Re: Why does pgindent's README say to download typedefs.list from the buildfarm?

I wrote:
> This works for me. One point that could stand discussion while we're
> here is whether the once-a-cycle run should use the verbatim buildfarm
> results or it's okay to editorialize on that typedefs list. I did a
> little of the latter in da256a4a7, and I feel like we should either
> bless that practice in this document or decide that it was a bad idea.

> For reference, what I added to the buildfarm's list was

> +InjectionPointCacheEntry
> +InjectionPointCondition
> +InjectionPointConditionType
> +InjectionPointEntry
> +InjectionPointSharedState
> +NotificationHash
> +ReadBuffersFlags
> +ResourceOwnerData
> +WaitEventExtension
> +WalSyncMethod

I realized that the reason the InjectionPoint typedefs were missing
is that none of the buildfarm animals that contribute typedefs are
building with --enable-injection-points. I rectified that on sifaka,
and now those are in the list available from the buildfarm.

As for the remainder, they aren't showing up because no variable
or field is declared using them, which means no debug symbol
table entry is made for them. This means we could just drop those
typedefs and be little the worse off notationally. I experimented
with a patch for that, as attached. (In the case of NotificationHash,
I thought it better to arrange for there to be a suitable variable;
but it could certainly be done the other way too.) Is this too anal?

regards, tom lane