Miss any of our Open RFC calls?Watch the recordings here! »

pdf-extraction

1.0.2 • Public • Published

pdf-extraction

Pure javascript cross-platform module to extract texts from PDFs.

version downloads node status

Installation

npm install pdf-extraction

Basic Usage - Local Files

const fs = require("fs");
const pdf = require("pdf-extraction");
 
let dataBuffer = fs.readFileSync("path to PDF file...");
 
pdf(dataBuffer).then(function (data) {
    // number of pages
    console.log(data.numpages);
    // number of rendered pages
    console.log(data.numrender);
    // PDF info
    console.log(data.info);
    // PDF metadata
    console.log(data.metadata);
    // PDF.js version
    // check https://mozilla.github.io/pdf.js/getting_started/
    console.log(data.version);
    // PDF text
    console.log(data.text);
});

Basic Usage - HTTP

You can use crawler-request which uses the pdf-extraction

Exception Handling

const fs = require("fs");
const pdf = require("pdf-extraction");
 
let dataBuffer = fs.readFileSync("path to PDF file...");
 
pdf(dataBuffer)
    .then(function (data) {
        // use data
    })
    .catch(function (error) {
        // handle exceptions
    });

Extend

// default render callback
function render_page(pageData) {
    //check documents https://mozilla.github.io/pdf.js/
    let render_options = {
        //replaces all occurrences of whitespace with standard spaces (0x20). The default value is `false`.
        normalizeWhitespace: false,
        //do not attempt to combine same line TextItem's. The default value is `false`.
        disableCombineTextItems: false,
    };
 
    return pageData.getTextContent(render_options).then(function (textContent) {
        let lastY,
            text = "";
        for (let item of textContent.items) {
            if (lastY == item.transform[5] || !lastY) {
                text += item.str;
            } else {
                text += "\n" + item.str;
            }
            lastY = item.transform[5];
        }
        return text;
    });
}
 
let options = {
    pagerender: render_page,
};
 
let dataBuffer = fs.readFileSync("path to PDF file...");
 
pdf(dataBuffer, options).then(function (data) {
    //use new format
});

Options

const DEFAULT_OPTIONS = {
    // internal page parser callback
    // you can set this option, if you need another format except raw text
    pagerender: render_page,
    // max page number to parse
    max: 0,
    //check https://mozilla.github.io/pdf.js/getting_started/
    version: "v1.10.100",
};

pagerender (callback)

If you need another format except raw text.

max (number)

Max number of page to parse. If the value is less than or equal to 0, parser renders all pages.

version (string, pdf.js version)

check pdf.js

  • 'default'
  • 'v1.9.426'
  • 'v1.10.100'
  • 'v1.10.88'
  • 'v2.0.550'

default uses version v1.10.100
mozilla.github.io/pdf.js

Test

Support

I use this package actively myself, so it has my top priority. You can chat on WhatsApp about any infos, ideas and suggestions.

Submitting an Issue

If you find a bug or a mistake, you can help by submitting an issue to GitLab Repository

Creating a Merge Request

GitLab calls it merge request instead of pull request.

License

MIT licensed and all it's dependencies are MIT or BSD licensed.

Install

npm i pdf-extraction

DownloadsWeekly Downloads

120

Version

1.0.2

License

MIT

Unpacked Size

33.3 MB

Total Files

770

Last publish

Collaborators

  • avatar