all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
From: Jambunathan K <kjambunathan@gmail.com>
To: Juri Linkov <juri@jurta.org>
Cc: 13687@debbugs.gnu.org
Subject: bug#13687: /srv/bzr/emacs/trunk r111878: * lisp/replace.el(read-regexp): Let-bind `default' to the first
Date: Mon, 18 Mar 2013 12:54:09 +0530	[thread overview]
Message-ID: <87vc8pmaqu.fsf@gmail.com> (raw)
In-Reply-To: <87li9vruup.fsf@mail.jurta.org> (Juri Linkov's message of "Sun, 10 Mar 2013 20:28:14 +0200")


Juri

(For the sake of record)

Feel free to take over this bug.

In light of recent developments on the mailing list, I am a bit hesitant
to take these discussions any further.

Jambunathan K.


Juri Linkov <juri@jurta.org> writes:

>> EXPERIMENTAL and ABANDONED PATCH
>
> I think your patch is useful, please don't abandon it
> except its `this-command' part.  As you already noted
> `this-command' is very fragile and flaky.  Removing everything
> related to `this-command' would leave other useful parts of your patch
> that adds a new defcustom `read-regexp-defaults' and especially this part:
>
> +               (choice :tag "Function to retrieve the regexp"
> +                       (const :tag "Use no defaults" nil)
> +                       (radio
> +                        (function-item find-tag-default-as-regexp)
> +                        (function-item find-tag-default)
> +                        (function-item :tag "Regexp history"
> +                                       (lambda nil
> +                                         "Use regexp history."
> +                                         (car regexp-history)))
> +                        function)))))
>
> and a new function `read-regexp-defaults'.
>
> Instead of using `this-command', look for ideas to other similar features.
> For example, many invocations of minibuffer functions specify their
> HISTORY argument as a symbol that divides history variables into groups.
> The DEFAULTS argument could use a similar grouping, i.e. when
> `read-regexp' uses the symbol `regexp-history' in a call like:
>
>   (read-regexp "Regexp to highlight" (car regexp-history) 'regexp-history)
>
> This could be changed to specify DEFAULTS as the symbol `read-regexp-defaults':
>
>   (read-regexp "Regexp to highlight" 'read-regexp-defaults 'regexp-history)
>
> where `read-regexp-defaults' is a symbol name of the function that uses
> the value of the defcustom `read-regexp-defaults' the get the default value
> or returns nil.
>
> We could add as many additional default-providing functions
> as the number of places that call `read-regexp'.  But I think
> it should be enough to have just two functions:
>
> (defun read-regexp-defaults-or-history ()
>   (or (read-regexp-defaults)
>       (car regexp-history)))
>
> (defun read-regexp-defaults-or-tag ()
>   (or (read-regexp-defaults)
>       (find-tag-default-as-regexp)))
>
> These two functions are necessary to keep the current status quo
> where some commands traditionally provide the last history element
> as the default (`highlight-regexp', `occur-read-primary-args', `how-many',
> `flush-lines', `keep-lines'), and other commands provide the tag at point
> (`rgrep', `query-replace', `multi-occur-in-matching-buffers').
> This is an artificial division existing solely for historical reasons.
> These functions could be used e.g. in `occur-read-primary-args' as:
>
>   (read-regexp "List lines matching regexp" 'read-regexp-defaults-or-history 'regexp-history)
>
> and e.g. in `grep-read-regexp':
>
>   (read-regexp "Search for" 'read-regexp-defaults-or-tag 'regexp-history)
>
> This implementation will satisfy three goals:
>
> 1. Backward compatibility for the current traditional defaults,
> 2. Allow easy customization in one place (defcustom `read-regexp-defaults'),
> 3. Allow fine-tuning with function redefinitions, i.e. when
>    the users will ask for more specific functions they could be added as:
>
> (defun occur-read-regexp-defaults ()
>   (read-regexp-defaults-or-history))
>
> (defun grep-read-regexp-defaults ()
>   (read-regexp-defaults-or-tag))
>
> and can be used in `occur-read-primary-args':
>
>   (read-regexp "List lines matching regexp" 'occur-read-regexp-defaults 'regexp-history)
>
> and in `grep-read-regexp':
>
>   (read-regexp "Search for" 'grep-read-regexp-defaults 'regexp-history)





  reply	other threads:[~2013-03-18  7:24 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <E1UA5QC-0006Vm-Qb@vcs.savannah.gnu.org>
     [not found] ` <877glsyecw.fsf@gmail.com>
2013-02-28 18:12   ` bug#13687: /srv/bzr/emacs/trunk r111878: * lisp/replace.el (read-regexp): Let-bind `default' to the first Juri Linkov
2013-03-03  9:31     ` Juri Linkov
2013-03-06 18:00       ` Jambunathan K
2013-03-08 13:02       ` Jambunathan K
2013-03-08 15:29         ` bug#13687: /srv/bzr/emacs/trunk r111878: * lisp/replace.el(read-regexp): " Drew Adams
2013-03-08 16:53           ` Jambunathan K
2013-03-08 17:10             ` Drew Adams
2013-03-08 17:27               ` Jambunathan K
2013-03-08 17:28           ` Juri Linkov
2013-03-08 18:16             ` Drew Adams
2013-03-08 18:30               ` Jambunathan K
2013-03-08 18:53                 ` Drew Adams
2013-03-08 19:03                   ` Jambunathan K
2013-03-08 21:08                     ` Drew Adams
2013-03-09  8:47                 ` Jambunathan K
2013-03-09 15:08                   ` Drew Adams
2013-03-09 16:21                     ` Jambunathan K
2013-03-09 16:37                       ` Drew Adams
2013-03-09 16:59                         ` Jambunathan K
2013-03-09 17:14                           ` Drew Adams
2013-03-09 17:25                             ` Jambunathan K
2013-03-09 17:55                               ` Drew Adams
2013-03-10 18:28                   ` Juri Linkov
2013-03-18  7:24                     ` Jambunathan K [this message]
2013-03-04  5:46     ` bug#13687: /srv/bzr/emacs/trunk r111878: * lisp/replace.el (read-regexp): " Jambunathan K
2013-03-04  9:28       ` Juri Linkov
2013-03-06 18:03         ` Jambunathan K

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

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

  git send-email \
    --in-reply-to=87vc8pmaqu.fsf@gmail.com \
    --to=kjambunathan@gmail.com \
    --cc=13687@debbugs.gnu.org \
    --cc=juri@jurta.org \
    /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 external index

	https://git.savannah.gnu.org/cgit/emacs.git
	https://git.savannah.gnu.org/cgit/emacs/org-mode.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.