[FEATURE] Simplify PDF Generation – Avoid Puppeteer Overhead

[malay77patra]

Problem Statement

Currently, Puppeteer is used to convert HTML to PDF, but this approach is excessive for our needs. It is also difficult to set up in cloud environments.

Users input data through a form, and we extract this data as JSON. Since we only need structured data from the form, converting HTML to PDF is unnecessary.

Proposed Solution

Instead of rendering HTML and converting it to PDF, we can directly generate PDFs using a JavaScript-based PDF library. The form data can be processed as JSON and used to build the PDF without requiring an intermediate HTML representation.

Suggested Alternatives

• Use jsPDF or a similar JavaScript library to generate PDFs directly from form data.
• This would simplify the implementation, reduce dependencies, and improve cloud compatibility.

Additional Information

If that sounds useful I am happy to help you out with the implementation.

[al1abb]

Sounds cool. I remember trying this approach first, but I couldn't get it to work, so I used puppeteer instead

[curious-companion]

@al1abb @malay77patra Since jsPDF directly converts JSON to a PDF format, how can we apply Tailwind styles (from TAILWIND_CDN) to the PDF? Since Tailwind is designed for HTML, would rendering the styled content to an image using html2canvas and then embedding it into the PDF be a suitable approach?

[malay77patra]

Hi @al1abb !

I found a way to generate a PDF from HTML using the react-pdf (with react-pdf-html) libraries, which can completely eliminate the need for Puppeteer.
However, it requires raw HTML and CSS code with styles applied.

Is it possible to get the raw HTML and CSS of the templates?

I noticed that currently, it retrieves the HTML with Tailwind classes and adds a Tailwind CDN to the page for styling.