Skip to content

Latest commit

 

History

History
48 lines (30 loc) · 2.19 KB

README.md

File metadata and controls

48 lines (30 loc) · 2.19 KB

upsheet

uploads a simple plain-text timesheet into Jira as a bunch of worklog entries. Given a timesheet.txt containing

20130130
  1945-2015 UPSH-22: cleaning up the code
20130131
  0530-0600 UPSH-21: writing a README
  0600-0630 UPSH-22: setting up a GitHub repository

when you say

$ upsheet timesheet.txt

it creates one worklog entry for issue UPSH-21 and two for UPSH-22.

Configuration

upsheet is a simple Python script with all configuration defined as constants at the beginning:

# Configuration
URL  = "http://localhost:8080/rest/api/latest/issue/%s%s"
USER = "admin"
PASS = "" # if empty, user will be prompted for password

If you need to upload timesheets to different Jira instances you'll have to make multiple copies of the script, adjusting the settings as needed.

Rationale

I work as a contractor on a project where Jira worklogs are used as basis for invoicing so it's pretty important to keep them up to date. Jira is a fine piece of software overall but firing up the worklog entry form each time I'm switching tasks is a pain. Appending a line to a file is much easier so I ended up keeping a plain-text timesheet and updating Jira once a day.

That was the plan, anyway. I ended up spoon-feeding Jira for a solid hour at the end of each month. Not fun. I always had a hunch that Jira probably had a REST API and when I got around to reading the docs it didn't look that hard. The present script is the result of a few months of gradual tweaking.

Future

Some colleagues prefer a timesheet file format where only the start time of an activity is logged:

20130131
  0830 PROJ-3998 writing Selenium tests for the assignment page
  1200 lunch
  1230 PROJ-3995 fixing the assignment page
  1800 out for the day

You can see the trade-off: breaks in activity have to be mentioned explicitly. upsheet is not friends with such files but I can imagine a version that would be. It could even auto-detect which format is being used and adapt.

Supporting other time-tracking systems besides Jira is also an option but I'm afraid it would pollute what is now a sharply focused tool.

More realistic objectives are to be found in the issue tracker.