Pltt is a Gitlab Time Tracker client for the command line. It mimics the interface of kriskbx/gitlab-time-tracker and is compatible to the config and frame database. Thus, it works as a drop-in-replacement.
Goals/differences to gtt:
- More helpful actions:
pltt start- can be called without issue id and displays a CLI select to choose the issue to book ongtt stop- possible to fix issue start time / duration theregtt create- creates issue on Gitlab immediately, can input title, select labels, enter description and also optionally create a branch/MR (like the button in Gitlab Web) and immediately switch to that
- More robust interface with Gitlab:
- Not possible to book on non-existing or closed issues
- Time Entries are booked on stop by default, so no date shifting for forgotten frames
- pltt status shows issue body, milestone etc. too
- Fast interface, even though it's ruby, the boot time is comparable or faster than the Nodejs version
gtt status 0,87s user 0,10s system 110% cpu 0,883 totalpltt status 0,49s user 0,04s system 71% cpu 0,741 total- -> even though pltt status does more (fetches whole issue title, description etc from Gitlab), it is faster
gtt start 79 0,87s user 0,10s system 111% cpu 0,867 totalpltt start 79 0,47s user 0,06s system 70% cpu 0,758 total- -> again, pltt does more - it makes an API request to gitlab to validate that issue exists and is not closed.
- optimizations:
- Dependency minimizations - only necessary deps (gitlab api, tty, thor, oj, hashids)
- most actions, like status etc only looks for the most recent frames in the framedir before parsing the JSON, so a larger framedir has a smaller effect on performance
- smaller code size, e.g. currently ~500 LOCs Ruby vs. >3000 LOCs JS
- Gitlab's Time Tracking database is more or less append-only. So, once synced entries are undeletable by this client. If you need to fix some time tracking entry, you need to manually subtract that in the Issue, like "-3h"
- Gitlab's Time Tracking API (and web client too) has no means of specifying the date/time of the spend entry. This is why, the entry is always posted as "right now". There is an Issue about that. One of the motivations of this client was, that after stopping entries, they are synced immediately
Install it yourself as:
$ gem install pltt
# start without issue id -> shows a selection menu of 30 most recent issues in project
pltt start
# start with issue id -> validates issue exists and is not closed
pltt start 123
# create new issue with interactive question for issue creation
# issue will be created instantly, not on sync
# You can enter issue title, select labels, and description
pltt create
# starts tracking on the issue that was most recently tracked before
pltt resume# stops time tracking & syncs to gitlab
# 2 interactive questions can easly fix missing started entries
# can adjust start time and duration (minutes)
pltt stop
# Edit current entry in EDITOR
pltt edit
pltt edit 19ad20
# Cancel = delete current running entry
pltt cancel
# normally not needed, as sync is made after stop, only in case of error
pltt sync# show all issues in project as table
pltt list
pltt list --my
pltt list --label BugMissing before release:
- gtt start with no arguments shows the list of open issues to select one from
- gtt create can optionally create a MR + branch
- gtt stop / sync note creation
- First-Start Guide, Generator which creates .gtt.yml in a project, gets the project url from .git/config
- Timezone
- Frame Cleanup, everything older than X month can delete
Missing, but not planned soon, as not needed by us:
- Logging/reporting of all bookings
- No Booking on Merge requests implemented
- No delete implemented, as cancel is enough. Deleting frames is also not very useful, when they are already synced to Gitlab
Bug reports and pull requests are welcome on GitHub at https://github.com/pludoni/pltt.
The gem is available as open source under the terms of the MIT License.