SG if you also considered the style block. I suppose the risks of having
this in a feature-specific block is much lower.
On Tue, Dec 7, 2021, 20:14 Sam McCall via Phabricator <
revi...@reviews.llvm.org> wrote:
> sammccall added inline comments.
>
>
> ================
> Comment at: clang-tools-extra/clangd/ConfigFragment.h:271
> + /// Describes hover preferences.
> + struct HoverBlock {
> + /// Whether hover show a.k.a type.
> ----------------
> kadircet wrote:
> > lh123 wrote:
> > > sammccall wrote:
> > > > One question is whether the setting should control hover
> specifically, or whether it covers "in places we print types" more
> generally. But it doesn't seem likely we'll make this configurable for
> diagnostics, and I don't have other examples. Most of our settings are
> per-feature. So I think this is right as it is.
> > > In the future, the AKA type can also be displayed in the signature
> help, but I don't know the best place for this option.（for now, it should
> be fine to put this setting in hover）
> > we've got a `Style` section actually, which might be more suitable for
> extensibility but I am also afraid of pushing ourselves into a corner by
> putting too much meaning into a boolean flag.
> >
> > maybe we should just go with a command line flag until we figure out
> what to do here (as it's less invasive)?
> I thought about style but the idea is that reflects coding style of the
> project, where this is really more of a user preference.
>
> I'm a little skeptical whether we'll want this in sig help/completion (vs
> just being smarter about whether to desugar)
>
> Command line flag is tempting but annoying to wire up. My favorite is
> probably what we have here
>
>
> Repository:
> rG LLVM Github Monorepo
>
> CHANGES SINCE LAST ACTION
> https://reviews.llvm.org/D114665/new/
>
> https://reviews.llvm.org/D114665
>
>
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
