PDF printout plugin

Development Plugins
1

I am currently working on a plugin that can convert a PDF to an image printout and insert the images into a note.
I have a working setup but it is very much hacked together, as it calls a Python script I've written for everything related to PDF conversion. Naturally, I would like to re-write this as a native Joplin Plugin.

I am currently stuck with this as I cannot get a minimal example with PDF.js to work.

The minimal example would be something like this:

var pdfjs = require("pdfjs-dist");
const doc = await pdfjs.getDocument('/<path-to-pdf>').promise;

The error I am getting is the following:

    ERROR in ./node_modules/pdfjs-dist/build/pdf.js 924:33
    Module parse failed: Unexpected token (924:33)
    You may need an appropriate loader to handle this file type, currently no loaders are configured to process this file. See https://webpack.js.org/concepts#loaders
    |   const httpHeaders = src.httpHeaders || null;
    |   const withCredentials = src.withCredentials === true;
    >   const password = src.password ?? null;
    |   const rangeTransport = src.range instanceof PDFDataRangeTransport ? src.range : null;
    |   const rangeChunkSize = Number.isInteger(src.rangeChunkSize) && src.rangeChunkSize > 0 ? src.rangeChunkSize : DEFAULT_RANGE_CHUNK_SIZE;
     @ ./src/index.ts 15:12-33

It seems to be related to the nullish coalescing operator not being defined. Asking ChatGPT and from trying to Google the answer, it seems like I have two options:

  1. Change the loader in webpack.config.js to use Babel and include the plugin
  2. Update the target versino inside tsconfig.json

I've had success with neither and would very much appreciate some help porting my script.

2

Can't you find a version of pdf.js that's compiled as commonJS? Or maybe build one yourself? Then you can require it as normal, without having to change the build system

3

That might be a better approach. I'll give it a shot and let you know of my findings.

Thanks!

4

Did it work ?

5

I am afraid not. It ended up being beyond my TypeScript skills so I unfortunately abandoned the attempt.

If anyone wants to pick it up and want the Python script for inspiration then feel free to message me

6

Can you share your python script ?

Did you search for other solutions that can convert pdf to images ?

7

This GitHub issue is related — hopefully it will soon be easier to convert PDFs to images from plugins:

8

The Python script can be found here: GitHub - mortenhaahr/joplin-pdf-ocr-upload
It currently also runs OCR on the imported PDF and adds it as metadata to the images. I would recommend skipping this on an initial version of the plugin.

Furthermore, I have the Joplin Plugin which is located here: GitHub - mortenhaahr/JoplinPDFPrintout
It works (tested on Linux) if the user installs it as a dev plugin and successfully installs the Python script.
I am glad to help more if needed

9

I just learned of the Joplin OCR feature included in v2.14.6 so I guess that part is no longer necessary.