Question about RevlogEntry mapping about fsrs-rs

[L-M-Sherlock]

Oops. I made a mistake. I will fix it. And use const.

https://github.com/open-spaced-repetition/fsrs-optimizer/blob/95694b787bb71ac9883db1201af09e334ee5ee0b/src/fsrs_optimizer/fsrs_optimizer.py#L372-L373

https://github.com/open-spaced-repetition/fsrs-optimizer/blob/95694b787bb71ac9883db1201af09e334ee5ee0b/src/fsrs_optimizer/fsrs_optimizer.py#L381

https://github.com/open-spaced-repetition/fsrs-optimizer/blob/95694b787bb71ac9883db1201af09e334ee5ee0b/src/fsrs_optimizer/fsrs_optimizer.py#L398-L398

[L-M-Sherlock]

I want to keep consistency between the Python version and Rust version. Here is the schema:

Column Name	Data Type	Description	Constraints
review_state	integer	The state of the card at the time of review. This describes the learning phase of the card	Optional, Values: {0 (New), 1 (Learning), 2 (Review), 3 (Relearning)}

https://github.com/open-spaced-repetition/fsrs-optimizer#review-logs-schema

[dae]

Won't the current Rust code (the second line I showed) always fail, because review_kind will never be 0 after the addition? What if instead of changing the numbers, you used constants like "Review" instead of comparing numbers? Then it should be clear in both.

[dae]

Eg const Review = 1; if review_kind == Review { ... }

[dae]

Closed by #38