Default source select

[v3nd3774]

teamwork with @kmeng01 to address:

with:

Try it out on the dev copy here.

[kmeng01]

also contains the time series query as requested by @YapheeetS

[824zzy]

When I click up-arrow in the left side panel, it didn't jump to county level and show the data.

Can you check which part of your codes cause this bug?

[v3nd3774]

@824zzy sure thing! Good catch.

[damianj]

Selecting a default source other than JHU causes counties to show undefined. For context this is what the settings look like:

[kmeng01]

@damianj Yes, because there is only county-level data from JHU. What would you like to be displayed? I can modify the API to return it

[damianj]

Right, but I'm only changing the state source not the county source. So shouldn't there still be values displayed there?

[kmeng01]

Hmm, that seems like a frontend issue. maybe @v3nd3774 can offer more insight

[kmeng01]

On second thought, this actually involves both frontend and API. We have to allow the query to specify the root default source and child default source, since the Texas default source is different from the county one.

Do we want this to be in this PR or another one? @damianj @v3nd3774

[v3nd3774]

So it may be related to the data but I'm not certain yet. I pushed the latest data files that were already present on the server to git so I could use the same data that the server is using and I was able to reproduce the error in this screenshot:

Is it the same error visible at the query: https://idir.uta.edu/covid-19-api-dev-2/api/v1/statquery?node=Gila%20County&date=2020-04-11&dsrc=JHU?

[v3nd3774]

@damianj I think the JHU source is still in the payload to the api as the query from the website currently shows dsrc=JHU.

[kmeng01]

The problem is that the API assumes that the parent and child node use the same default source. But because we added the settings, that assumption is no longer valid

[kmeng01]

I'll have to modify the API on the backend to accept both dsrc_parent and dsrc_child so that they can be differentiated. And then teh frontend must be edited to support that query

[kmeng01]

Where should we make this change?

[v3nd3774]

As you've suggested

[v3nd3774]

If you change it to accept those parameters I can change the frontend to send them.

[kmeng01]

Sure. Which branch?

[v3nd3774]

Same set of acceptable string values for each of those parameters right?

You can just branch off and I'll pull your changes when you're done 👍

[kmeng01]

Wait I'm not that familiar with some parts of git lol Can I just make the changes on default_source_select?

[kmeng01]

And yeah, same set of string values

[v3nd3774]

@kmeng01 yeah that sounds ok too

[v3nd3774]

Ok @damianj I think that the changes we've made address the bug you've highlighted.

I tried it with the settings you had could you confirm if it's been addressed?

[v3nd3774]

@824zzy I've made some changes to populate the parent variable in showPlace when a county is selected from the panel. I think it then flows through the code you had previously created the same way.

It results in addressing the bug you mentioned, could you confirm this is the case or let me know of a counter-example?

I tested it out below:

The changes should be visible at the dev endpoint.

[damianj]

That issue is fixed. Another issue I just noticed is at the county level when switching a state source or even leaving all on JHU, clicking "Apply changes" causes the following This then turns into

[v3nd3774]

@damianj ok, I think it's been addressed could you take a look and confirm/deny?

[damianj]

lgtm

[v3nd3774]

lgtm

In that case, could you merge it into master here on git?

[damianj]

Okay, Joey said your branch and his might have conflicts should we get those resolved first? :+1:

[v3nd3774]

@damianj @kmeng01 @824zzy thanks for the hard work, the combined changes from @kmeng01 , @824zzy and myself are on the dev version. Does it look ready to merge to master?

[824zzy]

LGTM~