Just as an extra info. Not onl can pdf-lib create new pdfs. It also can edit pdfs. I used a lot in my environment. But jear it’s big.
What about a pure “Fill PDF” node? Would be clearer and more limited in scope.
pdf-lib would be clear preference. I was able to build a quick prototype with pdf-lib in the function-item node.
Welcome to the community @vcrwr!
There is already a node for APITemplate.io which allows PDF creation.
إعجاب واحد (1)
Quite nice. Thanks @jan
Unfortunately, we need to keep these docs internal. Not only for GDPR reasons. But also for them, Singapore is probably going to be difficult. 
Thats how we came to n8n. >Zapier kann sich warm anziehen<
I was trying to understand, whether there would be a broader interest in such a “Fill PDF” Node and if so, offering to work on it.
i have made a workflow to make PDF’s and its all local
try using this
إعجابَين (2)
Hi,
@RedPacketSec thats a different workflow than the one we have in mind. We need to fill a pre-designed PDF with Data from a CRM.
@jan does it make sense to work on a “Fill PDF” node (with pdf-lib) on our side, or would you rather recommend to stick with a custom function-item workflow?
Yes you can do that easily.
You pre design the .odt file and have the keynames as template position holders and fill it with the crm data when executed.
I do it already but using API json data from a different system.
Just send the commands via SSH to my system running that software and then pulls it back into n8n when done.
Just to clarify. I’m not trying to solve my own problem. This is already solved with function-item and pdf-lib. I just thought about providing this “Fill-PDF” workflow as a standard node for n8n and was interested in opinions about it. So clap along (gimme a heart) if you think, its worth the dev efforts and should make its way into n8n.
8 إعجابات
I for one would love to see such 'fill-pdf" node !
i’m currently messing around with PDF’s that I read, then translate the text part, but then I want to write a “last page” with my translation but I cannot find a way to write my pdf from my binary blob.
Wondering if that wouldn’t be the solution, would be happy to test !
إعجاب واحد (1)
Anyone know how I can send PDFs created by APITemplate.io via HTTPS Request (with Customers coming in from GoogleSheets?)
إعجاب واحد (1)
for me it worked using Gotenberg, I found a video explaining how to use it in n8n
إعجابَين (2)
I’m the founder of APITemplate.io and CraftMyPDF.com.
If you’re looking for a way to create PDFs from HTML templates, APITemplate.io is the way to go. We have a native node developed for n8n.
Our other product, CraftMyPDF.com, offers a drag-and-drop editor and you can import existing PDFs, and we’ve recently built a community node for n8n. Here’s the tutorial: Automate PDF Generation with N8n and CraftMyPDF - CraftMyPDF.com
إعجابَين (2)
There is also a community node if you want to do it locally.
DocX Template → Insert data with JSON format → PDF (libreoffice)
إعجاب واحد (1)
I wrote a module that could help with this as well
@couleetech/n8n-nodes-pandoc - npm
You should be able to convert html or markdown to pdf ( I have only tested Word Doc to Markdown, but the other options are in there )
إعجاب واحد (1)
If you want to generate PDFs that can also deal with complex data, like tables, charts, qr codes, images, you can use my SaaS: pdforge
It has a no-code template builder that’s really fast to use and the product was created thinking about transforming data into visual reports as n8n workflow output.
I also have an n8n community node to make your life easier: n8n-nodes-pdforge
Hope it helps and feedbacks are welcome!
also thought of this
PDFMunk has a really cool node on n8n for this called ‘HTML to PDF’ that allows dynamic PDF generation.
Hi all
Official Carbone.io integration is now available : Carbone integrations | Workflow automation with n8n
إعجاب واحد (1)
+1 for carbone. Really fast and stable pdf creation. Really good template creation on their online studio. Been using it for a couple years now after switching from pdf.co
its more pricey than pdf.co, but speed and templates mostly makes up for that
A few dependencies listed in this thread targets JavaScript, but we could rely on a trusted system level tool instead : Pandoc for instance https://pandoc.org/ is available on Linux, and could be wrapped with some JS glue code.
This CLI approach is typically how AI agent tend to run PDF→text conversion in the other direction, via poppler-utils pdftotext (pdftotext(1) — poppler-utils — Debian testing — Debian Manpages) and a similar solution on Windows (https://www.xpdfreader.com/pdftotext-man.html). So I guess it could work for text→PDF