Extract Text from PDF using Python

Table of contents

1. Introduction
2. Prerequisites
3. Code Example
4. Return result to a callback url
5. Configuration Options
6. Upload by URL
7. Using Authentication
8. Further details

Introduction

The following tutorial shows you how to extract text from PDFs using a hosted JPedal cloud API. You can set up your own self-hosted JPedal microservice

Whilst the above service can be accessed with plain old HTTP requests, this tutorial uses our open source Python IDRCloudClient which provides a simple Python wrapper around the REST API.

Prerequisites

Using pip, install the IDRCloudClient package with the following command:

pip install IDRCloudClient

Code Example

Here is a basic code example to extract text from PDFs. Configuration options and advanced features can be found below.

from IDRSolutions import IDRCloudClient

client = IDRCloudClient('https://my-self-hosted-service.com/' + IDRCloudClient.JPEDAL)

try:
    result = client.convert(
        input=IDRCloudClient.UPLOAD, 
        file='/path/to/exampleFile.pdf',
        ,
        settings='{"mode":"extractText","type":"plainText"}'
    )
    outputURL = result['downloadUrl']

    client.downloadResult(result, 'path/to/output/dir')

    if outputURL is not None:
        print("Download URL: " + outputURL)

except Exception as error:
    print(error)

Return result to a callback url

The JPedal Microservice accepts a callback url to send the status of a extraction on completion. Using a callback url removes the need to poll the service to determine when the extraction is complete.
The callback url can be provided to the convert method as shown below.

result = client.convert(
    input=IDRCloudClient.UPLOAD,
    callbackUrl='http://listener.url',
    file='/path/to/exampleFile.pdf',
    ,
        settings='{"mode":"extractText","type":"plainText"}'
)

Configuration Options

The JPedal API accepts a stringified JSON object containing key value pair configuration options to customise your extraction. The settings should be provided to the convert method. A full list of the configuration options to extract text from PDFs can be found here.

settings='{"key":"value","key":"value"}'

Upload by URL

As well as uploading a local file you can also provide a URL which the JPedal Microservice will download and then perform the extraction. To do this you should replace the input and file values in the convert method with the following.

input=IDRCloudClient.DOWNLOAD
url='http://exampleURL/exampleFile.pdf'

Using Authentication

If you have deployed your own JPedal Microservice that requires a username and password to extract text from PDFs, you will need to provide them with each conversion. These are provided by passing a variable named auth to the convert method as shown below.

auth=('username', 'password'))

Further details

IDRCloudClient on GitHub
IDRCloudClient on Pypi
JPedal Microservice API
JPedal Microservice Use