Perl 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. Perl

use chilkat();

$success = 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

$pdf = chilkat::CkPdf->new();
$success = $pdf->LoadFile("qa_data/pdf/hidden_text.pdf");
if ($success == 0) {
    print $pdf->lastErrorText() . "\r\n";
    exit;
}

$json = chilkat::CkJsonObject->new();
$jsonOut = chilkat::CkJsonObject->new();

#  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.
$success = $pdf->ExtractText("text",$json,$jsonOut);
if ($success == 0) {
    print $pdf->lastErrorText() . "\r\n";
    exit;
}

print "---- default ----" . "\r\n";
print $jsonOut->stringOf("text") . "\r\n";

#  The quality report says what was left out.
$i = 0;
$count_i = $jsonOut->SizeOfArray("quality.pages[0].issues");
while ($i < $count_i) {
    $jsonOut->put_I($i);
    print "issue: " . $jsonOut->stringOf("quality.pages[0].issues[i].code") . "\r\n";
    $i = $i + 1;
}

#  Include everything.
$json->UpdateString("clippedText","include");
$json->UpdateString("offPageText","include");
$json->UpdateString("obscuredText","include");
$success = $pdf->ExtractText("text",$json,$jsonOut);
if ($success == 0) {
    print $pdf->lastErrorText() . "\r\n";
    exit;
}

print "---- clippedText, offPageText, obscuredText: include ----" . "\r\n";
print $jsonOut->stringOf("text") . "\r\n";

#  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.