Swift Requires Chilkat v11.7.0+

Get the Text Block Below a PDF Caption

See more PDF Text Extraction Examples

Demonstrates the "label" option with labelValue "below", which captures the block of text directly under a caption such as "Ship To", ending at the next gap or heading. Combined with layout "lines", the lines of an address block are kept separate. Swift

func chilkatTest() {
    var success: Bool = false

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

    //  The packing slip used in this example is available at packing_slip.pdf

    let pdf = CkoPdf()!
    success = pdf.loadFile(filePath: "qa_data/pdf/packing_slip.pdf")
    if success == false {
        print("\(pdf.lastErrorText!)")
        return
    }

    let json = CkoJsonObject()!
    let jsonOut = CkoJsonObject()!

    //  The packing slip has the captions "Bill To" and "Ship To" side by side,
    //  each with an address block below it.
    json.updateString(jsonPath: "label", value: "Ship To")

    //  "below" means the value is the block of text directly under the label.
    json.updateString(jsonPath: "labelValue", value: "below")

    //  With the default "flow" layout the lines of the block would be joined into one paragraph.
    //  "lines" keeps each line of the address on its own line.
    json.updateString(jsonPath: "layout", value: "lines")

    success = pdf.extractText(op: "text", jsonIn: json, jsonOut: jsonOut)
    if success == false {
        print("\(pdf.lastErrorText!)")
        return
    }

    print("Ship To:")
    print("\(jsonOut.string(of: "text")!)")
    print("")

    json.updateString(jsonPath: "label", value: "Bill To")
    success = pdf.extractText(op: "text", jsonIn: json, jsonOut: jsonOut)
    if success == false {
        print("\(pdf.lastErrorText!)")
        return
    }

    print("Bill To:")
    print("\(jsonOut.string(of: "text")!)")
    print("")

    //  A caption over a short paragraph works the same way.  Here the flow layout is better,
    //  because the two lines are one piece of text.
    json.updateString(jsonPath: "label", value: "Special Instructions")
    json.updateString(jsonPath: "layout", value: "flow")
    success = pdf.extractText(op: "text", jsonIn: json, jsonOut: jsonOut)
    if success == false {
        print("\(pdf.lastErrorText!)")
        return
    }

    print("Special Instructions:")
    print("\(jsonOut.string(of: "text")!)")

    //  Sample output:

    //  Ship To:
    //  Frank Delgado
    //  c/o Harbor Pet Boarding
    //  9 Ferry Lane
    //  Portside, WA 98001
    //  
    //  Bill To:
    //  Frank Delgado
    //  Apartment 3B
    //  41 Mill Street
    //  Portside, WA 98000
    //  
    //  Special Instructions:
    //  Leave the carton with the boarding office if no one answers. Do not leave it outside; the cat tree is not packed for rain.

}