Less all-time data on BETA & streaks not matching

[Arianthell]

I've checked the previous issues but couldn't see anything else matching to this. This user have less All-time Data on BETA, and claims ver1 data is correct one. I thought this would be the same issue as "Decrease of all time hours + lost streaks #26". But the difference is, the user has not lost the streaks on BETA. Their streak is gone on ver1 (could be the streak loss due to frozen ver1 from a few days ago). Do let me know if it's fine to correct their streaks on ver1.

" Hello the Study!CounterBeta version seems to be off on the hours. of by around 10.8 hours and while I assumed that maybe it wasn't calculating or counting the data prior to its initialisation, doesn't seem like it counted my data today either. so after a 2 hour session today while the leaderboard google sheet updated itself, the BETA bot said 0 hours when I did ~me" - 31/01

"the BETA bot reads my all-time as 126.7 hours while the study together leaderboards google docs sheet reads 139.2 which is the correct metric. the monthly seems fine considering that the roles and data have just reset although I was experiencing issues with the montly metric in January" - 02/02

[Zackhardtoname]

V1 is not controlled by me at all. It's up @Gugu7264 and @nadir7379 to fix this.

[Arianthell]

V1 is not controlled by me at all. It's up @Gugu7264 and @nadir7379 to fix this.

The user says V1 hours is correct and V2 is incorrect. I can fix their streaks for V1 too if they indeed studied everyday, I just need confirmation for that. But All-Time hours issue is on V2, and happened before the reset day. I copied their message under the images above, if you missed.

[Zackhardtoname]

V1 is not controlled by me at all. It's up @Gugu7264 and @nadir7379 to fix this.

The user says V1 hours is correct and V2 is incorrect. I can fix their streaks for V1 too if they indeed studied everyday, I just need confirmation for that. But All-Time hours issue is on V2, and happened before the reset day. I copied their message under the images above, if you missed.

I see. https://github.com/Zackhardtoname/study_together/blob/2814fa8e290219d5522e83430b01b209314c156f/time_counter.py#L217-L221

[Arianthell]

Feel free to close this issue if that makes sense.

No, because I still don't know what to say to the user about their missing hours on V2

[Zackhardtoname]

No, because I still don't know what to say to the user about their missing hours on V2

Yeah I am afraid the missing hours is related to #28 Since they are the same in nature, we can just reference that issue (you already have).

[Zackhardtoname]

If you want, I can refetch the data so v2 has the same monthly, all-time, and streak values as v1 (except for those who have just changed their names and v1 has not updated for them.)

[Arianthell]

If you want, I can refetch the data so v2 has the same monthly, all-time, and streak values as v1 (except for those who have just changed their names and v1 has not updated for them.)

No, majority of the server has more hours on v2 because v1 failed to count for a few days. V2 hours seem to be more accurate for most people. Refetching the v1 data to v2 would mean data loss for a big amount of people, very discouraging, and more work for staff too.

If you can do it specifically for this user only, it's fine I guess? I'd wait for Nadir's instructions however.

[Zackhardtoname]

It should be fixed for this month. See this The problem might come up again at the transition of the next month so I will have to be there monitoring...

[Zackhardtoname]

Just needa watch out for #28

[Zackhardtoname]

There is a grace period as @nadir7379 asked for a fluid streak mechanism in v2 that v1 doesn't have. Thus, v2 is gonna have a "grace period" as commented in the code above (yeah I know, technically a user can study like 50% of the days, at least right now)

Better explanation here: #47