Code inspector issue for License__c SObject

[Eric Kintzer]

‌

Scott – weird stuff here after updating to the the last two IC2 releases.

I have an object called License__c that you can perform operations on - e.g. reference fields and methods such as shown in the image. Code compiles fine, executes fine. But the code inspector thinks that license is not an SObject and displays the method name or field name in red

Initially, I did a offline symbol table generation of Sobjects only - this did not resolve the issue
So, I did a full offline symbol table regeneration and the issue was resolved.

I only bring this to your attention in case there is something about objects called License that might be confusing the code inspector when you release new versions

Eric

[Scott Wells]

Hi, Eric. I'm not seeing that locally. I just modeled an empty custom object called License__c, regenerated the project's OST, and added basically the same code to an anonymous Apex tab with no reported errors:

Is this issue reproducible in, say, a brand new scratch org project with such a dummy License__c object? Do you see the getPopulatedFieldsAsMap() method in the License__c stub SObject type in the OST? What happens if you regenerate the OST?

[Eric Kintzer]

when I regenerated the OST (full), the issue went away. I only noticed this problem after updating to latest two releases of IC2 whereas before the updates, the code inspector never had an issue with License__c (an sObject that has been around for more than 10 years and my IC2 project has been active for 4+ years).

I leave it to you to decide if there’s some deeper issue or this falls into the “regen the OST” when you see stuff like this. I’m happy if you close this issue.

[Scott Wells]

Yeah, I'm not really sure. I know that a couple of folks saw some weird behavior with the 2024.1 IDE update where caches/indices needed to be rebuilt for things to be in a proper state...and not just for IC2 but also for their own first-class languages. I didn't see that for my main IDEs, but I did see it on a couple of my test instances.

It's also possible that there's a lingering issue with IC2's caches/indices where things can get stale, though I'd need a (semi-)reproducible test case to be able to nail something like that down.

I'll resolve this for now since it's "fixed" for you, but if you see it happen again -- and particularly if you can identify a pattern -- please let me know.