I currently don't have the time to actively maintain this repo. That being said I still believe that this library is of good use to anyone looking to work with PDFs in Node as it simplifies a lot of what's cumbersome with PDFs, but currently I can't recommend anyone to use it in production as I cannot actively maintain it. I do however greatly welcome PRs and issues and will try to respond ASAP.
If your company would benefit from using this package, please consider sponsoring me and the package's continued development and feel free to reach out with any questions regarding maintenance.
simple-pdf
aims to be a simple drop-in module for extracting text and images
from PDF files. It exposes a promise-based and an event-based API.
- Extracts both text and images
- Handles most image encodings
Let's be real. This might not be the library for you. Here are a few reasons why.
- Slow with images - Images can be embedded in a PDF in many different ways. To ensure that all types of images can be extracted we render the whole PDF and then use sharp to extract the images from the rendered page. This adds extra processing time for pages that contains images (provided that you don't disable image extraction).
- New to the game - This library is brand new and hasn't been battle tested yet. If you're looking for a reliable solution, this library might not be the best choice for you.
- No automated testing - Though I'm working on this 🙃
Minimal example:
const fs = require('fs')
const { SimplePDFParser } = require('simple-pdf')
const fileBuffer = fs.readFileSync('somefile.pdf')
const parser = new SimplePDFParser(fileBuffer)
parser.parse().then((result) => {
console.log(result)
})
More examples can be found in the examples
directory and can be run with the following commands:
npm run example:events
npm run example:promises
npm i simple-pdf
The only exposed interface is the SimplePDFParser
class. It takes a Buffer
containing a PDF file as well as an optional options object.
new SimplePDFParser(fileBuffer, {
// options
})
Option | Value type | Default value | Description |
---|---|---|---|
paragraphThreshold |
integer | 25 |
The minimum distance between two lines on the y-axis to consider them part of separate paragraphs. This option only affects the parse method. |
lineThreshold |
integer | 1 |
The minimum distance between two lines on the y-axis to consider them part of the same line. PDFs usually suffer from issues with floating point numbers. This value is used to give a little room for error. You shouldn't have to change this value unless you're dealing with PDFs generated with OCR or other odd PDFs. |
imageScale |
integer | 2 |
Scaling applied to the PDF before extrating images. Higher value results in greater image resolution, but quadratically increases rendering times. |
extractImages |
boolean | true |
Controls whether or not to extract images. Image extraction requires rendering of each page, which might take a long time depending on the size of the PDF, configured imageScale and underlying hardware. If you don't need to extract images, setting this option to false is recommended. |
ignoreEmptyText |
boolean | true |
Controls whether or not to ignore empty text elements. Text elements are considered empty if their text content contains nothing by whitespace. |
joinParagraphs |
boolean | false |
Controls whether or not to join paragraphs. Enabling this option will join each line that's not separated by a non-text element (paragraph break or image) which will effectively make each line contain a paragraph. Paragraph breaks will be omitted from the final output. This option only affects the parse method. |
imageOutputFormat |
string | png |
Controls what format the image is exported as. Defaults to 'png'. Passed directly to Sharp: https://sharp.pixelplumbing.com/api-output#toformat |
This is probaly the easiest way to use this library. It parses all pages in parallel and returns the result when finished. Paragraphs and lines are automatically joined based on the options passed to the constructor.
Example:
const parser = new SimplePDFParser(fileBuffer)
const result = await parser.parse()
Result:
[
{
"type": "text",
"pageIndex": 0,
"items": [
{
"text": "Lorem ipsum",
"font": "g_d0_f1"
}
]
},
{
"type": "image",
"pageIndex": 0,
"imageBuffer": Buffer
}
]
If you need more granuar control of the resulting data structure you might want to use the advanced parsing. You can choose to either just await the result or use the events to process each page as it is finished parsing. Note that pages are not guaranteed to be returned in order.
Example:
const parser = new SimplePDFParser(fileBuffer)
// Called with each page
parser.on('page', (page) => {
console.log(`Page ${page.index}:`)
console.log('Text elements: ', page.textElements)
console.log('Image elements:', page.imageElements)
})
// Called when the parsing is finished
parser.on('done', () => {
console.log('Parser done')
})
// This must be run even if you just use the events API, but then you may ignore the return value
const result = await parser.parseRaw()
Result (each page):
{
index: 0, // Page index
textElements: [{
x: 123.456,
y: 654.321,
items: [{
text: 'Lorem ipsum',
font: 'g_d0_f1'
}]
}],
imageElements: [{
x: 4.2,
y: 83.11,
width: 120,
height: 80,
imageBuffer: Buffer
}]
}
More of a todo, but let's call it a roadmap
- Tests
- Better coverage
- Windows - Something is wrong either with the library or the tests (https://github.com/scriptcoded/simple-pdf/runs/1048499489)
- Make a logo (everyone likes a logo)
- Rewrite codebase in TypeScript
- Improve image extraction
- Set up automatic CI/CD pipeline for NPM deployment
- Simplify the API
Tests can be run with with the following commands:
npm run build
npm run test
Contributions and PRs are very welcome! PRs should go towards the develop
branch.
We use the Airbnb style guide. Please run ESLint before committing any changes:
npx eslint src
npx eslint src --fix
This project is licensed under the MIT license.