Newspaper generator

From Hackers & Designers
Revision as of 17:11, 23 February 2022 by Hd-onions (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
MainNavigation No

NDSM-news.001.png

Repository

NDSM newspaper 2019 generator

Description

Script to auto-generate a book from a bunch of documents downloaded from Google docs as HTML files. The script will remove most of the CSS and combine all the documents in one file: build/book.html. Then Weasyprint converts the HTML to PDF. The fonts used are all open source fonts.

Installation

Make sure you have weasyprint installed as per instructions : https://weasyprint.readthedocs.io/en/stable/install.html

Mac

On mac you'll have to install some stuff with Homebrew, so read the guide.

Install the requirements from file:

pip install -r requirements.txt

Usage

Export the documents as HTML (File > Download > HTML page...) from Google docs and place in ./srcdocs

Run clean command

Google docs places a lot of representational css inline in the document. This command cleans most of that up. We only leave the font-weight and font-style rules because we actually need those. The cleaned documents will be placed in ./build/clean. Adjust those as needed. By default is searches for files in the .srcdocs folder

python generate.py --clean

Or specify source folder: python generate.sh --clean --input ./srcdocs_en

Run build command

This will combine all the files in ./build/clean by getting the contents of the document body and wrapping that in an <article> tag. The combined document will be saved as ./build/book.html and this file is used by Weasyprint to generate ./build/book-<date>.html and ./build/book-<date>.pdf

python generate.py --build

optionally Set the output filename by adding --output [filename]

python generate.py --build --output ndsm_papger.pdf

Create PDF from preexisting html

This option allow for small text changes. To do that edit the generated /build/book-<date>.html and pass that to the --html flag

python generate.sh --html book-2019-0920-143629.html

Examples

Clean the input html and generate a Pdf

python generate.sh --clean --input ./srcdocs_en && python generate.sh --build

Resources

https://www.w3.org/TR/css-page-3/#cascading-and-page-context https://www.smashingmagazine.com/2015/01/designing-for-print-with-css/

Troubleshooting

We had a problem with a Cairo version, so we had to force a version with pip install cairocf==0.9.0

We had locale errors on some machines. Set locale with:

export LC_ALL=en_US.UTF-8
export LANG=en_US.UTF-8

or add those to the shells .rc file

Fonts

  • Todo. Add links to font authors