unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
* Re: master 5044c19 2/2: project.el: A project has only one main root now
       [not found] ` <20200523015942.848612043D@vcs0.savannah.gnu.org>
@ 2020-05-25 17:17   ` João Távora
  2020-05-25 19:06     ` Dmitry Gutov
  0 siblings, 1 reply; 7+ messages in thread
From: João Távora @ 2020-05-25 17:17 UTC (permalink / raw)
  To: emacs-devel; +Cc: Dmitry Gutov

dgutov@yandex.ru (Dmitry Gutov) writes:

> branch: master
> commit 5044c19001fe608f2eac621add2e05cbca6c804b
> Author: Dmitry Gutov <dgutov@yandex.ru>
> Commit: Dmitry Gutov <dgutov@yandex.ru>
>
>     project.el: A project has only one main root now
>     
>     Practice shows that the vast majority of projects only use one main
>     root.  The users of this API very often make this assumption as well.
>     The rest of the "roots" should be possible to express through
>     project-external-roots.

I welcome this change, it makes perfect sense and simplifies a bit of of
Eglot. So thanks!

However, it is breaking Eglot's tests.  Why?  Well Eglot runs its tests
against master and has recently decided to make every warning a test
failure (and project-roots is an obsoleted variable).

Perhaps Eglot should do neighter of those things, but perhaps if you
bumped the version in project.el to 0.3.0 then Eglot could just start
relying on that.  Or aren't you confident enough in this new API?

João



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: master 5044c19 2/2: project.el: A project has only one main root now
  2020-05-25 17:17   ` master 5044c19 2/2: project.el: A project has only one main root now João Távora
@ 2020-05-25 19:06     ` Dmitry Gutov
  2020-05-25 19:52       ` João Távora
  0 siblings, 1 reply; 7+ messages in thread
From: Dmitry Gutov @ 2020-05-25 19:06 UTC (permalink / raw)
  To: João Távora, emacs-devel

On 25.05.2020 20:17, João Távora wrote:
> However, it is breaking Eglot's tests.  Why?  Well Eglot runs its tests
> against master and has recently decided to make every warning a test
> failure (and project-roots is an obsoleted variable).

I like this idea on paper, but pretty sure you're going to have this 
situation again. Some obsolete functions will remain outside of core 
packages, right? And yet necessary to support older Emacs releases.

> Perhaps Eglot should do neighter of those things, but perhaps if you
> bumped the version in project.el to 0.3.0 then Eglot could just start
> relying on that.  Or aren't you confident enough in this new API?

I'll bump it now. There will be other breakages in the future, though. ;)



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: master 5044c19 2/2: project.el: A project has only one main root now
  2020-05-25 19:06     ` Dmitry Gutov
@ 2020-05-25 19:52       ` João Távora
  2020-05-25 20:00         ` Dmitry Gutov
  0 siblings, 1 reply; 7+ messages in thread
From: João Távora @ 2020-05-25 19:52 UTC (permalink / raw)
  To: Dmitry Gutov; +Cc: emacs-devel

Dmitry Gutov <dgutov@yandex.ru> writes:

> On 25.05.2020 20:17, João Távora wrote:
> I like this idea on paper, but pretty sure you're going to have this
> situation again. Some obsolete functions will remain outside of core 
> packages, right? And yet necessary to support older Emacs releases.

Yes, I'm not particularly convinced by the idea myself.  Anyway when
Eglot goes in the core that problem will be gone.  At any rate, the
"warning means error" will be a policy of Emacs tests, not Eglot's.
eglot-tests.el will be an abiding citizen of test/lisp like every other
one.

>> Perhaps Eglot should do neighter of those things, but perhaps if you
>> bumped the version in project.el to 0.3.0 then Eglot could just start
>> relying on that.  Or aren't you confident enough in this new API?
>
> I'll bump it now. There will be other breakages in the future,
> though. ;)

Thanks,
João



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: master 5044c19 2/2: project.el: A project has only one main root now
  2020-05-25 19:52       ` João Távora
@ 2020-05-25 20:00         ` Dmitry Gutov
  2020-05-25 20:24           ` João Távora
  0 siblings, 1 reply; 7+ messages in thread
From: Dmitry Gutov @ 2020-05-25 20:00 UTC (permalink / raw)
  To: João Távora; +Cc: emacs-devel

On 25.05.2020 22:52, João Távora wrote:
> Anyway when
> Eglot goes in the core that problem will be gone.

It won't support previous versions of Emacs?



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: master 5044c19 2/2: project.el: A project has only one main root now
  2020-05-25 20:00         ` Dmitry Gutov
@ 2020-05-25 20:24           ` João Távora
  2020-05-25 21:52             ` Dmitry Gutov
  0 siblings, 1 reply; 7+ messages in thread
From: João Távora @ 2020-05-25 20:24 UTC (permalink / raw)
  To: Dmitry Gutov; +Cc: emacs-devel

On Mon, May 25, 2020 at 9:00 PM Dmitry Gutov <dgutov@yandex.ru> wrote:

> On 25.05.2020 22:52, João Távora wrote:
> > Anyway when
> > Eglot goes in the core that problem will be gone.
>
> It won't support previous versions of Emacs?

Of course it will, but it will not be tested against them, as in `make check`.
Unless I'm mistaken. jsonrpc.el support Emacs 25. Are we testing against it?

Even if it is, which I doubt, and even if the current policy is
"error on warn", which I also doubt, we could change that
for old Emacs versions.

João



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: master 5044c19 2/2: project.el: A project has only one main root now
  2020-05-25 20:24           ` João Távora
@ 2020-05-25 21:52             ` Dmitry Gutov
  2020-05-25 22:01               ` João Távora
  0 siblings, 1 reply; 7+ messages in thread
From: Dmitry Gutov @ 2020-05-25 21:52 UTC (permalink / raw)
  To: João Távora; +Cc: emacs-devel

On 25.05.2020 23:24, João Távora wrote:
> Of course it will, but it will not be tested against them, as in `make check`.

Most of the time, when a package has to use an "obsolete" function, is 
because the "new" one is not yet available in older versions of Emacs 
that it still wants to support.



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: master 5044c19 2/2: project.el: A project has only one main root now
  2020-05-25 21:52             ` Dmitry Gutov
@ 2020-05-25 22:01               ` João Távora
  0 siblings, 0 replies; 7+ messages in thread
From: João Távora @ 2020-05-25 22:01 UTC (permalink / raw)
  To: Dmitry Gutov; +Cc: emacs-devel

On Mon, May 25, 2020 at 10:52 PM Dmitry Gutov <dgutov@yandex.ru> wrote:
>
> On 25.05.2020 23:24, João Távora wrote:
> > Of course it will, but it will not be tested against them, as in `make check`.
>
> Most of the time, when a package has to use an "obsolete" function, is
> because the "new" one is not yet available in older versions of Emacs
> that it still wants to support.

But in this case, :core packages take care of that nicely.  For other cases,
(no :core packages),  I see what you mean.  I doubt we're using that
"error on warn" policy in Emacs tests.

João



^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2020-05-25 22:01 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <20200523015940.19021.91647@vcs0.savannah.gnu.org>
     [not found] ` <20200523015942.848612043D@vcs0.savannah.gnu.org>
2020-05-25 17:17   ` master 5044c19 2/2: project.el: A project has only one main root now João Távora
2020-05-25 19:06     ` Dmitry Gutov
2020-05-25 19:52       ` João Távora
2020-05-25 20:00         ` Dmitry Gutov
2020-05-25 20:24           ` João Távora
2020-05-25 21:52             ` Dmitry Gutov
2020-05-25 22:01               ` João Távora

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).