Issue 1316: Skip measurable habit

[kalina559]

Resolves #1316. My solution is to add a 'SKIP' button to the numerical habit dialog. If it's clicked, the entry value is set to Entry.SKIP / 1000 = 0.003 and it's displayed in the same way as a skipped yes/no habit.

[hiqua]

Very nice, thanks!

[hiqua]

The ktlint stuff can be a bit annoying, you can have a look at https://github.com/iSoron/uhabits/blob/dev/docs/GUIDELINES.md#code-style to make it less annoying. I use the commit pre-hook personally.

[iSoron]

Hi @kalina559, thank you for the PR. Besides the comments above, I have a few other questions:

1. For numerical habits with daily frequency, how does a SKIP differ from an entry with value 0? It appears to me that the difference, at this point, is purely visual. Shouldn't SKIP also have an impact on the score and the "target" chart?
2. For numerical habits with weekly or monthly frequency, does it even make sense to allow SKIP? When should the user enter this option, instead of entering zero? What's the impact on score and target in this case?

[kalina559]

Hi @iSoron,

1. You're right, I'll look into how I can affect the score and target chart with this
2. User can enter SKIP in the history chart or in the reminder dialog. Maybe if all entries in a week/month are skipped, then the score and target are affected somehow?

[iSoron]

You're right, I'll look into how I can affect the score and target chart with this

For scores, I suggest copying the behavior from boolean habits: a SKIP keeps the score the same as the day before.

For the target chart, one thing that may make sense is to reduce the total. For example, if you have a daily numerical habit with target of 10, then the chart by default would show a monthly target of 300 (for a 30-day month). However, if you skip one day, the monthly target for that month could drop to 290. This might be difficult to implement, though. We would need to count the number of skips in a certain interval.

Regarding skips in non-daily numerical habits, I suggest disabling them for now until we understand that use case better. They don't seem to make a lot of sense to me at this point, but I'm open to suggestions.

[ke352802081770314]

Regarding skips in non-daily numerical habits, I suggest disabling them for now until we understand that use case better. They don't seem to make a lot of sense to me at this point, but I'm open to suggestions.

As you explained, skip should reduce the multi-day targets. so if I have a target "150 per week" and I have a skip day in that week, the target should be 6/7 =(7-skip_days)/7 of that number, so it is really what you already suggested, only the user-supplied value is on a different scale (weekly given directly instead of calculated as 7*daily etc.).
I must admit scoring isn't that transparent to me so I can't explain how to adjust scoring in that case. My conceptual logic would be "skip days are treated like they didn't exist in the calendar" i.e. skipping 2022-03-13 would make the scoring think that the next day after 2022-03-12 is 2022-03-14 and everything else follows logically from that.

It could be argued that the skip supports laziness, because if I intended to achieve the cumulative number not by doing it daily anyway, skip could be used to lower the target. But honesty is always up to the user, and if I mark three days as skip because I was bedridden sick it is correct that my jogging target for that week should only be 4/7 of when I had been healthy.

[iSoron]

Thank you for the PR, @kalina559. I think it looks mostly good.

@hiqua Please free to add any comments.

[iSoron]

Looks good to me. Thanks, @kalina559!