Unicode C

Regular Expression Replace Capture Groups

See more Regular Expressions Examples

Demonstrates replacing capture groups for a regular expression.

Note: Chilkat uses PCRE2. See PCRE2 Regular Expressions
Also see: PCRE2 Performance

Unicode C

#include <C_CkStringBuilderW.h>
#include <C_CkJsonObjectW.h>

void ChilkatSample(void)
    {
    BOOL success;
    const wchar_t *subject;
    const wchar_t *pattern;
    HCkStringBuilderW sb;
    HCkJsonObjectW json;
    int timeoutMs;
    int numMatches;
    int firstNameIdx;
    int lastNameIdx;
    HCkStringBuilderW sbTemp;
    int i;

    success = FALSE;

    subject = L"John Anders, +_+_+ Mary Robins $$$$";
    pattern = L"(?<first>\\w+)\\s+(?<last>\\w+)";

    sb = CkStringBuilderW_Create();
    CkStringBuilderW_Append(sb,subject);

    json = CkJsonObjectW_Create();
    CkJsonObjectW_putEmitCompact(json,FALSE);

    timeoutMs = 2000;
    numMatches = CkStringBuilderW_RegexMatch(sb,pattern,json,timeoutMs);
    if (numMatches < 0) {
        // Probably an error in the regular expression.
        // Suggestion: Use AI to help create and/or diagnose regular expressions.
        wprintf(L"%s\n",CkStringBuilderW_lastErrorText(sb));
        CkStringBuilderW_Dispose(sb);
        CkJsonObjectW_Dispose(json);
        return;
    }

    // Examine the matches:
    wprintf(L"%s\n",CkJsonObjectW_emit(json));

    // There are 2 matches:

    // {
    //   "named": {
    //     "first": 1,
    //     "last": 2
    //   },
    //   "match": [
    //     {
    //       "group": [
    //         {
    //           "cap": "John Anders",
    //           "idx": 0,
    //           "len": 11
    //         },
    //         {
    //           "cap": "John",
    //           "idx": 0,
    //           "len": 4
    //         },
    //         {
    //           "cap": "Anders",
    //           "idx": 5,
    //           "len": 6
    //         }
    //       ]
    //     },
    //     {
    //       "group": [
    //         {
    //           "cap": "Mary Robins",
    //           "idx": 19,
    //           "len": 11
    //         },
    //         {
    //           "cap": "Mary",
    //           "idx": 19,
    //           "len": 4
    //         },
    //         {
    //           "cap": "Robins",
    //           "idx": 24,
    //           "len": 6
    //         }
    //       ]
    //     }
    //   ]
    // }

    // To replace capture groups, write code to examine each capture group within
    // each match, and provide a replacement string.
    // Then call RegexReplace.

    // For example, let's capitalize the first names, and add append "on" to the last name.
    // After doing replacements, we should get:  JOHN Anderson, +_+_+ MARY Robinson $$$$

    firstNameIdx = CkJsonObjectW_IntOf(json,L"named.first");
    lastNameIdx = CkJsonObjectW_IntOf(json,L"named.last");

    sbTemp = CkStringBuilderW_Create();

    i = 0;
    numMatches = CkJsonObjectW_SizeOfArray(json,L"match");
    while (i < numMatches) {

        CkJsonObjectW_putI(json,i);

        // The replacement string for the first name will be all uppercase.
        CkJsonObjectW_putJ(json,firstNameIdx);
        CkStringBuilderW_Clear(sbTemp);
        CkJsonObjectW_StringOfSb(json,L"match[i].group[j].cap",sbTemp);
        CkStringBuilderW_ToUppercase(sbTemp);
        CkJsonObjectW_UpdateSb(json,L"match[i].group[j].rep",sbTemp);

        // Append "on" to the last name.
        CkJsonObjectW_putJ(json,lastNameIdx);
        CkStringBuilderW_Clear(sbTemp);
        CkJsonObjectW_StringOfSb(json,L"match[i].group[j].cap",sbTemp);
        CkStringBuilderW_Append(sbTemp,L"on");
        CkJsonObjectW_UpdateSb(json,L"match[i].group[j].rep",sbTemp);

        i = i + 1;
    }

    // The JSON now has replacement strings:
    wprintf(L"%s\n",CkJsonObjectW_emit(json));

    // {
    //   "named": {
    //     "first": 1,
    //     "last": 2
    //   },
    //   "match": [
    //     {
    //       "group": [
    //         {
    //           "cap": "John Anders",
    //           "idx": 0,
    //           "len": 11
    //         },
    //         {
    //           "cap": "John",
    //           "idx": 0,
    //           "len": 4,
    //           "rep": "JOHN"
    //         },
    //         {
    //           "cap": "Anders",
    //           "idx": 5,
    //           "len": 6,
    //           "rep": "Anderson"
    //         }
    //       ]
    //     },
    //     {
    //       "group": [
    //         {
    //           "cap": "Mary Robins",
    //           "idx": 19,
    //           "len": 11
    //         },
    //         {
    //           "cap": "Mary",
    //           "idx": 19,
    //           "len": 4,
    //           "rep": "MARY"
    //         },
    //         {
    //           "cap": "Robins",
    //           "idx": 24,
    //           "len": 6,
    //           "rep": "Robinson"
    //         }
    //       ]
    //     }
    //   ]
    // }

    // Call RegexReplace to update the StringBuilder with the replacements.
    success = CkStringBuilderW_RegexReplace(sb,json);
    if (success == FALSE) {
        wprintf(L"%s\n",CkStringBuilderW_lastErrorText(sb));
        CkStringBuilderW_Dispose(sb);
        CkJsonObjectW_Dispose(json);
        CkStringBuilderW_Dispose(sbTemp);
        return;
    }

    wprintf(L"Result after doing replacements:\n");
    wprintf(L"%s\n",CkStringBuilderW_getAsString(sb));

    // Result after doing replacements:
    // JOHN Anderson, +_+_+ MARY Robinson $$$$


    CkStringBuilderW_Dispose(sb);
    CkJsonObjectW_Dispose(json);
    CkStringBuilderW_Dispose(sbTemp);

    }