print --with-totals should accept --now

[alex-sparus]

klog print --with-totals is the only view that shows per-entry durations
alongside the record total, so it's what I reach for to answer "what did I
spend today, and on what". But it rejects --now:

$ klog print --today --with-totals --now
Invocation error: unknown flag --now

While I'm still clocked in, the open entry shows 0m and the day total
excludes it — which is exactly the moment I want that number.

--now is already supported by total, today, report, tags and
json, so print is the odd one out.

One design question worth deciding deliberately

NowArgs is a shared mixin, so wiring it in is small. The wrinkle is that
ApplyNow → CloseOpenRanges mutates the records, so where it's called
changes the output:

• Before SerialiseRecords — the line prints as 11:00 - 13:17, a
  synthesised end time that isn't in the file.
• After — Line.Text is already a fixed string while Line.Record is
  still a live reference, so the line keeps showing 11:00 - ? and only the
  duration column counts the elapsed time.

The second seems right for print, whose job is to reflect the file. Tried
it locally against v7.1 (three lines in print.go, existing test suite still
passes) and it gives:

 3h47m  |  2026-08-27
 1h30m  |      9:00 - 10:30 finished work
 2h17m  |      11:00 - ? still running

Happy to open a PR if you'd take it.

[jotaen]

Thanks for bringing this up, this seems to be a valid point and I haven’t thought about it yet. I think this would be worth addressing.

One “concern” would be that supporting --now on klog print may feel odd without --with-totals at the same time, because if they are separate flags, you could legitimately do klog print --now. That would be no different than klog print, however, which may be confusing (because --now wouldn’t do anything on its own).

Which makes me wonder whether klog print --now should actually substitute the ? placeholder with the current time, so that this invocation would make sense on its own. I.e.,

$ klog print --with-totals --now

3h47m  |  2026-08-27
1h30m  |      9:00 - 10:30 finished work
2h17m  |      11:00 - 13:17 still running

and

$ klog print --now

2026-08-27
   9:00 - 10:30 finished work
   11:00 - 13:17 still running

The obvious counter argument is that you couldn’t easily tell anymore which entries are open-ended – except for the subtle colouring difference in the output. I feel this wouldn’t matter as much for klog print --now, but for klog print --with-totals --now it may be odd. A potential middle-ground could be something like

$ klog print --with-totals --now

3h47m   |  2026-08-27
1h30m   |      9:00 - 10:30 finished work
(2h17m) |      11:00 - 13:17 still running

or

$ klog print --with-totals --now

3h47m   |  2026-08-27
1h30m   |      9:00 - 10:30 finished work
2h17m+? |      11:00 - 13:17 still running

[alex-sparus]

I don't have an opinion how it should look like as long as it answers my question: how much did I work today in total and on each item, even when they're still open.

The "fake" --now-induced time can be put in brackets, for example. I personally wouldn't mind even if it stays ?

Thanks a lot for looking at this!

[jotaen]

Feel free to open a PR! If you don’t mind, it would be cool if you could enable maintainer commits on the PR. I’d like to play around with this a bit to get a better feeling for the approaches/variants we discussed above. That said, feel free to share your branch as is for now.

I think I wouldn’t want to put the “fake” end time in brackets, because the klog print command should always output legitimate syntax. So I’d lean towards denoting this on the left-hand side somehow, in the “totals” column.