Tcl Requires Chilkat v11.7.0+

Get a PDF Field Value by its Label (Value to the Right)

See more PDF Text Extraction Examples

Demonstrates the "label" option, which captures the text that belongs to a caption such as "Invoice Number:" without needing coordinates. With labelValue "right", the value is the rest of the label's line. Label matching ignores case, extra spaces and a trailing colon. Tcl

load ./chilkat.dll

set success 0

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

#  The invoice used in this example is available at invoice_labels.pdf

set pdf [new_CkPdf]

set success [CkPdf_LoadFile $pdf "qa_data/pdf/invoice_labels.pdf"]
if {$success == 0} then {
    puts [CkPdf_lastErrorText $pdf]
    delete_CkPdf $pdf
    exit
}

set json [new_CkJsonObject]

set jsonOut [new_CkJsonObject]

#  The invoice has lines such as "Invoice Number:   INV-2026-0412".
#  The label option restricts the "text" operation to the text belonging to the label.
CkJsonObject_UpdateString $json "label" "Invoice Number"

#  "right" means the value is the rest of the label's line.
#  (The default, "auto", uses the rest of the line if there is text there, otherwise the block below.)
CkJsonObject_UpdateString $json "labelValue" "right"

set success [CkPdf_ExtractText $pdf "text" $json $jsonOut]
if {$success == 0} then {
    puts [CkPdf_lastErrorText $pdf]
    delete_CkPdf $pdf
    delete_CkJsonObject $json
    delete_CkJsonObject $jsonOut
    exit
}

#  The value is returned as the text.
puts "Invoice Number: [CkJsonObject_stringOf $jsonOut text]"

#  The labelMatches array tells where the label was found and how it appeared in the PDF.
puts "Found on page [CkJsonObject_IntOf $jsonOut {labelMatches[0].page}] as \"[CkJsonObject_stringOf $jsonOut {labelMatches[0].label}]\""

#  Matching ignores case and the trailing colon, so these all work:
CkJsonObject_UpdateString $json "label" "CUSTOMER:"
set success [CkPdf_ExtractText $pdf "text" $json $jsonOut]
if {$success == 0} then {
    puts [CkPdf_lastErrorText $pdf]
    delete_CkPdf $pdf
    delete_CkJsonObject $json
    delete_CkJsonObject $jsonOut
    exit
}

puts "Customer: [CkJsonObject_stringOf $jsonOut text]"

CkJsonObject_UpdateString $json "label" "due date"
set success [CkPdf_ExtractText $pdf "text" $json $jsonOut]
if {$success == 0} then {
    puts [CkPdf_lastErrorText $pdf]
    delete_CkPdf $pdf
    delete_CkJsonObject $json
    delete_CkJsonObject $jsonOut
    exit
}

puts "Due Date: [CkJsonObject_stringOf $jsonOut text]"

#  The totals at the foot of the invoice are right-aligned labels with the amount beside them.
CkJsonObject_UpdateString $json "label" "Total Due"
set success [CkPdf_ExtractText $pdf "text" $json $jsonOut]
if {$success == 0} then {
    puts [CkPdf_lastErrorText $pdf]
    delete_CkPdf $pdf
    delete_CkJsonObject $json
    delete_CkJsonObject $jsonOut
    exit
}

puts "Total Due: [CkJsonObject_stringOf $jsonOut text]"

#  A label that is not found is not an error.  The text is empty, labelMatches is empty,
#  and the quality report has an issue with the code "labelNotFound".
CkJsonObject_UpdateString $json "label" "Tracking Number"
set success [CkPdf_ExtractText $pdf "text" $json $jsonOut]
if {$success == 0} then {
    puts [CkPdf_lastErrorText $pdf]
    delete_CkPdf $pdf
    delete_CkJsonObject $json
    delete_CkJsonObject $jsonOut
    exit
}

if {[CkJsonObject_SizeOfArray $jsonOut "labelMatches"] == 0} then {
    puts "Tracking Number: (not found) [CkJsonObject_stringOf $jsonOut {quality.pages[0].issues[0].code}]"
}

#  Sample output:

#  Invoice Number: INV-2026-0412
#  Found on page 1 as "Invoice Number:"
#  Customer: Northwind Traders
#  Due Date: April 13, 2026
#  Total Due: $1,234.53
#  Tracking Number: (not found) labelNotFound

delete_CkPdf $pdf
delete_CkJsonObject $json
delete_CkJsonObject $jsonOut