GNU bug report logs - #56449
29.0.50; Revisit target shortcuts in erc-message

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 56449 in the body.
You can then email your comments to 56449 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: "J.P." <jp <at> neverwas.me>
To: bug-gnu-emacs <at> gnu.org
Subject: 29.0.50; Revisit target shortcuts in erc-message
Date: Fri, 08 Jul 2022 06:21:29 -0700

[Message part 1 (text/plain, inline)]

Tags: patch

The purpose of the function `erc-message' is to provide a "target
shortcut" option to a handful of "slash" commands, mostly ones emitting
a PRIVMSG. When a user invoking such a command supplies a period or a
comma instead of a channel or a nick, whatever's stored in the relevant
slot of `erc-server-last-peers' gets substituted in the outgoing
message. IOW, issuing a

  ERC> /MSG , hi

recalls the last nick to speak over a given connection. And

  ERC> /MSG . hi

reuses the target (channel or query) from the last `erc-message'
invocation, which may have been something like

  ERC> /MSG bob hi

with no actual shortcut present. As it happens, these targets are
stashed in the default value shared across all connections, which is
almost certainly undesirable. But attempting to fix it forces us to
confront some basic usability concerns that vary based on whatever we
deem to be the feature's intended "scope":

1. Network (patch)

   The name of the variable in question, `erc-server-last-peers', may
   suggest that this feature should be scoped per IRC connection. But
   what's the point of blindly addressing the last speaker among all of
   a network's buffers without knowing exactly who you're engaging?

2. Buffer-local

   Using a "," for PM'ing the last speaker in a target buffer may lead
   to a surprise if anyone else speaks before you've submitted the
   input. As with (1), only the logs know for sure. Also, the "." form
   is only useful so long as you refrain from speaking.

For now, I propose temporarily adopting (1) by confining shared values
to server buffers. As far as attempting to renovate this feature goes
(rather than, say, tacitly retiring it), I think it's worth considering
that nick completion and abbrevs have long provided capable
alternatives.

In that vein, the last patch below claims to offer a compromise that
somewhat extends the existing pcomplete integration for the affected
slash commands so that dots and commas expand to their relevant targets
in the input area. This reduces potential targeting errors by offering
visual confirmation. IOW, if the last person to speak was Alice, typing

  ERC> /m ,[TAB]

would expand to

  ERC> /m alice _

And while this demo is built atop (1), the idea could easily be adapted
for (2) or another variant. BTW, if anyone has experience with
pcomplete, I'm pretty sure I committed an epic faceplant in trying to
apply the API. If we end up going with this, please help. But, for now,
the door remains wide open, and alternative solutions are more than
welcome.

Thanks,
J.P.

P.S. This bug isn't (yet) claimed by any particular ERC release.


In GNU Emacs 29.0.50 (build 1, x86_64-pc-linux-gnu, GTK+ Version 3.24.34, cairo version 1.17.6)
 of 2022-07-06 built on localhost
Repository revision: e6504c3eda12c72268d2db6598764f043b74c24d
Repository branch: master
Windowing system distributor 'The X.Org Foundation', version 11.0.12014000
System Description: Fedora Linux 36 (Workstation Edition)

Configured using:
 'configure --enable-check-lisp-object-type --enable-checking=yes,glyphs
 'CFLAGS=-O0 -g3'
 PKG_CONFIG_PATH=:/usr/lib64/pkgconfig:/usr/share/pkgconfig'

Configured features:
ACL CAIRO DBUS FREETYPE GIF GLIB GMP GNUTLS GPM GSETTINGS HARFBUZZ JPEG
JSON LCMS2 LIBOTF LIBSELINUX LIBSYSTEMD LIBXML2 M17N_FLT MODULES NOTIFY
INOTIFY PDUMPER PNG RSVG SECCOMP SOUND SQLITE3 THREADS TIFF
TOOLKIT_SCROLL_BARS WEBP X11 XDBE XIM XINPUT2 XPM GTK3 ZLIB

Important settings:
  value of $LANG: en_US.UTF-8
  value of $XMODIFIERS: @im=ibus
  locale-coding-system: utf-8-unix

Major mode: Lisp Interaction

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-mode: t
  eldoc-mode: t
  show-paren-mode: t
  electric-indent-mode: t
  mouse-wheel-mode: t
  tool-bar-mode: t
  menu-bar-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  blink-cursor-mode: t
  line-number-mode: t
  indent-tabs-mode: t
  transient-mark-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug message mailcap yank-media puny dired
dired-loaddefs rfc822 mml mml-sec password-cache epa derived epg rfc6068
epg-config gnus-util text-property-search time-date subr-x mm-decode
mm-bodies mm-encode mail-parse rfc2231 mailabbrev gmm-utils mailheader
cl-loaddefs cl-lib sendmail rfc2047 rfc2045 ietf-drums mm-util
mail-prsvr mail-utils rmc iso-transl tooltip eldoc paren electric
uniquify ediff-hook vc-hooks lisp-float-type elisp-mode mwheel
term/x-win x-win term/common-win x-dnd tool-bar dnd fontset image
regexp-opt fringe tabulated-list replace newcomment text-mode lisp-mode
prog-mode register page tab-bar menu-bar rfn-eshadow isearch easymenu
timer select scroll-bar mouse jit-lock font-lock syntax font-core
term/tty-colors frame minibuffer nadvice seq simple cl-generic
indonesian philippine cham georgian utf-8-lang misc-lang vietnamese
tibetan thai tai-viet lao korean japanese eucjp-ms cp51932 hebrew greek
romanian slovak czech european ethiopic indian cyrillic chinese
composite emoji-zwj charscript charprop case-table epa-hook
jka-cmpr-hook help abbrev obarray oclosure cl-preloaded button loaddefs
faces cus-face macroexp files window text-properties overlay sha1 md5
base64 format env code-pages mule custom widget keymap
hashtable-print-readable backquote threads dbusbind inotify lcms2
dynamic-setting system-font-setting font-render-setting cairo
move-toolbar gtk x-toolkit xinput2 x multi-tty make-network-process
emacs)

Memory information:
((conses 16 35545 5688)
 (symbols 48 5073 0)
 (strings 32 13303 1546)
 (string-bytes 1 429956)
 (vectors 16 9197)
 (vector-slots 8 145428 11407)
 (floats 8 21 25)
 (intervals 56 214 0)
 (buffers 992 10))

[0001-Demo-mutating-default-value-of-erc-server-last-peers.patch (text/x-patch, attachment)]

[0002-SQUASH-Avoid-mutating-default-value-of-erc-server-la.patch (text/x-patch, attachment)]

[0003-POC-FIXME-Add-erc-message-shortcuts-to-erc-pcomplete.patch (text/x-patch, attachment)]

Message #8 received at 56449 <at> debbugs.gnu.org (full text, mbox):

From: "J.P." <jp <at> neverwas.me>
To: 56449 <at> debbugs.gnu.org
Cc: emacs-erc <at> gnu.org
Subject: Re: bug#56449: 29.0.50; Revisit target shortcuts in erc-message
Date: Tue, 12 Jul 2022 01:10:58 -0700

[Message part 1 (text/plain, inline)]

"J.P." <jp <at> neverwas.me> writes:

> IOW, issuing a
>
>   ERC> /MSG , hi
>
> recalls the last nick to speak over a given connection. And
>
>   ERC> /MSG . hi
>
> reuses the target (channel or query) from the last `erc-message'
> invocation

Actually, thinking on this a bit has revealed that these two state items
(last received vs. last sent) may be less symmetrical than originally
believed in terms of use case and function: the "." form always expands
to a substring of something already typed, IOW "previous input", while
the "," form may well produce something novel.

So, if the dot form is more akin to everyday command-history recall,
then ERC may have an easier time providing a useful substitute by
improving upon its existing offering, namely, erc-ring. Indeed, adopting
in-house variants of the more powerful history-traversing commands seen
elsewhere in Emacs, such as `previous-complete-history-element' and (in
some sense) `previous-matching-history-element', wouldn't be the worst
call (IMO).

> For now, I propose temporarily adopting (1) by confining shared values
> to server buffers.

... and stopping there rather than further complicating the situation
for little perceived gain, convenience wise, by fiddling with imperfect
alternatives to niche features.

As for the comma form, in case the above wasn't clear, I'm in favor of
dialing things back and shelving the notion of enhanced recipient
recall. (In the future, we could conceivably retrieve candidates by
searching backward through text-properties to gather the most recent
speakers in a target buffer. These could be presented via completing
read, perhaps anchored by a transient to offer unified access to various
scopes, like buffer, network, etc.)

For the sake of completeness, I have attached the surviving changes,
minor as they are, and will be adding them or similar to ERC by week's
end unless anyone has additional insights. Thanks.

[0001-Avoid-mutating-default-value-of-erc-server-last-peer.patch (text/x-patch, attachment)]

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.