Python 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. Chilkat2-Python

import sys
import chilkat2

success = False

# 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

pdf = chilkat2.Pdf()
success = pdf.LoadFile("qa_data/pdf/invoice_labels.pdf")
if (success == False):
    print(pdf.LastErrorText)
    sys.exit()

json = chilkat2.JsonObject()
jsonOut = chilkat2.JsonObject()

# 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.
json.UpdateString("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.)
json.UpdateString("labelValue","right")

success = pdf.ExtractText("text",json,jsonOut)
if (success == False):
    print(pdf.LastErrorText)
    sys.exit()

# The value is returned as the text.
print("Invoice Number: " + jsonOut.StringOf("text"))

# The labelMatches array tells where the label was found and how it appeared in the PDF.
print("Found on page " + str(jsonOut.IntOf("labelMatches[0].page")) + " as \"" + jsonOut.StringOf("labelMatches[0].label") + "\"")

# Matching ignores case and the trailing colon, so these all work:
json.UpdateString("label","CUSTOMER:")
success = pdf.ExtractText("text",json,jsonOut)
if (success == False):
    print(pdf.LastErrorText)
    sys.exit()

print("Customer: " + jsonOut.StringOf("text"))

json.UpdateString("label","due date")
success = pdf.ExtractText("text",json,jsonOut)
if (success == False):
    print(pdf.LastErrorText)
    sys.exit()

print("Due Date: " + jsonOut.StringOf("text"))

# The totals at the foot of the invoice are right-aligned labels with the amount beside them.
json.UpdateString("label","Total Due")
success = pdf.ExtractText("text",json,jsonOut)
if (success == False):
    print(pdf.LastErrorText)
    sys.exit()

print("Total Due: " + jsonOut.StringOf("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".
json.UpdateString("label","Tracking Number")
success = pdf.ExtractText("text",json,jsonOut)
if (success == False):
    print(pdf.LastErrorText)
    sys.exit()

if (jsonOut.SizeOfArray("labelMatches") == 0):
    print("Tracking Number: (not found) " + jsonOut.StringOf("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