Research: Identify open-source python libraries for PDF generation

[rjvitorino]

Goal

Find and assess Python tools for generating high-quality PDFs suitable for LinkedIn carousels.

Tasks

• Evaluate libraries like ReportLab, WeasyPrint, FPDF, etc.
• Consider layout flexibility, visual quality, and carousel-style output

[cornzyblack]

Here is an overview of my research

🏆 Chosen PDF Library: WeasyPrint

At a Glance — Usage & Community Stats

WeasyPrint

---

• It has 7.6k Github stars which implies that it is strong and mature.
• It is also regularly maintained
• At the moment of writing this it has about 150 open issues on Github and it is regularly maintained as well judging from the last commits
• It's pip installation statistics, last month download of ~ 4 million and a last week download of ~ 1 million
• It has a BSD 3-Clause License, license which means it can be used in an open-source app as well.

fpdf2

---

• It has about 1.2k Github stars of fpdf2 so it is growing
• At the moment of writing this it has about 50 open issues on Github, and it is regularly maintained as well judging from the last commits
  -- It's pip installation statistics has a last month download of ~ 1 million and a last week download of ~ 480 thousand
• It has a LGPL license which means it can be used in an open-source app as well.

reportlab

---

• Could not find the open source repo for this, so mean it might be privately maintained by ReportLab, hence we are dependent on ReportLab directly for updates and changes
• It has a high usage and it's pip installation statistics has a last month download of ~ 6.8 million and a last week download of ~ 1.8 million
• It has a BSD license which means we can use the open-source version for our open-source app

Developer Experience

WeasyPrint

• Learning curve is relative, but genrally quite low since it builds layouts using HTML and CSS.
• There is more control in terms of styling a page, because of CSS functionalites
• Good support for images and different fonts even CDN hosted fonts (since it allows for HTML and CSS)
• Allows for reusable templates (HTML, CSS)
• Fit for different document styles

fpdf2

• Allows for reusable templates
• Seems to be better fit for tabular document styles, e.g invoices, etc
• Styling is also done programmatically

reportlab

• Learning curve is medium, as it not requires Python but also some coordinate geometry for positioning objects (text, images, etc) in PDFs
• Allows for templating as well (with python functions) but might be more complicated as it coordinate geometry is required

Evaluation

After evaluating the highlighted libraries above: including WeasyPrint, ReportLab, and FPDF2 for PDF generation libraries — I selected WeasyPrint as the most suitable tool for this project.

Why WeasyPrint Wins

HTML & CSS Styling

Layouts, spacing, fonts, and positioning are all handled through familiar CSS, removing the need to calculate (x, y) coordinates or use low-level drawing APIs.

Modern Design Capabilities

It relies on web standards, WeasyPrint supports modern fonts, flexbox/grid layouts, circular images, and overlays — ideal for polished, presentation-style PDFs.

Clean and Maintainable Codebase

It is easy to build a reusable base HTML/CSS template and update it for different episodes. This keeps the layout consistent while simplifying future content updates.

After evaluating popular open-source PDF generation libraries — including WeasyPrint, ReportLab, and FPDF2 — I selected WeasyPrint as the most suitable tool for this project.