[bug] set saved search instance for Discover based on dataset

[kavilla]

Description

Preventing any errors being thrown because of the index pattern not existing in the case of saved search was created with a dataset.

Also fix issue that prevented saved search instance not being set even though loaded properly.

The original fix that was removed fixed a previous issue but subsequent updates and refactors required an update that was uncaught.

Issues Resolved

n/a

Screenshot

Testing the changes

• verified saved search is loading for PPL
• verified reload working for non index pattern
• verified short url working for non index pattern
• verified copy url working for non index pattern

Changelog

• skip

Check List

• All tests pass
  • yarn test:jest
  • yarn test:jest_integration
• New functionality includes testing.
• New functionality has been documented.
• Update CHANGELOG.md
• Commits are signed per the DCO using --signoff

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 13 lines in your changes missing coverage. Please review.

Project coverage is 60.86%. Comparing base (85e0767) to head (a886620).
Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
...ic/application/view_components/utils/use_search.ts	0.00%	9 Missing ⚠️
...ommon/search/search_source/create_search_source.ts	0.00%	1 Missing and 2 partials ⚠️
...lication/utils/state_management/discover_slice.tsx	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8689      +/-   ##
==========================================
- Coverage   60.86%   60.86%   -0.01%     
==========================================
  Files        3793     3793              
  Lines       90456    90462       +6     
  Branches    14204    14207       +3     
==========================================
- Hits        55059    55057       -2     
- Misses      31908    31915       +7     
- Partials     3489     3490       +1     

Flag	Coverage Δ
Linux_1	29.08% <0.00%> (-0.01%)	⬇️
Linux_2	56.39% <0.00%> (-0.01%)	⬇️
Linux_3	37.68% <0.00%> (-0.01%)	⬇️
Linux_4	29.85% <0.00%> (-0.01%)	⬇️
Windows_1	29.09% <0.00%> (-0.01%)	⬇️
Windows_2	56.35% <0.00%> (-0.01%)	⬇️
Windows_3	37.68% <0.00%> (-0.01%)	⬇️
Windows_4	29.85% <0.00%> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ashwin-pc]

We should add a test here for the saved search flow, but can come in as a fast follow.

[virajsanghvi]

Looks like a test related to saved searches is failing?

[virajsanghvi]

just for my knowledge, this just reads from cache and doesn't make a request? If it does make a request, I'm a little confused by the sequence above (call, cache, call) - I'm assuming the cache call is actually what fetches given this usage.

[kavilla]

good question.

for anything that isn't an index pattern, it will read just from the cache. if nothing returns from the cache, then we will recreate the the dataset and save it as an index pattern when we call cache dataset. then we retrieve the actual index pattern we just created.

if it is an index pattern it will check the cache if it doesn't exist in the cache it will make it a request

[virajsanghvi]

Approving PR as test failure looks like test needs to be cleaned up, but can happen separately

[ananzh]

test is fixed here opensearch-project/opensearch-dashboards-functional-test#1601