Sebastian Wilzbach

> I will let @wilzbach decide. I would go with `experimental` because they are then at least partially exposed whereas `package` functions are entirely hidden from the user.

> You mean an internal function. say tryParse(), marked with package inside std.experimental.typecons? Why `stdx.typecons` and not `std.experimental.conv`?

Finally our CIs are giving sensible error messages :) ``` std/meta.d(86:8)[warn]: Public declaration 'Quote' has no documented example. [Inferior 1 (process 2276) exited with code 01] ```

> @wilzbach It'd better if the linter can output all the warnings at once. This message is related to new code you pushed - how should the linker have known...

> Yeah, if nobody is actively working on std.data.json, then we shouldn't be advertising it in the Phobos docs. Hmm, how should somebody pick up the torch if they don't...

First things first: I split this PR of into three parts: - `find` (this PR) - `findSplit` (#5968) - the typo + removed `std.stdio` logging (#5969) > On one hand,...

> I wish we just closed this. It created just work for everyone involved, and is liable to create more. Would you mind elaborating on this a bit? `find(a, 2,...

> Based on the commit message, it seems to be a problem related to CTFE and casting. I did do some looking over Tuple to try and understand it when...

> we do so by reusing the unnamed Tuple as layout for the named Tuple. That's already the case today - even for named tuples the layout is as follows:...

> we need to do this: > Tuple!(int, "x", int, "y") should have as its only state Tuple!(int, int). In the form of a data member that is. Call that...