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 <CkPdfW.h>
#include <CkJsonObjectW.h>

void ChilkatSample(void)
    {
    bool 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

    CkPdfW pdf;
    success = pdf.LoadFile(L"qa_data/pdf/three_receipts.pdf");
    if (success == false) {
        wprintf(L"%s\n",pdf.lastErrorText());
        return;
    }

    CkJsonObjectW json;
    CkJsonObjectW jsonOut;

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

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

    wprintf(L"First total: %s\n",jsonOut.stringOf(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.
    json.UpdateString(L"labelOccurrence",L"all");
    success = pdf.ExtractText(L"text",json,jsonOut);
    if (success == false) {
        wprintf(L"%s\n",pdf.lastErrorText());
        return;
    }

    int i = 0;
    int count_i = jsonOut.SizeOfArray(L"labelMatches");
    while (i < count_i) {
        jsonOut.put_I(i);
        wprintf(L"Page %d total: %s\n",jsonOut.IntOf(L"labelMatches[i].page"),jsonOut.stringOf(L"labelMatches[i].text"));
        i = i + 1;
    }

    //  A number selects one occurrence: 2 is the second.
    json.UpdateInt(L"labelOccurrence",2);
    success = pdf.ExtractText(L"text",json,jsonOut);
    if (success == false) {
        wprintf(L"%s\n",pdf.lastErrorText());
        return;
    }

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

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

    wprintf(L"Receipt number on page 3: %s\n",jsonOut.stringOf(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
    }