Watson Wrong total in aggregate report

I get the wrong total in an aggregate report:

Wednesday 25 March 2020 (8h 39m 10s)
        bf090fd  07:00 to 08:28   1h 27m 40s  redacted  [a]
        0055576  09:00 to 09:43      43m 32s  redacted  [b]
        18835d8  11:04 to 11:17      12m 58s  redacted  [c]
        9ecf7fa  14:00 to 15:55   1h 55m 21s  redacted  [c]
        6da0ed4  16:20 to 18:29   2h 08m 19s  redacted  [c]
        87ee03f  22:24 to 00:35   2h 11m 20s  redacted  [c]

Wed 25 March 2020 - 6h 27m 50s
  redacted - 6h 27m 50s
        [a  1h 27m 40s]
        [b     43m 32s]
        [c  4h 16m 38s]

Mar 26 '20 06:03 LarsJK

:wave: @LarsJK ! This has been recently fixed in #277 but not released yet. Stay tuned!

Duplicate of #248.

Mar 26 '20 09:03 jmaupetit

I can duplicate the same wrong calculation issue in version 1.10.0. Is this released?

Watson Aggregate Watson Log

The only suspicious thing which I could think of is the use of watson edit. As I sometimes update the entry (and sometimes watson add) the expected behavior is to update the core database such that both aggregate and log match.

Dec 09 '20 16:12 mizhozan

@jmaupetit, @jbarnoud, May I ask to re-open this bug? I also tested for today that if the edit/add might lead to the difference but that's not true as I today with only one segment the result between watson log and watson aggregate differ:

Log: Thursday 10 December 2020 (3h 06m 53s) 2fd6g93 00:48 to 3:55 3h 06m 53s test

Aggregate: Thu 10 December 2020 - 3h 25m 15s test - 3h 25m 15s

Dec 10 '20 03:12 mizhozan

@mizhozan aggregate and log commands do not behave exactly the same. For example, the former will split frames that cross day boundaries while the latter will not. Do you have any other information that could be useful to debug the issue?

Dec 31 '20 11:12 davidag