7]")
            );
#endif

    // check permute
    // NOTE: we use a TreeSet below to sort the output, which is not guaranteed to be sorted!

    Hashtable *permutations = new Hashtable(FALSE, status);
    permutations->setValueDeleter(uhash_deleteUnicodeString);
    UnicodeString toPermute("ABC");

    CanonicalIterator::permute(toPermute, FALSE, permutations, status);

    logln("testing permutation");
  
    expectEqual("Simple permutation ", "", collectionToString(permutations), "ABC, ACB, BAC, BCA, CAB, CBA");

    delete permutations;
    
    // try samples
    logln("testing samples");
    Hashtable *set = new Hashtable(FALSE, status);
    set->setValueDeleter(uhash_deleteUnicodeString);
    int32_t i = 0;
    CanonicalIterator it("", status);
    if(U_SUCCESS(status)) {
      for (i = 0; i < ARRAY_LENGTH(testArray); ++i) {
          //logln("Results for: " + name.transliterate(testArray[i]));
          UnicodeString testStr = CharsToUnicodeString(testArray[i][0]);
          it.setSource(testStr, status);
          set->removeAll();
          for (;;) {
              //UnicodeString *result = new UnicodeString(it.next());
              UnicodeString result(it.next());
              if (result.isBogus()) {
                  break;
              }
              set->put(result, new UnicodeString(result), status); // Add result to the table
              //logln(++counter + ": " + hex.transliterate(result));
              //logln(" = " + name.transliterate(result));
          }
          expectEqual(i + ": ", testStr, collectionToString(set), CharsToUnicodeString(testArray[i][1]));

      }
    } else {
      errln("Couldn't instantiate canonical iterator. Error: %s", u_errorName(status));
    }
    delete set;
}