This is a fat Jira client inspired by:
- How SLOW it is to use a thin web client to find an issue and log some work.
- How fast incremental search is AWESOME and changes your way of doing things (like fzf, helm, ivy).
- How keyboard based interfaces can be discoverable for beginners and muscle-memory building for advanced users (like magit).
- How switching to a browser to find an issue can be distracting and time consuming (and how a simple CLI is still insufficient).
If you've never used incremental search, for eg. fzf or Emacs with Helm/Ivy to
search through commands, buffers, file names, file contents you might not
appreciate fatjira.
I want to find and alter the issue as fast and effortlessly as I can commit code, move the commit to a feature branch (because I forgot earlier), amend it (cause I forgot to include a changelog) and push it in Magit.
Recording of an example session using a fake 10000 entries based on europarl parallel corpus dataset. Some searching and adding a simulated worklog. Video is a bit laggy compared to real app on my computer. Neither termtosvg nor ttyrec seemed to follow the high refresh rate of the app.
Works:
- Issue synchronisation,
- basic (but fast) search,
- viewing issue details and recent worklogs,
- adding worklogs.
Work in progress. :P And I probably won't be able to push it to be useful by myself (time :/), but maybe showcasing the concept will inspire someone? Pretty please?
Testing with 5500 jira issues, each loaded with 333 custom fields (not filtered yet). Current PoC searches them by 6 fields in unnoticeable time even in a synchronous way. It does take 400MB RAM and loading from disc takes around 2 seconds. On disc, local cache takes 110MB. Synchronisation (depends on many factors) of local cache takes around 6 minutes, and further daily resync up to 30 seconds (most time was taken by logging in).
When filtering the custom keys out during sync, local store takes 48MB, and loading takes 0.7s instead of 2s.
I considered combining it with orgassist project as it scratches a similar itch.
Currently from the GIT, no install over PIP is available.
pipenv install
pipenv shell
cp config.tmpl.py config.py
vim config.py # Set server data
./fj --update
./fj # try with --offline.