bug#74243: [PATCH] Speed up vc-hg-state by treating ignored files as unregistered

[Sean Whitton]

Hello,

On Tue 19 Nov 2024 at 08:28am -05, Spencer Baugh wrote:

> The original motivation of optimizing 'state is to speed up
> vc-refresh-state and vc-after-save, since they call vc-hg-state in
> find-file-hook and after-save-buffer, which adds very noticeable
> latency.  (I care less about the speed of vc-dir, which uses
> 'dir-status-files not 'state)
>
> So those functions would need to pass this argument.  But they (through
> vc-state-refresh) then store the returned state in 'vc-state in the VC
> per-file properties, so anyone accessing that will also see the effect
> of this argument.

I see what you mean.

Quoting vc-state,

| A return of nil from this function means we have no information on the
| status of this file.
| [...]
|  `ignored'          The file showed up in a dir-status listing with a flag
|                     indicating the version-control system is ignoring it,
|                     Note: This property is not set reliably (some VCSes
|                     don't have useful directory-status commands) so assume
|                     that any file with vc-state nil might be ignorable
|                     without VC knowing it.
|
|  `unregistered'     The file is not under version control."
|
|  ;; Note: we usually return nil here for unregistered files anyway
|  ;; when called with only one argument.  This doesn't seem to cause
|  ;; any problems.  But if we wanted to change that, we should
|  ;; probably opt for redefining the `registered' command to return
|  ;; non-nil even for unregistered files (maybe also rename it), and
|  ;; then make sure that all `state' implementations handle
|  ;; unregistered file appropriately.

(I think there's a mistake here: an ignored file is not a file "under
version control", so `unregistered' should say "not under version
control and not ignored".  Would you agree?)

Thanks for pointing out the involvement of find-file-hook and
after-save-hook.  The problem you describe is not at all Hg-specific:
vc-state gets called in a context where speed matters, but it's also the
primary entry point for any code that wants to know the state of a file,
some of which might care more about accuracy than speed.

To put it another way, the code assumes throughout that finding out the
file state will always be fast.  But it also assumes the information is
accurate if present.  This makes me queasy about your original patch.
It does not seem wise to return something we don't know to be true only
on the basis that it all works out fine for now.

The 'nil' return value might provide us with a way out, however.
Could we add an optional argument to vc-state that means "just return
nil if finding out the state properly might be slow"?
Could we make vc-after-save and the relevant find-file-hook entry pass
that option through, and do something sensible with a nil return value?

If they get nil, they would clear out the saved property, and possibly
update the mode line display to "????" or something.  Maybe we'd want a
user option (that could go in your large repo's .dir-locals.el, so it's
set-and-forget) to opt-in to not knowing the file state as often.

-- 
Sean Whitton