feat(issues): better top issue tooling #103984
Conversation
cvxluo
changed the base branch from
master
to
cvxluo/top-issue-debug-functions
github-actions
bot
added
the
Scope: Frontend
cvxluo
changed the title
| const BATCH_SIZE = 50; | ||
|
|
||
| /** Fetches issue stats for clusters in batches to avoid API limits. */ | ||
| function useBatchClusterStats(clusters: ClusterSummary[]): { |
There was a problem hiding this comment.
might be good to split this hook out into a separate file for readability
There was a problem hiding this comment.
good idea — i'm going to actually split this PR up a bit as well, so this will live in a separate file
| </Flex> | ||
|
|
||
| <ClusterStatsBar> | ||
| {cluster.fixability_score && ( |
There was a problem hiding this comment.
not sure if it would make sense for the fixability score to be 0 but maybe safer to check for nullness here instead of general falsiness?
cvxluo
force-pushed
the
cvxluo/better-top-issue-tooling
branch
from
1e40ad9 to
bb07727
Compare
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
| {t('confidence')} | ||
| </Text> | ||
| </StatItem> | ||
| )} |
There was a problem hiding this comment.
Bug: Fixability score zero incorrectly hidden from display
The condition {cluster.fixability_score && ( uses a truthiness check that treats 0 as falsy, preventing the fixability score component from rendering when the score is exactly 0. Since fixability_score is number | null, a value of 0 represents a valid 0% confidence score that should be displayed. Use a nullness check (!= null or !== null) instead of truthiness to distinguish between the missing value (null) and a valid zero score.
Add custom sorting logic so we can sort by cluster recency, # of events, fixability, and # of issues, or any combination of these factors.