Python Requires Chilkat v11.7.0+

Extract Tables from a PDF

See more PDF Text Extraction Examples

Demonstrates the "tables" operation, which returns the tables detected on the pages as rows and cells. Each cell has its row and column (counted from 0), its row and column span, and its text. 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 PDF used in this example is available at price_list.pdf

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

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

# The "tables" operation returns the tables found on the selected pages.
# Tables are recognized by their drawn cell borders.
success = pdf.ExtractText("tables",json,jsonOut)
if (success == False):
    print(pdf.LastErrorText)
    sys.exit()

print("Number of tables: " + str(jsonOut.IntOf("count")))

# Used to build each row of output.
sbRow = chilkat2.StringBuilder()

i = 0
count_i = jsonOut.SizeOfArray("tables")
while i < count_i :
    jsonOut.I = i
    print("Table on page " + str(jsonOut.IntOf("tables[i].page")) + ": " + str(jsonOut.IntOf("tables[i].rows")) + " rows x " + str(jsonOut.IntOf("tables[i].columns"))\
         + " columns, detected by " + jsonOut.StringOf("tables[i].detectedBy"))

    # The cells are listed row by row.  Print each row as its cells separated by " | ".
    currentRow = -1
    sbRow.SetString("")
    j = 0
    count_j = jsonOut.SizeOfArray("tables[i].cells")
    while j < count_j :
        jsonOut.J = j
        row = jsonOut.IntOf("tables[i].cells[j].row")
        if (row != currentRow):
            if (currentRow >= 0):
                print(sbRow.GetAsString())

            currentRow = row
            sbRow.SetString("  ")
        else:
            sbRow.Append(" | ")

        sbRow.Append(jsonOut.StringOf("tables[i].cells[j].text"))
        j = j + 1

    print(sbRow.GetAsString())
    i = i + 1

# The cells of the first table as JSON:
jsonOut.EmitCompact = False
print(jsonOut.Emit())

# Sample output:

# Number of tables: 1
# Table on page 1: 8 rows x 5 columns, detected by ruled
#   Code | Product | Unit | Price | In Stock
#   HT-101 | Claw hammer, 16 oz | each | $18.00 | 24
#   HT-102 | Adjustable wrench, 10 in | each | $14.50 | 31
#   HT-103 | Screwdriver set, 6 piece | set | $22.00 | 12
#   HT-104 | Tape measure, 25 ft | each | $9.75 | 40
#   HT-105 | Utility knife | each | $6.20 | 58
#   HT-106 | Pliers, needle nose | each | $11.00 | 19
#   HT-107 | Hand saw, 20 in | each | $21.40 | 8
# {
#   "count": 1,
#   "tables": [
#     {
#       "page": 1,
#       "detectedBy": "ruled",
#       "rows": 8,
#       "columns": 5,
#       "cells": [
#         {
#           "row": 0,
#           "col": 0,
#           "rowSpan": 1,
#           "colSpan": 1,
#           "text": "Code"
#         },
#         {
#           "row": 0,
#           "col": 1,
#           "rowSpan": 1,
#           "colSpan": 1,
#           "text": "Product"
#         },
#         ...
#       ]
#     }
#   ],
#   "quality": {
#     "rating": "good",
#     ...
#   }
# }