Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Extracting PDF Metadata and Text With Python

DZone's Guide to

Extracting PDF Metadata and Text With Python

In this post, we take a look at how to use Python and some cool Python packages to extract a few different types of data from PDFs. Read on to get started!

· Big Data Zone ·
Free Resource

The open source HPCC Systems platform is a proven, easy to use solution for managing data at scale. Visit our Easy Guide to learn more about this completely free platform, test drive some code in the online Playground, and get started today.

There are lots of PDF-related packages for Python. One of my favorites is PyPDF2. You can use it to extract metadata, rotate pages, split or merge PDFs, and more. It's kind of a Swiss-army knife for existing PDFs. In this article, we will learn how to extract basic information about a PDF using PyPDF2.

Getting Started

PyPDF2 doesn't come as a part of the Python Standard Library, so you will need to install it yourself. The preferred way to do so is to use pip.

pip install pypdf2

Now that we have PyPDF2 installed, let's learn how to get metadata from a PDF!

Extracting Metadata

You can use PyPDF2 to extract a fair amount of useful data from any PDF. For example, you can learn the author of the document, its title and subject, and how many pages there are. Let's find out how by downloading the sample of this book from Leanpub at https://leanpub.com/reportlab. The sample I downloaded was called "reportlab-sample.pdf".

Here's the code:

# get_doc_info.py

from PyPDF2 import PdfFileReader


def get_info(path):
    with open(path, 'rb') as f:
        pdf = PdfFileReader(f)
        info = pdf.getDocumentInfo()
        number_of_pages = pdf.getNumPages()

    print(info)

    author = info.author
    creator = info.creator
    producer = info.producer
    subject = info.subject
    title = info.title

if __name__ == '__main__':
    path = 'reportlab-sample.pdf'
    get_info(path)

Here we import the PdfFileReader class from PyPDF2. This class gives us the ability to read a PDF and extract data from it using various accessor methods. The first thing we do is create our own get_info function that accepts a PDF file path as its only argument. Then we open the file in read-only binary mode. Next, we pass that file handler into PdfFileReader and create an instance of it.

Now we can extract some information from the PDF by using the getDocumentInfo method. This will return an instance of PyPDF2.pdf.DocumentInformation, which has the following useful attributes, among others:

  • author
  • creator
  • producer
  • subject
  • title

If you print out the DocumentInformation object, this is what you will see:

{ '/Author': 'Michael Driscoll',
 '/CreationDate': "D:20180331023901-00'00'",
 '/Creator': 'LaTeX with hyperref package',
 '/Producer': 'XeTeX 0.99998',
 '/Title': 'ReportLab - PDF Processing with Python' }

We can also get the number of pages in the PDF by calling the getNumPages method.

Extracting Text From PDFs

PyPDF2 has limited support for extracting text from PDFs. It doesn't have built-in support for extracting images, unfortunately. I have seen some recipes on Stack Overflow that use PyPDF2 to extract images, but the code examples seem to be pretty hit or miss.

Let's try to extract the text from the first page of the PDF that we downloaded in the previous section:

# extracting_text.py

from PyPDF2 import PdfFileReader


def text_extractor(path):
    with open(path, 'rb') as f:
        pdf = PdfFileReader(f)

        # get the first page
        page = pdf.getPage(1)
        print(page)
        print('Page type: {}'.format(str(type(page))))

        text = page.extractText()
        print(text)


if __name__ == '__main__':
    path = 'reportlab-sample.pdf'
    text_extractor(path)

You will note that this code starts out in much the same way as our previous example. We still need to create an instance of PdfFileReader. But, this time, we grab a page using the getPage method. PyPDF2 is zero-based, much like most things in Python, so when you pass it a one, it actually grabs the second page. The first page, in this case, is just an image, so it wouldn't have any text.

Interestingly, if you run this example you will find that it doesn't return any text. Instead, all I got was a series of line break characters. Unfortunately, PyPDF2 has pretty limited support for extracting text. Even if it is able to extract text, it may not be in the order you expect and the spacing may be different as well.

To get this example code to work, you will need to try running it against a different PDF. I found one on the United States Internal Revenue Service website here: https://www.irs.gov/pub/irs-pdf/fw9.pdf

This is a W9 form for people who are self-employed or contract employees. It can be used in other situations too. Anyway, I downloaded it as w9.pdf and added it to the GitHub repository as well. If you use that PDF instead of the sample one, it will happily extract some of the text from page 2. I won't reproduce the output here as it is kind of lengthy though.

You may find that the pdfminer package works better for extracting text than PyPDF2 though.

Wrapping Up

The PyPDF2 package is quite useful. We were able to get some helpful information from PDFs using it. I could see using PyPDF on a folder of PDFs and using the metadata extraction technique to sort out the PDFs by creator name, subject, etc. Give it a try and see what you think!

Managing data at scale doesn’t have to be hard. Find out how the completely free, open source HPCC Systems platform makes it easier to update, easier to program, easier to integrate data, and easier to manage clusters. Download and get started today.

Topics:
big data ,python ,data extraction ,tutorial

Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}