How to Parse PDFs in Python: A Step-by-Step Guide

parse-pdf-in-python.md

Learn How to Parse PDFs in Python: A Step-by-Step Guide

parse-pdf-in-python_extract_all_text.py

# This code example shows how to extract text from all pages of a PDF document in Python
import aspose.pdf as ap

# Open PDF document
document = ap.Document("AddText.pdf")

# Create text absorber
text_absorber = ap.text.TextAbsorber()

# Call the accept method to process all pages
document.pages.accept(text_absorber)

# Retrieve the extracted text
extracted_text = text_absorber.text

# Define the file path
file_path = "extracted-text.txt"

# Open the file in write mode and write the extracted text
with open(file_path, "w", encoding="utf-8") as tw:
    tw.write(extracted_text + "\n")  # Write the extracted text with a newline

parse-pdf-in-python_extract_image.py

# This code example shows how to extract images from a PDF in Python
import aspose.pdf as ap

# Open document
document = ap.Document("Sample.pdf")

# Extract a particular image (first image from the first page)
x_image = document.pages[1].resources.images[1]

# Define the output image path
output_image_path = "OutputImage.jpg"

# Save the extracted image
with open(output_image_path, "wb") as output_image:
    output_image.write(x_image.to_stream().read())

parse-pdf-in-python_extract_specific_page_text.py

# This code example shows how to extract text from a specific page of a PDF document in Python
import aspose.pdf as ap

# Open PDF document
document = ap.Document("AddText.pdf")

# Create text absorber
text_absorber = ap.text.TextAbsorber()

# Call the accept method to process all pages
document.pages[1].accept(text_absorber)

# Retrieve the extracted text
extracted_text = text_absorber.text

# Define the file path
file_path = "extracted-text.txt"

# Open the file in write mode and write the extracted text
with open(file_path, "w", encoding="utf-8") as tw:
    tw.write(extracted_text + "\n")  # Write the extracted text with a newline

parse-pdf-in-python_extract_table.py

# This code example shows how to extract tables from a PDF document in Python
import aspose.pdf as ap

# Load PDF file
document = pdf.Document("sample.pdf")

# Process all pages
for page in document.pages:
    # Initialize TableAbsorber object
    absorber = ap.text.TableAbsorber()
    # Identify tables on the current page
    absorber.visit(page)
    # Loop through extracted tables
    for table in absorber.table_list:
        # Iterate through all the rows in the table
        for row in table.row_list:
            # Iterate through all the columns in the row
            for cell in row.cell_list:
                # Fetch the text fragments
                text_fragment_collection = cell.text_fragments
                # Iterate through the text fragments
                for fragment in text_fragment_collection:
                    # Print the text
                    print(fragment.text)

parse-pdf-in-python_extract_text_from_multi_column.py

# This code example shows how to extract text from a multi-column PDF in Python
import io
import aspose.pdf as ap

# Open PDF document
document = ap.Document("multi-column-sample.pdf")

# Create TextFragmentAbsorber object to extract text
text_fragment_absorber = ap.text.TextFragmentAbsorber()

# Accept the absorber for the first page
document.pages.accept(text_fragment_absorber)

# Get the collection of extracted text fragments
text_fragment_collection = text_fragment_absorber.text_fragments

# Reduce font size by at least 70% to improve text extraction
for text_fragment in text_fragment_collection:
    text_fragment.text_state.font_size *= 0.7

# Save the modified document to an in-memory stream
source_stream = io.BytesIO()
document.save(source_stream)

# Reload the document from the memory stream
source_stream.seek(0)
dest_document = ap.Document(source_stream)

# Initialize TextAbsorber to extract the updated text
text_absorber = ap.text.TextAbsorber()
dest_document.pages.accept(text_absorber)
extracted_text = text_absorber.text

# Save the extracted text to a file
with open("ExtractColumnsText_out.txt", "w", encoding="utf-8") as file:
    file.write(extracted_text)

parse-pdf-in-python_extract_text_from_page_region.py

# This code example shows how to extract text from a specific region of a page in a PDF document using Python
import aspose.pdf as ap

# Open PDF document
document = ap.Document("sample.pdf")

# Create TextAbsorber object to extract text
absorber = ap.text.TextAbsorber()
absorber.text_search_options.limit_to_page_bounds = True
absorber.text_search_options.rectangle = ap.Rectangle(100, 200, 250, 350, True)

# Accept the absorber for the first page
document.pages[1].accept(absorber)

# Get the extracted text
extracted_text = absorber.text

# Define the file path
file_path = "extracted-text.txt"

# Open the file in write mode and write the extracted text
with open(file_path, "w", encoding="utf-8") as tw:
    tw.write(extracted_text + "\n")  # Write the extracted text with a newline

parse-pdf-in-python_extract_text_with_scale_factor.py

# This code example shows how to extract text from a specific region of a page in a PDF document using Python
import aspose.pdf as ap

# Open PDF document
document = ap.Document("sample.pdf")

# Initialize TextAbsorber with text extraction options
text_absorber = ap.text.TextAbsorber()

# Set extraction options
extraction_options = ap.text.TextExtractionOptions(ap.text.TextExtractionOptions.TextFormattingMode.PURE)
extraction_options.scale_factor = 0.5  # Adjusts text recognition for better column detection
text_absorber.extraction_options = extraction_options

# Extract text from the specified page
document.pages.accept(text_absorber)

# Get extracted text
extracted_text = text_absorber.text

# Save extracted text to a file
with open("ExtractTextUsingScaleFactor_out.txt", "w", encoding="utf-8") as file:
    file.write(extracted_text)

parse-pdf-in-python_get_metadata.py

# This code example shows how to extract file information in Python
import aspose.pdf as ap

# Load the PDF document
document = ap.Document("Sample.pdf")

# Retrieve document information
doc_info = document.info

# Display document metadata
print(f"Author: {doc_info.author}")
print(f"Creation Date: {doc_info.creation_date}")
print(f"Keywords: {doc_info.keywords}")
print(f"Modify Date: {doc_info.mod_date}")
print(f"Subject: {doc_info.subject}")
print(f"Title: {doc_info.title}")

parse-pdf-in-python_highlight_annotation.py

import aspose.pdf as ap

# Load the PDF document
document = ap.Document("annotations.pdf")

# Loop through all annotations on the first page
for annotation in document.pages[1].annotations:
    if annotation.annotation_type == ap.annotations.AnnotationType.HIGHLIGHT:
        # Print annotation details
        print(f"Title: {annotation.full_name}")
        print(f"Annotation Rectangle: {annotation.rect}")

parse-pdf-in-python_line_annotation.py

import aspose.pdf as ap

# Load the PDF document
document = ap.Document("annotations.pdf")

# Loop through all annotations on the first page
for annotation in document.pages[1].annotations:
    if annotation.annotation_type == ap.annotations.AnnotationType.LINE:
        # Print annotation details
        print(f"Annotation Rectangle: {annotation.rect}")

parse-pdf-in-python_link_annotation.py

import aspose.pdf as ap

# Load the PDF document
document = ap.Document("annotations.pdf")

# Loop through all annotations on the first page
for annotation in document.pages[1].annotations:
    if annotation.annotation_type == ap.annotations.AnnotationType.LINK:
        # Print annotation details
        print(f"Annotation Rectangle: {annotation.rect}")

parse-pdf-in-python_text_annotation.py

import aspose.pdf as ap

# Load the PDF document
document = ap.Document("annotations.pdf")

# Loop through all annotations on the first page
for annotation in document.pages[1].annotations:
    if annotation.annotation_type == ap.annotations.AnnotationType.TEXT:
        # Print annotation details
        print(f"Title: {annotation.full_name}")
        print(f"Contents: {annotation.contents}")
        print(f"Annotation Rectangle: {annotation.rect}")