C++ Requires Chilkat v11.7.0+

Expand Ligatures and Normalize Unicode in Extracted PDF Text

See more PDF Text Extraction Examples

Demonstrates the "expandLigatures" and "normalize" clean-up options. Typeset PDFs use single glyphs for "fi" and "fl"; by default they are expanded to their letters. Unicode normalization is NFC by default; NFKC also folds compatibility characters such as circled digits, the trademark sign and vulgar fractions into plain characters, and "none" leaves the text as the PDF has it. C++

#include <CkPdf.h>
#include <CkJsonObject.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 is available at ligatures.pdf

    CkPdf pdf;
    success = pdf.LoadFile("qa_data/pdf/ligatures.pdf");
    if (success == false) {
        std::cout << pdf.lastErrorText() << "\r\n";
        return;
    }

    CkJsonObject json;
    CkJsonObject jsonOut;

    //  The first line is set with the fi and fl ligature glyphs; the second has circled digits,
    //  a trademark sign and a vulgar fraction.
    success = pdf.ExtractText("text",json,jsonOut);
    if (success == false) {
        std::cout << pdf.lastErrorText() << "\r\n";
        return;
    }

    std::cout << "---- default (expandLigatures: true, normalize: NFC) ----" << "\r\n";
    std::cout << jsonOut.stringOf("text") << "\r\n";

    //  Keep the ligature characters as they are in the PDF.
    json.UpdateBool("expandLigatures",false);
    json.UpdateString("normalize","none");
    success = pdf.ExtractText("text",json,jsonOut);
    if (success == false) {
        std::cout << pdf.lastErrorText() << "\r\n";
        return;
    }

    std::cout << "---- expandLigatures: false, normalize: none ----" << "\r\n";
    std::cout << jsonOut.stringOf("text") << "\r\n";

    //  NFKC folds the compatibility characters too (and expands ligatures on its own).
    json.UpdateString("normalize","NFKC");
    success = pdf.ExtractText("text",json,jsonOut);
    if (success == false) {
        std::cout << pdf.lastErrorText() << "\r\n";
        return;
    }

    std::cout << "---- normalize: NFKC ----" << "\r\n";
    std::cout << jsonOut.stringOf("text") << "\r\n";

    //  Sample output (shown here in ASCII: (fi) and (fl) stand for the single ligature characters, (1) and (2) for the circled digits,
    //  (TM) for the trademark sign and 1/2 for the vulgar fraction; the NFKC output has plain "1", "2", "TM" and "1", fraction slash, "2"):

    //  ---- default (expandLigatures: true, normalize: NFC) ----
    //  Typeset Text
    //  
    //  The final flight of the season was officially confirmed on the fifth. Rooms (1) and (2) are reserved; the Harbor Cafe(TM) serves 1/2 portions.
    //  ---- expandLigatures: false, normalize: none ----
    //  Typeset Text
    //  
    //  The (fi)nal (fl)ight of the season was of(fi)cially con(fi)rmed on the (fi)fth. Rooms (1) and (2) are reserved; the Harbor Cafe(TM) serves 1/2 portions.
    //  ---- normalize: NFKC ----
    //  Typeset Text
    //  
    //  The final flight of the season was officially confirmed on the fifth. Rooms 1 and 2 are reserved; the Harbor CafeTM serves 1/2 portions.
    }