Re: Output to directory patch

[Al Haji-Ali]

Hello Ikumi,

On 12/03/2021, Ikumi Keita wrote:
> 1. First of all, please write documentation for the new feature. Add
> appropriate section of paragraphs to texinfo files under doc/
> subdirectory.

How about this at the end of section "4.5 Controlling the output"

,----
| Addionally, output files produced by @AUCTeX{} can be placed in a seperate
| directory.
|
| @defopt TeX-output-dir
| Set this option to the path of a directory where output files will be
| placed. The output files include those that are produced by applications
| running under @AUCTeX{}, temporary files related to region processing and
| preview-latex files. If a relative path is specified, it is
| interpreted as being relative to the master file in a mutlifile document.
|
| This is a buffer local variable and must be set seprately for all
| documents and all files in a multifile document. Alternatively, you may
| use @code{setq-default} to set the default value of this option or set it
| as a directory local variable.
|
| Note that a non-nil value of @code{TeX-output-dir} might be incompatible
| with some @TeX{} commands and macros. In particular, the @TeX{} macro
| @samp{\include} is known to not work with this option. Some @TeX{} packages
| which produce intermediary files might also be incompatible. A possible
| workaround for those packages is to append the value of @code{TeX-output-dir}
| to the environment variables @samp{TEXINPUTS} and @samp{BIBINPUTS}.
| @end defopt
`----

(Aside, I like your way of quoting code in emails. Is there an emacs command 
that you use? I am having to do it manually at the moment.)

> 5-2. The default value nil doesn't match the `string' type, thus
>      customization buffer shown by M-x customize-option displays
>      "mismatch" like this:

Ah right, how about these two choices instead?
,----
| (choice (const  :tag "Directory of master file" nil)
|         (string :tag "Custom" "build"))
`----

> 6.(Option) It would be slightly better if 'FUNCTION-NAME-SYMBOLs are
> rewritten as #'FUNCTION-NAME-SYMBOLs in codes committed anew, but I
> leave it to your choice since it would make only thin difference.
I am all for conformity to a standard :)

Best regards,
-- Al