bug#46331: provided-mode-derived-p doesn't work on aliases

bug#46331: provided-mode-derived-p doesn't work on aliases

[Reuben Thomas via Bug reports for GNU Emacs, the Swiss army knife of text editors]

[-- Attachment #1: Type: text/plain, Size: 338 bytes --]

javascript-mode is an alias for js-mode, and some third-party modes (e.g.
json-mode) derive from it.

(provided-mode-derived-p 'js-mode 'prog-mode) == t

but

(provided-mode-derived-p 'javascript-mode 'prog-mode) == nil

Hence in json-mode, (derived-mode-p 'prog-mode) evaluates to nil, which
seems unfortunate.

-- 
https://rrt.sc3d.org

[-- Attachment #2: Type: text/html, Size: 1492 bytes --]

bug#46331: provided-mode-derived-p doesn't work on aliases

[Lars Ingebrigtsen]

Reuben Thomas <rrt@sc3d.org> writes:

> javascript-mode is an alias for js-mode, and some third-party modes (e.g.
> json-mode) derive from it.
>
> (provided-mode-derived-p 'js-mode 'prog-mode) == t
>
> but
>
> (provided-mode-derived-p 'javascript-mode 'prog-mode) == nil
>
> Hence in json-mode, (derived-mode-p 'prog-mode) evaluates to nil, which seems
> unfortunate.

Yup.  I've now changed the function to also work on MODE parameters that
are function aliases.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#46331: provided-mode-derived-p doesn't work on aliases

[Reuben Thomas via Bug reports for GNU Emacs, the Swiss army knife of text editors]

[-- Attachment #1: Type: text/plain, Size: 610 bytes --]

On Sat, 6 Feb 2021 at 11:04, Lars Ingebrigtsen <larsi@gnus.org> wrote:

> Reuben Thomas <rrt@sc3d.org> writes:
>
> > javascript-mode is an alias for js-mode, and some third-party modes (e.g.
> > json-mode) derive from it.
> >
> > (provided-mode-derived-p 'js-mode 'prog-mode) == t
> >
> > but
> >
> > (provided-mode-derived-p 'javascript-mode 'prog-mode) == nil
> >
> > Hence in json-mode, (derived-mode-p 'prog-mode) evaluates to nil, which
> seems
> > unfortunate.
>
> Yup.  I've now changed the function to also work on MODE parameters that
> are function aliases.
>

Many thanks!

-- 
https://rrt.sc3d.org

[-- Attachment #2: Type: text/html, Size: 1337 bytes --]

bug#46331: provided-mode-derived-p doesn't work on aliases

[Reuben Thomas via Bug reports for GNU Emacs, the Swiss army knife of text editors]

[-- Attachment #1: Type: text/plain, Size: 226 bytes --]

I just looked at the history of provided-mode-derived-p, and it seems that
a fix had already been made for this problem. See commit 48ff4c0b2f7. (I am
using emacs 26.3, which does not have this fix.)

-- 
https://rrt.sc3d.org

[-- Attachment #2: Type: text/html, Size: 502 bytes --]

bug#46331: provided-mode-derived-p doesn't work on aliases

[Reuben Thomas via Bug reports for GNU Emacs, the Swiss army knife of text editors]

[-- Attachment #1: Type: text/plain, Size: 397 bytes --]

On Sat, 6 Feb 2021 at 11:27, Reuben Thomas <rrt@sc3d.org> wrote:

> I just looked at the history of provided-mode-derived-p, and it seems that
> a fix had already been made for this problem. See commit 48ff4c0b2f7. (I am
> using emacs 26.3, which does not have this fix.)
>

Ah, sorry, the earlier fix is for when one of MODES is an alias, not for
when MODE is an alias!

-- 
https://rrt.sc3d.org

[-- Attachment #2: Type: text/html, Size: 1020 bytes --]

bug#46331: provided-mode-derived-p doesn't work on aliases

[Reuben Thomas via Bug reports for GNU Emacs, the Swiss army knife of text editors]

[-- Attachment #1: Type: text/plain, Size: 216 bytes --]

There is one remaining nit: the docstring of provided-mode-derived-p was
updated by that earlier commit to say "MODES or their aliases"; presumably
a similar change should be made to the docstring of derived-mode-p?

[-- Attachment #2: Type: text/html, Size: 344 bytes --]

bug#46331: provided-mode-derived-p doesn't work on aliases

[Lars Ingebrigtsen]

Reuben Thomas <rrt@sc3d.org> writes:

> I just looked at the history of provided-mode-derived-p, and it seems that a fix had
> already been made for this problem. See commit 48ff4c0b2f7. (I am using emacs
> 26.3, which does not have this fix.)

No, that fix was for the opposite problem.  :-)  It allows checking
parents that are aliases, but does nothing for checking a child that's an
alias.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#46331: provided-mode-derived-p doesn't work on aliases

[Lars Ingebrigtsen]

Reuben Thomas <rrt@sc3d.org> writes:

> There is one remaining nit: the docstring of provided-mode-derived-p
> was updated by that earlier commit to say "MODES or their aliases";
> presumably a similar change should be made to the docstring of
> derived-mode-p?

I think adding "MODES or their aliases" was a bit superfluous -- I think
it should be self-evident that that's what's meant here.  I didn't
remove that bit from the doc string, but I think adding it to
derived-mode-p isn't all that helpful.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#46331: provided-mode-derived-p doesn't work on aliases

[Reuben Thomas via Bug reports for GNU Emacs, the Swiss army knife of text editors]

[-- Attachment #1: Type: text/plain, Size: 542 bytes --]

On Sat, 6 Feb 2021 at 11:39, Lars Ingebrigtsen <larsi@gnus.org> wrote:

>
> I think adding "MODES or their aliases" was a bit superfluous -- I think
> it should be self-evident that that's what's meant here.  I didn't
> remove that bit from the doc string, but I think adding it to
> derived-mode-p isn't all that helpful.
>

I agree, but I think it should be consistent between the two functions,
otherwise it looks as though they work differently. (If one is reading only
the docstrings, not looking at the code.)

-- 
https://rrt.sc3d.org

[-- Attachment #2: Type: text/html, Size: 1143 bytes --]

bug#46331: provided-mode-derived-p doesn't work on aliases

[Lars Ingebrigtsen]

Reuben Thomas <rrt@sc3d.org> writes:

> On Sat, 6 Feb 2021 at 11:39, Lars Ingebrigtsen <larsi@gnus.org> wrote:
>
>  I think adding "MODES or their aliases" was a bit superfluous -- I think
>  it should be self-evident that that's what's meant here.  I didn't
>  remove that bit from the doc string, but I think adding it to
>  derived-mode-p isn't all that helpful.
>
> I agree, but I think it should be consistent between the two functions, otherwise it
> looks as though they work differently. (If one is reading only the docstrings, not
> looking at the code.)

That's true.  So I've now removed the "or their aliases" from the doc
string.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no