Skip to content

Latest commit

 

History

History
88 lines (55 loc) · 2.71 KB

instructions.md

File metadata and controls

88 lines (55 loc) · 2.71 KB
Raw

Creating your Zulip archive takes a few steps to set up.

Download dependencies

  • Clone this repo.
  • Download python3 if you don't already have it. (We require version 3.6 or higher.)
  • Install the dependencies, with pip3 install -r requirements.txt.

Get a Zulip API key

You will need an API key to get data from Zulip. Often you will do this by creating a bot, but you can also use your main user's API key.

  • Download a zuliprc file to zuliprc within this project.

Customize your settings

  • Run this command:

    cp default_settings.py settings.py

  • Then read settings.py and modify the settings to fit your needs. (There are comments in the file that explain each setting.)

  • Optionally, modify the code to fit your needs. This repo is based on the leanprover-community Jekyll setup.

Build JSON files from your Zulip instance

  • Create a directory to store JSON in (see settings.py for more details).
  • Run python3 archive.py -t to download a fresh archive. (This may take a long time. You may wish to experiment with just a few streams at first--see settings.py for details.)

Note: you will be able to update your archive later with python3 archive.py -i to get more messages.

Build the HTML files

Run this command to build your archive

python3 archive.py -b

Test your changes locally

You can use this command to serve your files:

python3 -m http.server 4000

Typically you will then view your files at http://127.0.0.1:4000/archive/.

Add assets

You may wish to copy the following assets into your site directory structure:

  • assets/img/zulip.svg

Go to production

Once you are satisfied with your local testing, you will want to host your archive publicly. See hosting.md for more details.

Other notes

archive.py

The main tool to familiarize yourself with is archive.py. It takes these options:

  • -t builds a fresh archive. This will download every message from the Zulip chat and might take a long time. Must be run at least once before using -i.
  • -i updates the archive with messages posted since the last scrape.
  • -b generates the markdown/html output.

github.py

This repostiory also contains a hacky tool for managing pushes to a repository hosted by GitHub Pages, which supports the following options. Be sure to read the warnings in github.py.

  • -f updates the git repository containing the script
  • -p pushes the generated files

Contributions are appreciated to make github.py no longer hacky.