add linwin to list of supported PLATFORM_DIRS

[zhumo]

Problem

The two tables intel_me_info and secureboot are being marked as available for all platforms, when it should only be on linux and windows ("linwin"). It is correctly marked as linwin, but the documentation is not generating correctly.

This is because the genwebsitejson.py file does not contemplate linwin, so it just assumes every platform is supported.

This PR adds a linwin key to the PLATFORM_DIRS constant so that it will generate the correct tagging in the JSON.

For broader consideration

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.

[directionless]

Thanks for digging into this!

I didn't test your PR, but the diff looks fine.

[directionless]

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.

[zhumo]

@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.

[directionless]

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.

[zhumo]

@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 👍