fix saving new sql lab queries

[suddjian]

CATEGORY

Choose one

• Bug Fix
• Enhancement (new features, refinement)
• Refactor
• Add tests
• Build / Development Environment
• Documentation

SUMMARY

This fixes an issue introduced in #8298 as reported by @graceguo-supercat. Due to a lack of FAB know-how on my part, the id field was required when POSTing a saved query, which prevents any new saved queries from being created.

BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF

TEST PLAN

Tested locally

ADDITIONAL INFORMATION

• Has associated issue:
• Changes UI
• Requires DB Migration.
• Confirm DB Migration upgrade and downgrade tested.
• Introduces new feature or API
• Removes existing feature or API

REVIEWERS

[graceguo-supercat]

@suddjian Save query feature works good. but i have another question: it seems query's label is not carried-over to the Save Query modal, please see my screenshot.

I saw this problem in the last commit (so it's not introduced from this PR). it will be great improvement if you can have this fixed as well.

[suddjian]

That's certainly an interesting bug, I'll see what I can do

[suddjian]

@graceguo-supercat I've done some investigation, and this bug seems to be connected to a larger issue where null values are being returned from the backend as "None".

I have pinned down exactly where it affects this part of the Saved Query feature: In the response to the request GET savedqueryviewapi/api/get/{id}, the query.description is set to "None". The SaveQuery component is being provided with either the query.description or query.title, so it displays the description which is "None".

I'm not sure how to best deal with this problem, though I have a couple very hack-y ideas. It should probably be reported as its own bug if it isn't already, as it's probably affecting other parts of the app that should be receiving nulls.

[graceguo-supercat]

fair. I agree to report the missing label as a separate issue.

[graceguo-supercat]

LGTM

[rubypollev]

Still seeing this issue in 0.35.0. Anyone else?

[suddjian]

@rubypollev are you referring to the label not being carried over? I don't think there's been an attempt to fix that, yet.

[rubypollev]

Ah, okay! Is there an issue I can subscribe to @suddjian?