Should we not have that try-catch statement? It allows for silent failures like this, if something happens in the future. I'm happy to make that change, if you think this is a good idea

I'm not sure what you mean by try-catch here. I think it would be reasonable to throw an error, but I'd do it via either a quick .get instead of [] or by checking for an null return and then throwing.

@directionless Sorry, not try-catch... try-except statement. See screenshot below

Basically, I think the lack of linwin as a key is a proximate cause for this issue, but imo the root cause is that we are try-excepting the "No Key" error and then automatically assuming that if there's no key, then the table is available for all platforms. I think that's how this linwin thing slipped through.

I don't know the history/context of why we are letting this error go through, but at first glance, I would think that we should let it fail so that someone will catch it and fix it. Glad to make that change if the team agrees.

Oh! I see what you mean. Yeah, I'm familiar with the pattern, but I misunderstood your comment -- I didn't realize we already had one.

I would bet it's there so that the all case can be specified with a lack of of designation. I doubt there's a deeper reason, and I'm in favor of changing it. I agree its error prone. It seems reasonable to remove the try and fix whatever breaks.

@directionless ok! I removed the catch statement. I ran the script twice: once as is and once with linwin commented out. It works as is and fails with KeyError when linwin is greyed out, which I think is expected behavior 👍