unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
From: Eli Zaretskii <eliz@gnu.org>
To: Stefan Kangas <stefan@marxist.se>
Cc: 44858@debbugs.gnu.org, larsi@gnus.org
Subject: bug#44858: [PATCH] Make byte-compiler warn about wide docstrings
Date: Fri, 11 Dec 2020 09:53:27 +0200	[thread overview]
Message-ID: <83lfe43hjc.fsf@gnu.org> (raw)
In-Reply-To: <CADwFkm=pXm4PDHGXgHY+xX1BT0DGESar1EWZBcSZsA2c3gF9Mg@mail.gmail.com> (message from Stefan Kangas on Thu, 10 Dec 2020 14:59:16 -0600)

> From: Stefan Kangas <stefan@marxist.se>
> Date: Thu, 10 Dec 2020 14:59:16 -0600
> Cc: 44858@debbugs.gnu.org
> 
> +** byte compiler

Please capitalize "Byte".

> +*** The byte-compiler now warns about too wide documentation strings.
> +By default, it will warn if a documentation string is wider than the
> +largest of 80 characters or 'fill-column'.  See the new user
> +option 'byte-compile-docstring-max-column'.

The last sentence will be more useful if you make it more informative,
e.g.

  This is controlled by the new user option
  'byte-compile-docstring-max-column'.

> +(defvar byte-compile--wide-docstring-substitution-len 3
> +  "Substitution width used in `byte-compile--wide-docstring-p'.")

The doc string should be more detailed.  Let's at least explain what
hides behind "substitution", and say that the value is a heuristic.

> +(defcustom byte-compile-docstring-max-column 80
> +  "Length that a doc string can be before the byte-compiler reports a warning."

I suggest "width", not "length", since you both use the former
elsewhere, and because "column" goes better with "width".

Also, "can be before" is not the clearest way of saying this.  I
suggest

    Recommended maximum width of doc string lines.
  The byte-compiler will emit a warning for doc-string lines
  wider than this.  If `fill-column' has a larger value, it will
  override this variable.

> +(defun byte-compile-docstring-length-warn (form)
> +  "Warn if documentation string of FORM is too wide.
> +It is too wide if it is longer than the largest of `fill-column'
> +and `byte-compile-docstring-max-column'."

"It is too wide if it is longer" is inaccurate, I think, because we
actually test each _line_ of the doc string against the limit, while
the above seems to imply the overall length of the doc string (which
could be multi-line) is tested.  So:

  It is too wide if some of its lines is wider than the largest of...

> --- /dev/null
> +++ b/test/lisp/emacs-lisp/bytecomp-resources/warn-long-docstring-autoload.el
> @@ -0,0 +1,3 @@
> +;;; -*- lexical-binding: t -*-
> +(autoload 'foox "foo"
> +  "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx")

I see that none of the tests use multi-line doc strings, is that
right?  If so, I think it would be good to have at least a couple, one
where the first line is too wide, and another where some line other
than the first is too wide.  Also, what about testing the fact that
fill-column overrides the value of the new option?

Thanks.





  parent reply	other threads:[~2020-12-11  7:53 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-25  1:36 bug#44858: [PATCH] Make byte-compiler warn about wide docstrings Stefan Kangas
2020-11-26 10:49 ` Lars Ingebrigtsen
2020-11-26 12:46   ` Stefan Kangas
2020-11-26 12:53     ` Lars Ingebrigtsen
2020-12-10 20:59       ` Stefan Kangas
2020-12-10 21:53         ` Stefan Kangas
2020-12-11  8:16           ` Eli Zaretskii
2020-12-11 20:03             ` Stefan Kangas
2020-12-11  7:33         ` Eli Zaretskii
2020-12-11 20:36           ` Stefan Kangas
2020-12-19 11:22             ` Eli Zaretskii
2020-12-19 16:50               ` Stefan Kangas
2020-12-19 17:14                 ` Eli Zaretskii
2020-12-29  1:27                   ` Basil L. Contovounesios
2020-12-29  2:16                     ` Lars Ingebrigtsen
2020-12-19 17:18                 ` Lars Ingebrigtsen
2020-12-19 23:48                   ` Stefan Kangas
2020-12-11  7:53         ` Eli Zaretskii [this message]
2020-12-19 23:55           ` Stefan Kangas
2020-12-20 17:53             ` Lars Ingebrigtsen
2020-12-28  5:18               ` Stefan Kangas
2020-12-11 15:13         ` Lars Ingebrigtsen
2020-12-30 12:07       ` Stefan Kangas
2020-12-31  4:42         ` Lars Ingebrigtsen
2020-11-26 14:19 ` Eli Zaretskii
2020-11-27  8:37   ` Lars Ingebrigtsen
2020-11-27 11:15     ` Stefan Kangas
2020-11-27 12:44       ` Eli Zaretskii
2020-12-06 11:09         ` Stefan Kangas
2020-12-06 11:19           ` Eli Zaretskii
2020-12-06 16:54           ` Drew Adams
2020-11-27 18:36     ` Drew Adams
2020-11-27 18:55       ` Drew Adams
2020-12-03 20:18 ` Tomas Nordin
2020-12-11 20:14   ` Stefan Kangas
2021-09-24 17:25 ` Stefan Kangas
2021-09-25  1:07   ` Lars Ingebrigtsen
2021-09-26 11:38     ` Stefan Kangas

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.gnu.org/software/emacs/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=83lfe43hjc.fsf@gnu.org \
    --to=eliz@gnu.org \
    --cc=44858@debbugs.gnu.org \
    --cc=larsi@gnus.org \
    --cc=stefan@marxist.se \
    /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.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).