add argument to a derived-mode

add argument to a derived-mode

[Joe Riel]

Is there a way to use define-derived-mode to
create a major mode that takes an argument if
the base mode does not?

-- 
Joe Riel

Re: add argument to a derived-mode

[Stefan Monnier]

> Is there a way to use define-derived-mode to
> create a major mode that takes an argument if
> the base mode does not?

Major modes do not take arguments, no.
Instead, you'll want to define 2 functions:
- a major mode
- the function you want, with its argument, and which internally calls
  the major mode.


        Stefan

Re: add argument to a derived-mode

[Joe Riel]

On Fri, 13 Jan 2017 09:37:07 -0500
Stefan Monnier <monnier@iro.umontreal.ca> wrote:

> > Is there a way to use define-derived-mode to
> > create a major mode that takes an argument if
> > the base mode does not?
> 
> Major modes do not take arguments, no.
> Instead, you'll want to define 2 functions:
> - a major mode
> - the function you want, with its argument, and which internally calls
>   the major mode.

Is it a requirement that major modes not take an argument?
In my case the major modes are not invoked interactively,
so having a required argument isn't a practical problem.
However, if it's considered bad form to pass arguments to them,
I'll make the change, which would allow using define-derived-mode.

-- 
Joe Riel

Re: add argument to a derived-mode

[Stefan Monnier]

> Is it a requirement that major modes not take an argument?

Technically, not quite (it may be called with no arguments, so any
argument it can accept should be optional).

> However, if it's considered bad form to pass arguments to them,
> I'll make the change, which would allow using define-derived-mode.

Yes, it's considered bad form and likely to lead to problems sooner
or later.


        Stefan

Re: add argument to a derived-mode

[Joe Riel]

On Sun, 15 Jan 2017 17:33:10 -0500
Stefan Monnier <monnier@IRO.UMontreal.CA> wrote:

> > Is it a requirement that major modes not take an argument?
> 
> Technically, not quite (it may be called with no arguments, so any
> argument it can accept should be optional).
> 
> > However, if it's considered bad form to pass arguments to them,
> > I'll make the change, which would allow using define-derived-mode.
> 
> Yes, it's considered bad form and likely to lead to problems sooner
> or later.

Thanks.  I made the changes.  Using define-derived-mode is certainly nicer.

-- 
Joe Riel