Tcl 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. 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 PDF used in this example (three receipts, one per page) is available at three_receipts.pdf

set pdf [new_CkPdf]

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

set json [new_CkJsonObject]

set jsonOut [new_CkJsonObject]

#  Each of the three pages is a receipt with a "Total:" line.
CkJsonObject_UpdateString $json "label" "Total"
CkJsonObject_UpdateString $json "labelValue" "right"

#  By default only the first occurrence (in reading order) is used.
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 "First total: [CkJsonObject_stringOf $jsonOut 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.
CkJsonObject_UpdateString $json "labelOccurrence" "all"
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
}

set i 0
set count_i [CkJsonObject_SizeOfArray $jsonOut "labelMatches"]
while {$i < $count_i} {
    CkJsonObject_put_I $jsonOut $i
    puts "Page [CkJsonObject_IntOf $jsonOut {labelMatches[i].page}] total: [CkJsonObject_stringOf $jsonOut {labelMatches[i].text}]"
    set i [expr $i + 1]
}

#  A number selects one occurrence: 2 is the second.
CkJsonObject_UpdateInt $json "labelOccurrence" 2
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 "Second total: [CkJsonObject_stringOf $jsonOut text]"

#  The occurrences are counted over the selected pages, so combining "pages" with the label
#  is another way to pick one receipt:
CkJsonObject_UpdateString $json "labelOccurrence" "first"
CkJsonObject_UpdateString $json "pages" "3"
CkJsonObject_UpdateString $json "label" "Receipt No."
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 "Receipt number on page 3: [CkJsonObject_stringOf $jsonOut 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

delete_CkPdf $pdf
delete_CkJsonObject $json
delete_CkJsonObject $jsonOut