chore: Go Sql Online Store

[kevjumba]

What this PR does / why we need it:
Implements a basic sql online store that we can use to perform integration testing within Go instead of requiring entry point to be through Python. (This makes it quite hard to do Go specific testing). Also contains integration tests for the sql online store and the getonlinefeatures grpc service to make sure the correct data is retrieved.

Which issue(s) this PR fixes:

Fixes #

[woop]

I dont agree with the idea of rewriting online stores into Go. We will end up duplicating a lot of logic that should only live in Python in my opinion. What is the motivation behind this change?

[achals]

This should probably be guarded by a lock to prevent concurrent access creating multiple dangling connections

[kevjumba]

added a mutex.

[kevjumba]

I dont agree with the idea of rewriting online stores into Go. We will end up duplicating a lot of logic that should only live in Python in my opinion. What is the motivation behind this change?

Adding @pyalex and @achals. Oleksii raised the concern of difficulties in writing certain tests for the Go feature server specifically and we talked about it and decided to use sqlite as a path as a simple testing backbone to do certain integration/unit testing.

[woop]

I dont agree with the idea of rewriting online stores into Go. We will end up duplicating a lot of logic that should only live in Python in my opinion. What is the motivation behind this change?

Adding @pyalex and @achals. Oleksii raised the concern of difficulties in writing certain tests for the Go feature server specifically and we talked about it and decided to use sqlite as a path as a simple testing backbone to do certain integration/unit testing.

Can you be more specific on the types of testing that isn't possible? Is the idea that we would reimplement all these update/destroy methods from the online store interface and make the Go implementation more than a reader/writer?

[achals]

I dont agree with the idea of rewriting online stores into Go. We will end up duplicating a lot of logic that should only live in Python in my opinion. What is the motivation behind this change?

Adding @pyalex and @achals. Oleksii raised the concern of difficulties in writing certain tests for the Go feature server specifically and we talked about it and decided to use sqlite as a path as a simple testing backbone to do certain integration/unit testing.

Can you be more specific on the types of testing that isn't possible? Is the idea that we would reimplement all these update/destroy methods from the online store interface and make the Go implementation more than a reader/writer?

I believe the idea was to have only the read path implemented for SQLite to make it easy to run a full test in pure go using a prebuilt SQLite file without crossing the boundary from python to go, or spinning up redis and writing to it before testing the go code that read it.

I agree that the update method should not be included in this pr, unless I'm missing something.

[codecov-commenter]

Codecov Report

Merging #2446 (941aea8) into master (ad5694e) will decrease coverage by 0.01%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #2446      +/-   ##
==========================================
- Coverage   84.94%   84.92%   -0.02%     
==========================================
  Files         127      127              
  Lines       10841    10841              
==========================================
- Hits         9209     9207       -2     
- Misses       1632     1634       +2     

Flag	Coverage Δ
integrationtests	75.23% <ø> (-0.24%)	⬇️
unittests	58.45% <ø> (ø)

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

Impacted Files	Coverage Δ
...ython/feast/embedded_go/online_features_service.py	95.08% <0.00%> (-1.64%)	⬇️
.../integration/online_store/test_universal_online.py	93.59% <0.00%> (-0.28%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ad5694e...941aea8. Read the comment docs.

[achals]

/lgtm

[feast-ci-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: achals, kevjumba

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [achals,kevjumba]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment