* Added proxy support for action types
* Fixed tests
* added rejectUnauthorizedCertificates config setting
* removed slack not used code
* Fixed Slack proxy
* fixed typecheck errors
* Cleanup code
* Fixed slack
* Added unit tests
* added proxy server for test
* Fixed build
* Added functional tests
* fixed due to comments
* Fixed tests and some changes due to comments
* Fixed functional tests
* fixed circular deps
* Added proxy unit test to action type
Our tests shouldn't rely on the DOM structure of Resolver (when its arbitrary) because that will make them brittle. If the user doesn't care about the DOM structure, then neither should our tests.
Note: sometimes the user does care about the DOM structure, and in those cases the tests should as well.
* Add routes
* Add version for use in doc link
* Set up basic router layout + WorkplaceSearchNav
* Update views to account for Layout
* Move version to common folder
* Fix version path
* Remove product button
No longer needed since we have all top-level app links in Kibana as a part of this PR
* You know, for search
* Remove comment
* Remove unused i18n properties from JSON
Tests were failing after removing component:
https://kibana-ci.elastic.co/job/elastic+kibana+pipeline-pull-request/67797/execution/node/382/log/
* Revert button and i18n copy removal
This reverts commit ba0535187e.
* Move Overview out of layout to hide nav
For now, the route for groups was added to avoid having comment out the code. Will add the groups component in a future PR
* Revert layout changes to Overview
Since there is no nav, the padding was missing and the view looked off. Reverting to the 7.9, centered column view
* Remove extra Overview component
Was causing tests to fail
* Revert error state to use EuiPage
Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>
This replaces the use of the old graphql-based `useSource` hook with the new plain JSON `useLogSource` hook.
It also fixes two more problems:
- A rendering problem with the source configuration loading screen and a `setState` race condition in the `useLogSource` hook.
- A non-backwards-compatible change of the `/link-to/:sourceId/logs` route in #61162.
* Initial work
* Expand generic support to alert instances
* Convert index threshold to use generics
* Make fixture alert types use generics
* Make alert instance related types use unknown
* Fix typecheck failures
* Cleanup + add instance generic support to registry.get API
* Shallow clone
* Rename some TS variables
* Fix failing api integration tests
* Change code for easier review and keep more history
* Fix
Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>
* _cluster/state/metadata/* and added a comment to function
* add another comment regarding why we are asking for *
* update jest test
* refactor and clean up use of cluster status to get index state
Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>
* Adding api test for custom link transaction example
* expecting specific fields
* expecting specific fields
Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>
The app can show more than 1 Resolver at a time. Each instance has a unique ID called the `resolverComponentInstanceID`.
When the user interacts with Resolver it will add values to the query string. The query string values will contain the `resolverComponentInstanceID`. This allows each Resolver to keep its state separate. When resolver unmounts it will remove any query string values related to it.
If Resolver's `resolverComponentInstanceID` changes it should remove query string values related to the old instance ID. It does not. This PR fixes that.
Note: I don't know if it was possible for this bug to actually happen. I can't make it happen, but depending on how Resolver is mounted by its consumers it *could*
* [Reporting] Update more Server Types for TaskManager
* remove some task manager references
* more strict
* more strict 2
* simplify
* fix test
* fix test
* routing validation unused types cleanup
* remove more casting in route handlers
* feedback changes
* original comment was fine
Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>
* Update design-specific GH code-owners
Created a `@elastic/stack-design` encompassing all product designers associated with the stack+solutions (Kibana repo contributors, not cloud). The `@elastic/kibana-design` team has been reduced to the actual Kibana design team. Then other individual design teams have been added as code-owners to their respective folders containing SASS changes.
* First few processors of the first batch
- Also refactored options to live in scoped objects to avoid
overriding type (important fix!)
- Have not polished copy or form layout.
* add type to shared imports
* Refactors for repeated fields and added forms
- date_index_name
- dissect
- dot_expander
- drop
Fields refactored:
- Field
- Ignore missing
* Fix broken imports and some other small refactors
* added text editor field and updated pattern and if fields
* Large copy improvements and updates and other small refactors
- Added help text for all fields
- Updated layout so that required fields are always on first
- Replaced circle radio group with a select drop down
* update circle shape type field to select
* Added "long" option for convert type
* fix path import
* fix types and i18n
* add validation for dot expander fix append value to be a combobox
* fix i18n
Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>
Fixes flaky tests in Task Manager and Alerting.
The fix in #73244 was correct, but it missed an edge case which causes the already running task to be rescheduled over and over.
This prevents that edge case which was effecting both TM in general and Alerting specifically.
