Xbase++ 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. Xbase++

LOCAL nSuccess
LOCAL oPdf
LOCAL oJson
LOCAL oJsonOut
LOCAL i
LOCAL nCount_i

nSuccess := 0

//  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

oPdf := CreateObject("Chilkat.Pdf")
nSuccess := oPdf:LoadFile("qa_data/pdf/three_receipts.pdf")
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    RETURN
ENDIF

oJson := CreateObject("Chilkat.JsonObject")
oJsonOut := CreateObject("Chilkat.JsonObject")

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

//  By default only the first occurrence (in reading order) is used.
nSuccess := oPdf:ExtractText("text", oJson, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "First total: " + oJsonOut:StringOf("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.
oJson:UpdateString("labelOccurrence", "all")
nSuccess := oPdf:ExtractText("text", oJson, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

i := 0
nCount_i := oJsonOut:SizeOfArray("labelMatches")
DO WHILE i < nCount_i
    oJsonOut:I := i
    ? "Page " + Str(oJsonOut:IntOf("labelMatches[i].page")) + " total: " + oJsonOut:StringOf("labelMatches[i].text")
    i := i + 1
ENDDO

//  A number selects one occurrence: 2 is the second.
oJson:UpdateInt("labelOccurrence", 2)
nSuccess := oPdf:ExtractText("text", oJson, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "Second total: " + oJsonOut:StringOf("text")

//  The occurrences are counted over the selected pages, so combining "pages" with the label
//  is another way to pick one receipt:
oJson:UpdateString("labelOccurrence", "first")
oJson:UpdateString("pages", "3")
oJson:UpdateString("label", "Receipt No.")
nSuccess := oPdf:ExtractText("text", oJson, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "Receipt number on page 3: " + oJsonOut:StringOf("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

oPdf:destroy()
oJson:destroy()
oJsonOut:destroy()