From: Leo Butler <Leo.Butler@umanitoba.ca>
To: Ihor Radchenko <yantar92@posteo.net>
Cc: Bastien <bzg@gnu.org>,
Lockywolf <for_org-bugs_2023-09-01@lockywolf.net>,
"emacs-orgmode@gnu.org" <emacs-orgmode@gnu.org>
Subject: Re: [MAINTENANCE] On how much we can expose internals into defcustom
Date: Wed, 6 Sep 2023 19:39:49 +0000 [thread overview]
Message-ID: <87cyyvdraz.fsf@t14.reltub.ca> (raw)
In-Reply-To: <87il8ovqeg.fsf@localhost> (Ihor Radchenko's message of "Tue, 05 Sep 2023 10:57:27 +0000")
On Tue, Sep 05 2023, Ihor Radchenko <yantar92@posteo.net> wrote:
> CCing Bastien, as he might want to intervene.
>
> Leo Butler <Leo.Butler@umanitoba.ca> writes:
>
> >> +(defcustom org-babel-maxima-command-arguments
> >> + "--very-quiet"
> >
> >> +(defcustom org-babel-maxima-batch/load
> >> + "batchload"
> >> +
> >> +(defcustom org-babel-maxima-graphic-file-format-string
> >> + "(set_plot_option ('[gnuplot_term, png]), set_plot_option ('[gnuplot_out_file, %S]))$"
> >> +
> >> +(defcustom org-babel-maxima-default-epilogue
> >> + "gnuplot_close ()$"
>
>>> IMHO, in their current state, if a user mindlessly customizes these
>>> options without knowing how ob-maxima internals work, ob-maxima may
>>> simply be broken.
>>
>> I think there is a fine line between being too rigid but working within
>> a limited scope (as ob-maxima is now), or providing enough customizable
>> options to let users do what they want. I would prefer the latter,
>> if the defaults provide a working configuration.
>>
>> Note that I do attempt to suggest other working options in the defcustom
>> definitions.
>
>>> As a general rule, we do not expose internal details that are _required_
>>> for things to work to users.
>>
>> I understand this principle, but, why not provide enough options for
>> users to configure a package to do what they want? Yes, that may mean
>> they break the package--but only temporarily, because returning to the
>> default options will return the package to a working state.
>
> That's a bit more tricky.
> Imagine, for example, that we commit
>
> +(defcustom org-babel-maxima-command-arguments
> + "--very-quiet"
>
> and some users will later customize the default value to
>
> "--very-quite --my-personal-switch-I-want"
>
> Then, in future, due to changes in Org or maxima, we might need to
> change "--very-quite" to something else in order to keep ob-maxima in
> working condition: "--very-quite=yes --another-useful-flag-we-absolutely-need-in-org"
>
> In order to make such a switch, we will have to force all the users
> change their customization - something we do not want to annoy users
> with.
I understand your general argument, but I doubt it is relevant to this
particular case.
In this specific case, that command-line option `--very-quiet' was
introduced in 17 years ago
(https://sourceforge.net/p/maxima/mailman/message/11796355/). The syntax
has not changed since it was introduced.
>
> So, leaving essential settings customizeable is not necessarily a good idea.
I understand your hesitation about full-blown customization using
`defcustom'. However, I would still like to have more dials to turn.
Perhaps we could add header arguments to get the desired customization?
E.g.
- :batch :: Control how the Maxima source is evaluated by Maxima.
1. Default. If nil or no, then use batchload with the --very-quiet
command-line flag.
2. If t or yes, then use batch with the --quiet command-line flag;
- :plot-engine :: Set the plotting package.
1. Default. If nil or no, the use `plot';
2. If `draw', then use `draw'.
- Similarly, we could do something like ob-R.el does, and construct the
graphics instructions using some additional header arguments and
grovelling the terminal from the filename (see
org-babel-R-construct-graphics-device-call).
My sense is that this would be more in keeping with how other ob-*.el
packages do things.
Best,
Leo
next prev parent reply other threads:[~2023-09-06 19:40 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-01 4:35 [BUG] Consider replacing bachload with batch in ob-maxima. [9.6.6 (release_9.6.6 @ /usr/share/emacs/30.0.50/lisp/org/)] Lockywolf
2023-09-01 18:33 ` Leo Butler
2023-09-02 7:19 ` Ihor Radchenko
2023-09-02 18:12 ` Leo Butler
2023-09-05 10:57 ` [MAINTENANCE] On how much we can expose internals into defcustom (was: [BUG] Consider replacing bachload with batch in ob-maxima. [9.6.6 (release_9.6.6 @ /usr/share/emacs/30.0.50/lisp/org/)]) Ihor Radchenko
2023-09-06 19:39 ` Leo Butler [this message]
2023-09-07 11:35 ` [MAINTENANCE] On how much we can expose internals into defcustom Ihor Radchenko
2023-09-12 21:09 ` [PATCH] ob-maxima.el, etc. (was Re: [MAINTENANCE] On how much we can expose internals into defcustom) Leo Butler
2023-09-15 9:41 ` Ihor Radchenko
2023-09-15 15:13 ` Leo Butler
2023-09-16 9:04 ` Ihor Radchenko
2023-09-19 19:25 ` Leo Butler
2023-09-20 9:17 ` Ihor Radchenko
2023-09-20 15:02 ` Leo Butler
2023-09-21 9:18 ` Ihor Radchenko
2023-09-21 14:03 ` Leo Butler
2023-09-22 9:43 ` Ihor Radchenko
2023-10-02 16:01 ` Leo Butler
2023-10-04 8:38 ` Ihor Radchenko
2023-10-04 13:07 ` Leo Butler
2023-09-02 7:06 ` [BUG] Consider replacing bachload with batch in ob-maxima. [9.6.6 (release_9.6.6 @ /usr/share/emacs/30.0.50/lisp/org/)] Ihor Radchenko
2023-09-02 18:20 ` Leo Butler
2023-09-03 5:25 ` Vladimir Nikishkin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://www.orgmode.org/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87cyyvdraz.fsf@t14.reltub.ca \
--to=leo.butler@umanitoba.ca \
--cc=bzg@gnu.org \
--cc=emacs-orgmode@gnu.org \
--cc=for_org-bugs_2023-09-01@lockywolf.net \
--cc=yantar92@posteo.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs/org-mode.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).