Prometheus: Fuzzy search for metric names in Code Mode (#85396)

* perf: limit metric names in Code Mode suggestions

* feat: configurable metric names limit

* feat: code mode autocomplete info/disclaimer

* chore: put new functionality behind new feature toggle

* refactor: avoid type assertions

* refactor: avoid explicit `any`

refactor: type guards

* refactor: type guards

* chore: add testdata results

* fix: add missing feature toggle guard

* perf: prefer array access to `Array.prototype.at`

* test: add missing config override

* test: refactor for brevity & clarity

* perf: avoid unnecessary mapping

* chore: undo testdata changes

* fix: use correct limit; perf optimizations

* refactor: avoid unnecessary `async`s

* types: simplify

* test: add missing tests

* fix: avoid hardcoding

* test: update mock path

* docs: fix typo

style: remove formatting artifact

style: remove formatting artifact

style: remove formatting artifact

* fix: event scope regression

* style: refactor for clarity

* refactor: prefer `useCallback` to in-effect handler

* refactor: simplify & broaden `filter`

* refactor: rename file to keep with conventions

* chore: mirror Prometheus package changes in app

* refactor: prefer no `@ts-ignore`

* chore: update betterer results

* docs: use type in TSDoc `@link` without `@ts-ignore`

* test: add missing provider

* test: fix jest mock path

* fix: display disclaimer in empty input case
This commit is contained in:
Nick Richmond
2024-04-04 16:38:23 -04:00
committed by GitHub
parent 0d7834c7f7
commit 559fab9dc6
32 changed files with 1292 additions and 132 deletions

View File

@@ -153,6 +153,7 @@ Experimental features might be changed or removed without prior notice.
| `kubernetesSnapshots` | Routes snapshot requests from /api to the /apis endpoint |
| `cachingOptimizeSerializationMemoryUsage` | If enabled, the caching backend gradually serializes query responses for the cache, comparing against the configured `[caching]max_value_mb` value as it goes. This can can help prevent Grafana from running out of memory while attempting to cache very large query responses. |
| `prometheusPromQAIL` | Prometheus and AI/ML to assist users in creating a query |
| `prometheusCodeModeMetricNamesSearch` | Enables search for metric names in Code Mode, to improve performance when working with an enormous number of metric names |
| `alertmanagerRemoteSecondary` | Enable Grafana to sync configuration and state with a remote Alertmanager. |
| `alertmanagerRemotePrimary` | Enable Grafana to have a remote Alertmanager instance as the primary Alertmanager. |
| `alertmanagerRemoteOnly` | Disable the internal Alertmanager and only use the external one defined. |