Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Pedro Andres Aranda Gutierrez]

[-- Attachment #1: Type: text/plain, Size: 595 bytes --]

Hi

reading removed FIXME in the second hunk:

# FIXME: Don't install the dump file into the app bundle when
# self-contained install is disabled.

I understand that we should install if self-contained is enabled.

However, I see

+ifeq (${ns_self_contained},no)

Shouldn't that read as

+ifeq (${ns_self_contained},yes)

I'm confused...

/PA
-- 
Fragen sind nicht da, um beantwortet zu werden,
Fragen sind da um gestellt zu werden
Georg Kreisler

Headaches with a Juju log:
unit-basic-16: 09:17:36 WARNING juju.worker.uniter.operation we should run
a leader-deposed hook here, but we can't yet

[-- Attachment #2: Type: text/html, Size: 3016 bytes --]

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Gerd Möllmann]

On 23-09-21 8:29 , Pedro Andres Aranda Gutierrez wrote:
> Hi
> 
> reading removed FIXME in the second hunk:
> 
> # FIXME: Don't install the dump file into the app bundle when
> # self-contained install is disabled.
> 
> I understand that we should install if self-contained is enabled.
> 
> However, I see
> 
> +ifeq (${ns_self_contained},no)
> 
> Shouldn't that read as
> 
> +ifeq (${ns_self_contained},yes)
> 
> I'm confused...

I assume that is nextstep/Makefile.in.  We want to cp the .pdmp file 
only if we don't do a self-contained build (...=no).

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Pedro Andres Aranda Gutierrez]

[-- Attachment #1: Type: text/plain, Size: 1083 bytes --]

Thank you for the clarification!
The FIXME was sort of misleading here. A double negation always is ;-)

Best, /PA

On Thu, 21 Sept 2023 at 10:47, Gerd Möllmann <gerd.moellmann@gmail.com>
wrote:

> On 23-09-21 8:29 , Pedro Andres Aranda Gutierrez wrote:
> > Hi
> >
> > reading removed FIXME in the second hunk:
> >
> > # FIXME: Don't install the dump file into the app bundle when
> > # self-contained install is disabled.
> >
> > I understand that we should install if self-contained is enabled.
> >
> > However, I see
> >
> > +ifeq (${ns_self_contained},no)
> >
> > Shouldn't that read as
> >
> > +ifeq (${ns_self_contained},yes)
> >
> > I'm confused...
>
> I assume that is nextstep/Makefile.in.  We want to cp the .pdmp file
> only if we don't do a self-contained build (...=no).
>
>

-- 
Fragen sind nicht da, um beantwortet zu werden,
Fragen sind da um gestellt zu werden
Georg Kreisler

Headaches with a Juju log:
unit-basic-16: 09:17:36 WARNING juju.worker.uniter.operation we should run
a leader-deposed hook here, but we can't yet

[-- Attachment #2: Type: text/html, Size: 1763 bytes --]

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Gerd Möllmann]

On 23-09-21 11:59 , Pedro Andres Aranda Gutierrez wrote:
> Thank you for the clarification!
> The FIXME was sort of misleading here. A double negation always is ;-)

You're welcome.

Please don't be not untroubled ;-)

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Alan Third]

On Thu, Sep 21, 2023 at 10:46:56AM +0200, Gerd Möllmann wrote:
> On 23-09-21 8:29 , Pedro Andres Aranda Gutierrez wrote:
> > Hi
> > 
> > reading removed FIXME in the second hunk:
> > 
> > # FIXME: Don't install the dump file into the app bundle when
> > # self-contained install is disabled.
> > 
> > I understand that we should install if self-contained is enabled.
> > 
> > However, I see
> > 
> > +ifeq (${ns_self_contained},no)
> > 
> > Shouldn't that read as
> > 
> > +ifeq (${ns_self_contained},yes)
> > 
> > I'm confused...
> 
> I assume that is nextstep/Makefile.in.  We want to cp the .pdmp file only if
> we don't do a self-contained build (...=no).

Now I'm confused!

We only have to install the pdmp file in the app bundle when doing a
self-contained build.

If your other change in Makefile.in already does that, then this cp
isn't needed at all.

-- 
Alan Third

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Gerd Möllmann]

On 23-09-21 18:50 , Alan Third wrote:
> On Thu, Sep 21, 2023 at 10:46:56AM +0200, Gerd Möllmann wrote:
>> On 23-09-21 8:29 , Pedro Andres Aranda Gutierrez wrote:
>>>> I'm confused...
>>
>> I assume that is nextstep/Makefile.in.  We want to cp the .pdmp file only if
>> we don't do a self-contained build (...=no).
> 
> Now I'm confused!

I suspect you're all trying to confuse me :-).

> We only have to install the pdmp file in the app bundle when doing a
> self-contained build.
> 
> If your other change in Makefile.in already does that, then this cp
> isn't needed at all.
> 
My aim was to not change anything in the case the build is not 
self-contained, because I can't test that.

In nextstep/Makefile.in, the cp
- is not done in the self-contained case
- is done for the non-self-contained case
(This cp was formerly done for a naked make at toplevel, not only for a 
make install.  That was the problem resulting in an inconistent Emacs.app.)

In the toplevel Makefile.in
- the install is done in the self-contained case, as a substitute for 
the now missing cp in nextstep/Makefile.in
- nothing is changed for the non-self-contained case
(This happens when we make install.)

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Alan Third]

On Thu, Sep 21, 2023 at 07:06:20PM +0200, Gerd Möllmann wrote:
> On 23-09-21 18:50 , Alan Third wrote:
> > On Thu, Sep 21, 2023 at 10:46:56AM +0200, Gerd Möllmann wrote:
> > > On 23-09-21 8:29 , Pedro Andres Aranda Gutierrez wrote:
> > > > > I'm confused...
> > > 
> > > I assume that is nextstep/Makefile.in.  We want to cp the .pdmp file only if
> > > we don't do a self-contained build (...=no).
> > 
> > Now I'm confused!
> 
> I suspect you're all trying to confuse me :-).
> 
> > We only have to install the pdmp file in the app bundle when doing a
> > self-contained build.
> > 
> > If your other change in Makefile.in already does that, then this cp
> > isn't needed at all.
> > 
> My aim was to not change anything in the case the build is not
> self-contained, because I can't test that.
> 
> In nextstep/Makefile.in, the cp
> - is not done in the self-contained case
> - is done for the non-self-contained case
> (This cp was formerly done for a naked make at toplevel, not only for a make
> install.  That was the problem resulting in an inconistent Emacs.app.)
> 
> In the toplevel Makefile.in
> - the install is done in the self-contained case, as a substitute for the
> now missing cp in nextstep/Makefile.in
> - nothing is changed for the non-self-contained case
> (This happens when we make install.)

OK. As the author of that line of code and the FIXME I would recommend
removing the cp in nextstep/Makefile.in for the non-self-contained
case as well, but I don't think I can prove that it's not going to
break anything as I can't really test it either! :)

I guess I can just remove it myself...
-- 
Alan Third

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Gerd Möllmann]

On 23-09-21 19:29 , Alan Third wrote:
> OK. As the author of that line of code 

Are you sure you're the culprit?  Git blame shows Daniel C. added that 
for pdumper...

Re: Question on https://git.savannah.gnu.org/cgit/emacs.git/commit/?id=d0b1e3647fb4e6d889f9f999388c53daf0e67f0d

[Alan Third]

On Thu, Sep 21, 2023 at 07:58:43PM +0200, Gerd Möllmann wrote:
> On 23-09-21 19:29 , Alan Third wrote:
> > OK. As the author of that line of code
> 
> Are you sure you're the culprit?  Git blame shows Daniel C. added that for
> pdumper...

Alas yes, it's my fault:

https://lists.gnu.org/archive/html/emacs-devel/2018-03/msg00974.html

-- 
Alan Third