On Sun, Dec 4, 2022, 04:23 Stefan Monnier <monnier@iro.umontreal.ca> wrote:

> > +Despite the original authors's best efforts,
> > +TRY-COMPLETION-FUNCTION is still a poorly understood
> > +implementation detail.  If you understand what it's for, you
>
> FWIW, I'm not really the original author of try-completion.
> More specifically, the `try-completion` function itself predates my
> involvement with Emacs, and its generalization in
> `partial-completion-mode` as well.  I do take responsability for
> `completion-try-completion`, admittedly, where I tried to give an
> "official" interface to that operation, but I was just trying to
> preserve existing semantics.



Thanks for the insight. But I meant the original author of this function,
of course. I don't have a gripe with try-completion. I'm sure it's doing
useful stuff. I just don't know what to write in this docstring for it to
teach the caller how to use the argument effectively.  Please help me, it's
the last thing that needs to be done, i think.

And of course I'll squash all this silly history afterwards, this is just a
scratch branch.

>
>