Initialization warnings without compehensive information about problem

Initialization warnings without compehensive information about problem

[Heime]

I do not like that when there is a problem the "*Warnings*" buffer only says

Warning (initialization): An error occurred while loading ‘/home/hagbard/.emacs’:
Symbol's function definition is void: wordscal-ktrigr

Providing little useful information regarding the file, its location and the corresponding
line number where the problem was detected.  Such problems occur frequently when one is using   
a number of initialisation files for one's emacs setup. 

Re: Initialization warnings without compehensive information about problem

[Emanuel Berg]

Heime wrote:

> I do not like that when there is a problem the "*Warnings*"
> buffer only says
>
> Warning (initialization): An error occurred while loading
> ‘/home/hagbard/.emacs’:
> Symbol's function definition is void: wordscal-ktrigr

Byte-compile your Elisp ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Initialization warnings without compehensive information about problem

[Bruno Barbier]

Heime <heimeborgia@protonmail.com> writes:

> I do not like that when there is a problem the "*Warnings*" buffer only says
>
> Warning (initialization): An error occurred while loading ‘/home/hagbard/.emacs’:
> Symbol's function definition is void: wordscal-ktrigr
>
> Providing little useful information regarding the file, its location and the corresponding
> line number where the problem was detected. 

You need to start Emacs with the "--debug-init". That way, it will stop
and give you all the details about the error.

See (info "(emacs) Initial Options")

Bruno

Re: Initialization warnings without compehensive information about problem

[Heime]

------- Original Message -------
On Sunday, November 13th, 2022 at 1:55 PM, Bruno Barbier <brubar.cs@gmail.com> wrote:


> Heime heimeborgia@protonmail.com writes:
> 
> > I do not like that when there is a problem the "Warnings" buffer only says
> > 
> > Warning (initialization): An error occurred while loading ‘/home/hagbard/.emacs’:
> > Symbol's function definition is void: wordscal-ktrigr
> > 
> > Providing little useful information regarding the file, its location and the corresponding
> > line number where the problem was detected.
> 
> 
> You need to start Emacs with the "--debug-init". That way, it will stop
> and give you all the details about the error.
> 
> See (info "(emacs) Initial Options")

It would be more useful for the "*Warnings*" to give good information rather than the generic
.emacs has a problem with a function definition being void.  It is a bother that for any problem
one has to rerun again with "--debug-init".

Re: Initialization warnings without compehensive information about problem

[Bruno Barbier]

Heime <heimeborgia@protonmail.com> writes:

> ------- Original Message -------
> On Sunday, November 13th, 2022 at 1:55 PM, Bruno Barbier <brubar.cs@gmail.com> wrote:
>
> It would be more useful for the "*Warnings*" to give good information rather than the generic
> .emacs has a problem with a function definition being void.  It is a bother that for any problem
> one has to rerun again with "--debug-init".
>  

You just need to always start emacs with '--debug-init', if you really
want Emacs to always immediately stop on any config issue.

Re: Initialization warnings without compehensive information about problem

[Emanuel Berg]

Bruno Barbier wrote:

>> It would be more useful for the "*Warnings*" to give good
>> information rather than the generic .emacs has a problem
>> with a function definition being void. It is a bother that
>> for any problem one has to rerun again with "--debug-init".
>
> You just need to always start emacs with '--debug-init', if
> you really want Emacs to always immediately stop on any
> config issue.

But that's 0.018 seconds slower ...

$ time emacs -f kill-emacs
emacs -f kill-emacs  2.16s user 0.21s system 84% cpu 2.809 total

$ time emacs --debug-init -f kill-emacs
emacs --debug-init -f kill-emacs  2.23s user 0.19s system 85% cpu 2.826 total

-- 
underground experts united
https://dataswamp.org/~incal

Re: Initialization warnings without compehensive information about problem

[Heime]

------- Original Message -------
On Sunday, November 13th, 2022 at 3:22 PM, Emanuel Berg <incal@dataswamp.org> wrote:


> Bruno Barbier wrote:
> 
> > > It would be more useful for the "Warnings" to give good
> > > information rather than the generic .emacs has a problem
> > > with a function definition being void. It is a bother that
> > > for any problem one has to rerun again with "--debug-init".
> > 
> > You just need to always start emacs with '--debug-init', if
> > you really want Emacs to always immediately stop on any
> > config issue.
> 
> 
> But that's 0.018 seconds slower ...
> 
> $ time emacs -f kill-emacs
> emacs -f kill-emacs 2.16s user 0.21s system 84% cpu 2.809 total
> 
> $ time emacs --debug-init -f kill-emacs
> emacs --debug-init -f kill-emacs 2.23s user 0.19s system 85% cpu 2.826 total

I would be far more convenient if the "*Warnings*" could provide a little bit
more information about the problem.  Not at the level of --debug-init or
byte-compilation, some something one can act on.

Re: Initialization warnings without compehensive information about problem

[Bruno Barbier]

Heime <heimeborgia@protonmail.com> writes:

> ------- Original Message -------
> On Sunday, November 13th, 2022 at 3:22 PM, Emanuel Berg <incal@dataswamp.org> wrote:
>
> I would be far more convenient if the "*Warnings*" could provide a little bit
> more information about the problem.  Not at the level of --debug-init or
> byte-compilation, some something one can act on.  

Emacs provides the exact file location and the variable name; I'm not
sure what else I would need.

With the provided information, I would do M-x ffap on the filename to
open it, followed by C-s to find that function name.

Maybe there could be a link to show where that function name occurs, something like:
   (with-current-buffer (find-file "/home/hagbard/.emacs") (occur "wordscal-ktrigr"))



Bruno