Xbase++ Requires Chilkat v11.7.0+

Clipped, Off-Page and White-on-White Text in PDF Extraction

See more PDF Text Extraction Examples

Demonstrates the "clippedText", "offPageText" and "obscuredText" filters. Text that a reader never sees - hidden by a clipping path, outside the crop box, or drawn in the same color as the box beneath it - is excluded by default, and the quality report notes what was skipped. Each kind can be included. 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 is available at hidden_text.pdf

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

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

//  The page has a visible paragraph plus three hidden lines: one drawn outside its clipping
//  rectangle, one below the crop box, and one in white on a white box.
nSuccess := oPdf:ExtractText("text", oJson, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "---- default ----"
? oJsonOut:StringOf("text")

//  The quality report says what was left out.
i := 0
nCount_i := oJsonOut:SizeOfArray("quality.pages[0].issues")
DO WHILE i < nCount_i
    oJsonOut:I := i
    ? "issue: " + oJsonOut:StringOf("quality.pages[0].issues[i].code")
    i := i + 1
ENDDO

//  Include everything.
oJson:UpdateString("clippedText", "include")
oJson:UpdateString("offPageText", "include")
oJson:UpdateString("obscuredText", "include")
nSuccess := oPdf:ExtractText("text", oJson, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "---- clippedText, offPageText, obscuredText: include ----"
? oJsonOut:StringOf("text")

//  Sample output:

//  ---- default ----
//  Visible Text
//  
//  This paragraph is visible. The page also carries three kinds of text a reader never sees.
//  
//  End of the visible text.
//  issue: obscuredTextSkipped
//  issue: clippedTextSkipped
//  ---- clippedText, offPageText, obscuredText: include ----
//  Visible Text
//  
//  This paragraph is visible. The page also carries three kinds of text a reader never sees.
//  
//  CLIPPED: this line is drawn outside its clipping rectangle.
//  
//  OBSCURED: this line is white on a white box.
//  
//  End of the visible text.
//  
//  OFFPAGE: this line sits below the crop box.

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