Unicode C Requires Chilkat v11.7.0+

Get Every Occurrence of a Label in a PDF

See more PDF Text Extraction Examples

Demonstrates the "labelOccurrence" option. When the same label appears more than once (for example "Total:" on each of several receipts), "first" (the default) uses the first occurrence, "all" returns every occurrence with its page, and a number selects one occurrence. Unicode C

#include <C_CkPdfW.h>
#include <C_CkJsonObjectW.h>

void ChilkatSample(void)
    {
    BOOL success;
    HCkPdfW pdf;
    HCkJsonObjectW json;
    HCkJsonObjectW jsonOut;
    int i;
    int count_i;

    success = FALSE;

    //  This example requires the Chilkat API to have been previously unlocked.
    //  See Global Unlock Sample for sample code.

    //  The PDF used in this example (three receipts, one per page) is available at three_receipts.pdf

    pdf = CkPdfW_Create();
    success = CkPdfW_LoadFile(pdf,L"qa_data/pdf/three_receipts.pdf");
    if (success == FALSE) {
        wprintf(L"%s\n",CkPdfW_lastErrorText(pdf));
        CkPdfW_Dispose(pdf);
        return;
    }

    json = CkJsonObjectW_Create();
    jsonOut = CkJsonObjectW_Create();

    //  Each of the three pages is a receipt with a "Total:" line.
    CkJsonObjectW_UpdateString(json,L"label",L"Total");
    CkJsonObjectW_UpdateString(json,L"labelValue",L"right");

    //  By default only the first occurrence (in reading order) is used.
    success = CkPdfW_ExtractText(pdf,L"text",json,jsonOut);
    if (success == FALSE) {
        wprintf(L"%s\n",CkPdfW_lastErrorText(pdf));
        CkPdfW_Dispose(pdf);
        CkJsonObjectW_Dispose(json);
        CkJsonObjectW_Dispose(jsonOut);
        return;
    }

    wprintf(L"First total: %s\n",CkJsonObjectW_stringOf(jsonOut,L"text"));

    //  labelOccurrence "all" returns every occurrence.  The "text" member holds the values joined
    //  by line breaks, and labelMatches has one entry per occurrence with its page.
    CkJsonObjectW_UpdateString(json,L"labelOccurrence",L"all");
    success = CkPdfW_ExtractText(pdf,L"text",json,jsonOut);
    if (success == FALSE) {
        wprintf(L"%s\n",CkPdfW_lastErrorText(pdf));
        CkPdfW_Dispose(pdf);
        CkJsonObjectW_Dispose(json);
        CkJsonObjectW_Dispose(jsonOut);
        return;
    }

    i = 0;
    count_i = CkJsonObjectW_SizeOfArray(jsonOut,L"labelMatches");
    while (i < count_i) {
        CkJsonObjectW_putI(jsonOut,i);
        wprintf(L"Page %d total: %s\n",CkJsonObjectW_IntOf(jsonOut,L"labelMatches[i].page"),CkJsonObjectW_stringOf(jsonOut,L"labelMatches[i].text"));
        i = i + 1;
    }

    //  A number selects one occurrence: 2 is the second.
    CkJsonObjectW_UpdateInt(json,L"labelOccurrence",2);
    success = CkPdfW_ExtractText(pdf,L"text",json,jsonOut);
    if (success == FALSE) {
        wprintf(L"%s\n",CkPdfW_lastErrorText(pdf));
        CkPdfW_Dispose(pdf);
        CkJsonObjectW_Dispose(json);
        CkJsonObjectW_Dispose(jsonOut);
        return;
    }

    wprintf(L"Second total: %s\n",CkJsonObjectW_stringOf(jsonOut,L"text"));

    //  The occurrences are counted over the selected pages, so combining "pages" with the label
    //  is another way to pick one receipt:
    CkJsonObjectW_UpdateString(json,L"labelOccurrence",L"first");
    CkJsonObjectW_UpdateString(json,L"pages",L"3");
    CkJsonObjectW_UpdateString(json,L"label",L"Receipt No.");
    success = CkPdfW_ExtractText(pdf,L"text",json,jsonOut);
    if (success == FALSE) {
        wprintf(L"%s\n",CkPdfW_lastErrorText(pdf));
        CkPdfW_Dispose(pdf);
        CkJsonObjectW_Dispose(json);
        CkJsonObjectW_Dispose(jsonOut);
        return;
    }

    wprintf(L"Receipt number on page 3: %s\n",CkJsonObjectW_stringOf(jsonOut,L"text"));

    //  Sample output:

    //  First total: $19.86
    //  Page 1 total: $19.86
    //  Page 2 total: $17.34
    //  Page 3 total: $23.62
    //  Second total: $17.34
    //  Receipt number on page 3: 1003


    CkPdfW_Dispose(pdf);
    CkJsonObjectW_Dispose(json);
    CkJsonObjectW_Dispose(jsonOut);

    }