GNU bug report logs - #40863
[PATCH] Improve the display-time-world UI

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 40863 in the body.
You can then email your comments to 40863 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: Stefan Kangas <stefan <at> marxist.se>
To: bug-gnu-emacs <at> gnu.org
Subject: [PATCH] Improve the display-time-world UI
Date: Sun, 26 Apr 2020 10:56:03 +0200

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

I have made some improvements to the display-time-world UI.  I divided
them up into four patches to ease review and merging of the individual
features.  Please let me know what you think.

(Of course I can squash the patches before pushing if that is preferable.)

Patch 4 adds an alias 'world-clock'.  Ideally, I would like to rename
the somewhat obscurely named 'display-world-time' to 'world-clock' and
make the old names into obsolete aliases.  It would be good to hear
any opinions on that too.

I'm also not sure if any of this should go into NEWS, so I didn't do
that work for now.  Let me know if that's desirable.

Best regards,
Stefan Kangas

[0001-Kill-display-time-world-buffer-on-exit.patch (text/x-patch, attachment)]

[0002-New-face-for-display-time-world-label.patch (text/x-patch, attachment)]

[0003-Move-point-to-new-buffer-for-display-time-world.patch (text/x-patch, attachment)]

[0004-Improve-the-display-time-world-UI.patch (text/x-patch, attachment)]

Message #8 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sun, 26 Apr 2020 17:08:33 +0300

> From: Stefan Kangas <stefan <at> marxist.se>
> Date: Sun, 26 Apr 2020 10:56:03 +0200
> 
> -(defcustom display-time-world-buffer-name "*wclock*"
> +(defcustom display-time-world-buffer-name "*World Clock*"

Is this part really necessary? what if some application or user out
there expects the old name?

Thanks.

Message #11 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sun, 26 Apr 2020 16:19:04 +0200

Eli Zaretskii <eliz <at> gnu.org> writes:

>> -(defcustom display-time-world-buffer-name "*wclock*"
>> +(defcustom display-time-world-buffer-name "*World Clock*"
>
> Is this part really necessary? what if some application or user out
> there expects the old name?

Good point, I'll leave that part out.

Thanks for reviewing.

Best regards,
Stefan Kangas

Message #14 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Noam Postavsky <npostavs <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 40863 <at> debbugs.gnu.org, Stefan Kangas <stefan <at> marxist.se>
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Mon, 27 Apr 2020 13:20:40 -0400

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Stefan Kangas <stefan <at> marxist.se>
>> Date: Sun, 26 Apr 2020 10:56:03 +0200
>> 
>> -(defcustom display-time-world-buffer-name "*wclock*"
>> +(defcustom display-time-world-buffer-name "*World Clock*"
>
> Is this part really necessary? what if some application or user out
> there expects the old name?

If an application expects a particular name, then it's already broken
because the user can change the defcustom at any time, right?  And if a
user expects the old name, they can customize it back to the old value.

Message #19 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Noam Postavsky <npostavs <at> gmail.com>
Cc: 40863 <at> debbugs.gnu.org, stefan <at> marxist.se
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Mon, 27 Apr 2020 21:31:51 +0300

> From: Noam Postavsky <npostavs <at> gmail.com>
> Cc: Stefan Kangas <stefan <at> marxist.se>,  40863 <at> debbugs.gnu.org
> Date: Mon, 27 Apr 2020 13:20:40 -0400
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> >> From: Stefan Kangas <stefan <at> marxist.se>
> >> Date: Sun, 26 Apr 2020 10:56:03 +0200
> >> 
> >> -(defcustom display-time-world-buffer-name "*wclock*"
> >> +(defcustom display-time-world-buffer-name "*World Clock*"
> >
> > Is this part really necessary? what if some application or user out
> > there expects the old name?
> 
> If an application expects a particular name, then it's already broken
> because the user can change the defcustom at any time, right?  And if a
> user expects the old name, they can customize it back to the old value.

So you are saying that we should change these names from time to time
just to make a point?

Message #22 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Noam Postavsky <npostavs <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 40863 <at> debbugs.gnu.org, stefan <at> marxist.se
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Mon, 27 Apr 2020 15:00:37 -0400

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Noam Postavsky <npostavs <at> gmail.com>
>> Cc: Stefan Kangas <stefan <at> marxist.se>,  40863 <at> debbugs.gnu.org
>> Date: Mon, 27 Apr 2020 13:20:40 -0400
>> 
>> Eli Zaretskii <eliz <at> gnu.org> writes:
>> 
>> >> From: Stefan Kangas <stefan <at> marxist.se>
>> >> Date: Sun, 26 Apr 2020 10:56:03 +0200
>> >> 
>> >> -(defcustom display-time-world-buffer-name "*wclock*"
>> >> +(defcustom display-time-world-buffer-name "*World Clock*"
>> >
>> > Is this part really necessary? what if some application or user out
>> > there expects the old name?
>> 
>> If an application expects a particular name, then it's already broken
>> because the user can change the defcustom at any time, right?  And if a
>> user expects the old name, they can customize it back to the old value.
>
> So you are saying that we should change these names from time to time
> just to make a point?

No, I think changing the name here makes it more readable (i.e., it's
not a random change, it's an improvement).  And I think this benefit
outweighs the possible existence of broken applications, or users
possibly preferring the current value (I will admit this weighting also
somewhat depends on the probability of these hypotheticals actually
turning out to be real, which I've no real basis for; you might judge
them to be more likely than I do).

Message #25 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Noam Postavsky <npostavs <at> gmail.com>
Cc: 40863 <at> debbugs.gnu.org, stefan <at> marxist.se
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Mon, 27 Apr 2020 22:35:58 +0300

> From: Noam Postavsky <npostavs <at> gmail.com>
> Cc: 40863 <at> debbugs.gnu.org,  stefan <at> marxist.se
> Date: Mon, 27 Apr 2020 15:00:37 -0400
> 
> No, I think changing the name here makes it more readable (i.e., it's
> not a random change, it's an improvement).  And I think this benefit
> outweighs the possible existence of broken applications, or users
> possibly preferring the current value (I will admit this weighting also
> somewhat depends on the probability of these hypotheticals actually
> turning out to be real, which I've no real basis for; you might judge
> them to be more likely than I do).

I guess it's a matter of personal preferences, then.  E.g., we have
*scratch* and *xref*.

Message #28 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Dmitry Gutov <dgutov <at> yandex.ru>
To: Eli Zaretskii <eliz <at> gnu.org>, Noam Postavsky <npostavs <at> gmail.com>
Cc: 40863 <at> debbugs.gnu.org, stefan <at> marxist.se
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Tue, 28 Apr 2020 01:00:32 +0300

On 27.04.2020 22:35, Eli Zaretskii wrote:
> I guess it's a matter of personal preferences, then.  E.g., we have
> *scratch*  and*xref*.

Scratch is scratch. No longer name needed (I wouldn't know what to use).

Regarding xref, someday we'll use longer names. E.g. ones including the 
search input.

Message #31 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: "Basil L. Contovounesios" <contovob <at> tcd.ie>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Mon, 27 Apr 2020 23:58:31 +0100

Stefan Kangas <stefan <at> marxist.se> writes:

> I have made some improvements to the display-time-world UI.  I divided
> them up into four patches to ease review and merging of the individual
> features.  Please let me know what you think.

Thanks for working on this, see my comments below.

> (Of course I can squash the patches before pushing if that is preferable.)
>
> Patch 4 adds an alias 'world-clock'.  Ideally, I would like to rename
> the somewhat obscurely named 'display-world-time' to 'world-clock' and
> make the old names into obsolete aliases.  It would be good to hear
> any opinions on that too.

No strong feelings either way here.

> I'm also not sure if any of this should go into NEWS, so I didn't do
> that work for now.  Let me know if that's desirable.

A new entry-point command alias intended as a replacement for the
current command should definitely be announced.  Not sure about the
rest.

> Best regards,
> Stefan Kangas
>
> From 3d8c091a03c25826755b3735eca9da4b342826ba Mon Sep 17 00:00:00 2001
> From: Stefan Kangas <stefankangas <at> gmail.com>
> Date: Sun, 26 Apr 2020 09:38:03 +0200
> Subject: [PATCH 1/4] Kill display-time-world buffer on exit
>
> * lisp/time.el (display-time-world-mode-map): New defvar.
> (display-time-world-exit): New defun to kill buffer on exit.
> (display-time-world): Doc fix.
> ---
>  lisp/time.el | 15 ++++++++++++++-
>  1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/lisp/time.el b/lisp/time.el
> index 44fd1a7e33..a5e5b9b4a7 100644
> --- a/lisp/time.el
> +++ b/lisp/time.el
> @@ -509,6 +509,13 @@ display-time-mode
>  		 'display-time-event-handler)))
>  
>  
> +(defvar display-time-world-mode-map
> +  (let ((map (make-sparse-keymap)))
> +    (define-key map "g" nil)

Rather than remove the revert-buffer binding inherited from
special-mode-map, wouldn't it be better to implement a
revert-buffer-function that updates the world clocks displayed?

> +    (define-key map "q" 'display-time-world-exit)

[...]

> +(defun display-time-world-exit ()
> +  "Quit the world time buffer."
> +  (interactive)
> +  (quit-window t))

I don't think this is a change for the better.  The existing binding of
quit-window, inherited by and consistent across many special-mode
derivatives, already takes a prefix argument for killing the
corresponding buffer if a user prefers that to burying.  The proposed
change would make killing the only option by default.

>  ;;;###autoload
>  (defun display-time-world ()
>    "Enable updating display of times in various time zones.
> +\\<display-time-world-mode-map>
>  `display-time-world-list' specifies the zones.
> -To turn off the world time display, go to that window and type `q'."
> +To turn off the world time display, go to that window and type `\\[display-time-world-exit]'."

If you agree with me then this can be changed to \\[quit-window].

[...]

> @@ -540,7 +544,10 @@ display-time-world-display
>  	  (setq max-width width))))
>      (setq fmt (concat "%-" (int-to-string max-width) "s %s\n"))
>      (dolist (timedata (nreverse result))
> -      (insert (format fmt (car timedata) (cdr timedata))))
> +      (insert (format fmt
> +                      (propertize (car timedata)
> +                                  'face 'display-time-world-label)
> +                      (cdr timedata))))

Just curious: when do we use 'face' and when 'font-lock-face'?

>      (delete-char -1))
>    (goto-char (point-min)))
>  
> -- 
> 2.26.2
>
>
> From 0f2ca954e03b857554071ae48a98d44ba7030d69 Mon Sep 17 00:00:00 2001
> From: Stefan Kangas <stefankangas <at> gmail.com>
> Date: Sun, 26 Apr 2020 10:44:00 +0200
> Subject: [PATCH 3/4] Move point to new buffer for display-time-world
>
> * lisp/time.el (display-time-world): Move point to created buffer.
> ---
>  lisp/time.el | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/lisp/time.el b/lisp/time.el
> index 44885f74d1..84c0071b7d 100644
> --- a/lisp/time.el
> +++ b/lisp/time.el
> @@ -566,11 +566,11 @@ display-time-world
>    (when (and display-time-world-timer-enable
>               (not (get-buffer display-time-world-buffer-name)))
>      (run-at-time t display-time-world-timer-second 'display-time-world-timer))
> -  (with-current-buffer (get-buffer-create display-time-world-buffer-name)
> -    (display-time-world-display (time--display-world-list))
> -    (display-buffer display-time-world-buffer-name
> -		    (cons nil '((window-height . fit-window-to-buffer))))
> -    (display-time-world-mode)))
> +  (switch-to-buffer-other-window (get-buffer-create display-time-world-buffer-name))
> +  (display-time-world-display (time--display-world-list))
> +  (display-buffer display-time-world-buffer-name
> +		  (cons nil '((window-height . fit-window-to-buffer))))

switch-to-buffer and friends have historically not played nicely with
display-buffer-alist and switch-to-buffer-preserve-window-point when
used in Elisp libraries, with adverse user-visible effects.  Unless
there's a specific reason to use switch-to-buffer, pop-to-buffer and
friends should generally be preferred instead.

Note that using pop-to-buffer or similar here could remove the need for
calling both get-buffer-create and display-buffer, while remaining fully
customisable by the user.

> +  (display-time-world-mode))
>  
>  (defun display-time-world-timer ()
>    (if (get-buffer display-time-world-buffer-name)
> -- 
> 2.26.2
>
>
> From 47401a1e10213d89cbdb6eaa2b66b44c2be37dfe Mon Sep 17 00:00:00 2001
> From: Stefan Kangas <stefankangas <at> gmail.com>
> Date: Sun, 26 Apr 2020 10:47:17 +0200
> Subject: [PATCH 4/4] Improve the display-time-world UI
>
> * lisp/time.el (display-time-world-buffer-name): Change default
> buffer name.
> (display-time-world): Add usage instructions to modeline.

Are there many other modes that do this?  Is it really that helpful to
have a standard special-mode binding displayed prominently in the mode
line by default, with no easy way to remove it?

FWIW, on entry some modes display useful key bindings in the echo area,
which I find far less intrusive than a more permanent addition to the
mode line.

[...]

> @@ -570,7 +570,11 @@ display-time-world
>    (display-time-world-display (time--display-world-list))
>    (display-buffer display-time-world-buffer-name
>  		  (cons nil '((window-height . fit-window-to-buffer))))
> -  (display-time-world-mode))
> +  (display-time-world-mode)
> +  (setq mode-line-format (format "%15s %13s" "World Clock" "q to quit")))

Doesn't this override the entire mode line, erasing all other default
constructs?

What's wrong with the status quo, which displays the mode name, "World
clock", like every other mode does?

> +;; This would be a more intuitive name.
> +(defalias 'world-clock 'display-time-world)
>  
>  (defun display-time-world-timer ()
>    (if (get-buffer display-time-world-buffer-name)
> -- 
> 2.26.2

Otherwise LGTM.

-- 
Basil

Message #34 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Noam Postavsky <npostavs <at> gmail.com>
To: Dmitry Gutov <dgutov <at> yandex.ru>
Cc: Eli Zaretskii <eliz <at> gnu.org>, stefan <at> marxist.se, 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Mon, 27 Apr 2020 19:04:10 -0400

Dmitry Gutov <dgutov <at> yandex.ru> writes:

> On 27.04.2020 22:35, Eli Zaretskii wrote:
>> I guess it's a matter of personal preferences, then.  E.g., we have
>> *scratch*  and*xref*.
>
> Scratch is scratch. No longer name needed (I wouldn't know what to use).

I guess "*Scratch*" vs "*scratch*", but I don't think that's a
significant enough difference to even look at.

> Regarding xref, someday we'll use longer names. E.g. ones including
> the search input.

Since the term "xref" is used all over in package name, command names,
documentation, etc; users already need to be familiar with the short
form.  That's not the case for "wclock".  Anyway, none of this is
especially important.  Leaving the name as it is won't do any harm.

Message #37 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: Noam Postavsky <npostavs <at> gmail.com>
Cc: Eli Zaretskii <eliz <at> gnu.org>, 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 01 May 2020 17:03:10 +0200

Noam Postavsky <npostavs <at> gmail.com> writes:

>>> If an application expects a particular name, then it's already broken
>>> because the user can change the defcustom at any time, right?  And if a
>>> user expects the old name, they can customize it back to the old value.
>>
>> So you are saying that we should change these names from time to time
>> just to make a point?
>
> No, I think changing the name here makes it more readable (i.e., it's
> not a random change, it's an improvement).  And I think this benefit
> outweighs the possible existence of broken applications, or users
> possibly preferring the current value

I think Noam presents a compelling argument here.

It might be somewhat minor, and it might to some degree be a matter of
preference.  But even minor improvements can add up and contribute to
the overall user experience.

Best regards,
Stefan Kangas

Message #40 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org, npostavs <at> gmail.com
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 01 May 2020 18:13:50 +0300

> From: Stefan Kangas <stefan <at> marxist.se>
> Cc: Eli Zaretskii <eliz <at> gnu.org>,  40863 <at> debbugs.gnu.org
> Date: Fri, 01 May 2020 17:03:10 +0200
> 
> > No, I think changing the name here makes it more readable (i.e., it's
> > not a random change, it's an improvement).  And I think this benefit
> > outweighs the possible existence of broken applications, or users
> > possibly preferring the current value
> 
> I think Noam presents a compelling argument here.

Noam said several things about this, not just what you cite above.

Message #43 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 40863 <at> debbugs.gnu.org, npostavs <at> gmail.com
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 01 May 2020 18:26:28 +0200

Eli Zaretskii <eliz <at> gnu.org> writes:

>> > No, I think changing the name here makes it more readable (i.e., it's
>> > not a random change, it's an improvement).  And I think this benefit
>> > outweighs the possible existence of broken applications, or users
>> > possibly preferring the current value
>> 
>> I think Noam presents a compelling argument here.
>
> Noam said several things about this, not just what you cite above.

Well, I quoted the part I agree with, and tried not to quote unfairly.
Maybe I failed to include something important, but I'm not sure what
that is.

I take it that you aren't convinced?

Best regards,
Stefan Kangas

Message #46 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org, npostavs <at> gmail.com
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 01 May 2020 20:57:11 +0300

> From: Stefan Kangas <stefan <at> marxist.se>
> Cc: 40863 <at> debbugs.gnu.org,  npostavs <at> gmail.com
> Date: Fri, 01 May 2020 18:26:28 +0200
> 
> >> I think Noam presents a compelling argument here.
> >
> > Noam said several things about this, not just what you cite above.
> 
> Well, I quoted the part I agree with, and tried not to quote unfairly.
> Maybe I failed to include something important, but I'm not sure what
> that is.
> 
> I take it that you aren't convinced?

My bias is towards not breaking backward compatibility unless
necessary.  Noam in effect said that he tends towards your proposal,
but the issue is not important enough.  Where does that leave us?

Message #49 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Dmitry Gutov <dgutov <at> yandex.ru>
To: Eli Zaretskii <eliz <at> gnu.org>, Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org, npostavs <at> gmail.com
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 1 May 2020 21:07:38 +0300

On 01.05.2020 20:57, Eli Zaretskii wrote:
> My bias is towards not breaking backward compatibility unless
> necessary.  Noam in effect said that he tends towards your proposal,
> but the issue is not important enough.  Where does that leave us?

Do we vote?

I like "World Clock".

Message #52 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Dmitry Gutov <dgutov <at> yandex.ru>
Cc: 40863 <at> debbugs.gnu.org, stefan <at> marxist.se, npostavs <at> gmail.com
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 01 May 2020 21:18:23 +0300

> Cc: 40863 <at> debbugs.gnu.org, npostavs <at> gmail.com
> From: Dmitry Gutov <dgutov <at> yandex.ru>
> Date: Fri, 1 May 2020 21:07:38 +0300
> 
> On 01.05.2020 20:57, Eli Zaretskii wrote:
> > My bias is towards not breaking backward compatibility unless
> > necessary.  Noam in effect said that he tends towards your proposal,
> > but the issue is not important enough.  Where does that leave us?
> 
> Do we vote?

Do we have to?

Message #55 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: "Basil L. Contovounesios" <contovob <at> tcd.ie>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sat, 02 May 2020 13:50:53 +0200

"Basil L. Contovounesios" <contovob <at> tcd.ie> writes:

> Thanks for working on this, see my comments below.

Thank you for taking the time to make a thorough review.

I agree with your suggestions and will send an updated set of patches.

> Just curious: when do we use 'face' and when 'font-lock-face'?

I think this is explained in (info "(elisp) Special Properties").

Best regards,
Stefan Kangas

Message #58 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: "Basil L. Contovounesios" <contovob <at> tcd.ie>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sat, 02 May 2020 18:10:19 +0200

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

"Basil L. Contovounesios" <contovob <at> tcd.ie> writes:

> Stefan Kangas <stefan <at> marxist.se> writes:
>
>> I have made some improvements to the display-time-world UI.  I divided
>> them up into four patches to ease review and merging of the individual
>> features.  Please let me know what you think.
>
> Thanks for working on this, see my comments below.
>
>> (Of course I can squash the patches before pushing if that is preferable.)
>>
>> Patch 4 adds an alias 'world-clock'.  Ideally, I would like to rename
>> the somewhat obscurely named 'display-world-time' to 'world-clock' and
>> make the old names into obsolete aliases.  It would be good to hear
>> any opinions on that too.
>
> No strong feelings either way here.

So here's a revised set of patches:

1. Fix all Basil's comments, and remove the buffer rename that Eli
   didn't approve of.

2. Rename 'display-time-world' to 'world-clock', and add an alias for
   the old name.

Then comes the more ambitious part:

3. Rename 'display-time-world-*' functions and variables into
   'world-clock-*' equivalents and make the old names into obsolete
   aliases.

   Move the world-clock options from the defgroup display-time
   (sorting under the mode-line and mail), into a new defgroup
   world-clock (sorting under applications).

4. Rearrange the defcustoms time.el to clearly separate world-clock
   from display-time-mode.  Cleanup of :group args.

I have seen no to the proposals in patches 1-2 so far, and I hope that
we can agree to make the changes also in patches 3-4. I think the
result is much more logical from pretty much any point of view.

Comments are obviously very welcome.

Best regards,
Stefan Kangas

[0001-Improve-display-time-world-UI-Bug-40863.patch (text/x-diff, attachment)]

[0002-Make-display-time-world-into-an-alias-for-world-cloc.patch (text/x-diff, attachment)]

[0003-Rename-display-time-world-to-world-clock-Bug-40863.patch (text/x-diff, attachment)]

[0004-Rearrange-and-cleanup-code-in-time.el-Bug-40863.patch (text/x-diff, attachment)]

Message #61 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: "Basil L. Contovounesios" <contovob <at> tcd.ie>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sat, 02 May 2020 20:00:48 +0200

Stefan Kangas <stefan <at> marxist.se> writes:

> 4. Rearrange the defcustoms time.el to clearly separate world-clock
                             ^ in
>    from display-time-mode.  Cleanup of :group args.
>
> I have seen no to the proposals in patches 1-2 so far, and I hope that
                ^ objections
> we can agree to make the changes also in patches 3-4. I think the
> result is much more logical from pretty much any point of view.

Message #64 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: "Basil L. Contovounesios" <contovob <at> tcd.ie>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 22 May 2020 22:26:43 -0700

Stefan Kangas <stefan <at> marxist.se> writes:

> So here's a revised set of patches:
>
> 1. Fix all Basil's comments, and remove the buffer rename that Eli
>    didn't approve of.
>
> 2. Rename 'display-time-world' to 'world-clock', and add an alias for
>    the old name.
>
> Then comes the more ambitious part:
>
> 3. Rename 'display-time-world-*' functions and variables into
>    'world-clock-*' equivalents and make the old names into obsolete
>    aliases.
>
>    Move the world-clock options from the defgroup display-time
>    (sorting under the mode-line and mail), into a new defgroup
>    world-clock (sorting under applications).
>
> 4. Rearrange the defcustoms time.el to clearly separate world-clock
>    from display-time-mode.  Cleanup of :group args.
>
> I have seen no objections to the proposals in patches 1-2 so far, and
> I hope that we can agree to make the changes also in patches 3-4. I
> think the result is much more logical from pretty much any point of
> view.
>
> Comments are obviously very welcome.

No comments within 3 weeks.  Can I therefore assume there are no
objections to any of the above?  Thanks.

Best regards,
Stefan Kangas

Message #67 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: "Basil L. Contovounesios" <contovob <at> tcd.ie>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sat, 23 May 2020 14:43:18 +0100

Thanks for the ping and sorry about not getting back sooner.

I'm generally fine with the proposed changes, except for some comments
below.

> From e4e5012abdece7ce334900fd45f7e5ba06185ecc Mon Sep 17 00:00:00 2001
> From: Stefan Kangas <stefankangas <at> gmail.com>
> Date: Sun, 26 Apr 2020 10:16:06 +0200
> Subject: [PATCH 1/4] Improve display-time-world UI (Bug#40863)

[...]

> +(defvar display-time-world-mode-map
> +  (let ((map (make-sparse-keymap)))
> +    (define-key map "g" #'display-time-world-timer)

I think it's better to keep the standard default binding of
revert-buffer inherited from special-mode-map and set
revert-buffer-function in display-time-world-mode instead, as I
suggested in my last review.

Then display-time-world-timer won't need an interactive spec, which
doesn't sit too well with the function's current name and purpose.

> +    map)
> +  "Keymap for `display-time-world-mode'.")

[...]

> @@ -541,18 +554,17 @@ display-time-world-display
>  (defun display-time-world ()
>    "Enable updating display of times in various time zones.
>  `display-time-world-list' specifies the zones.
> -To turn off the world time display, go to that window and type `q'."
> +To turn off the world time display, go to that window and type `\\[quit-window]'."
>    (interactive)
>    (when (and display-time-world-timer-enable
>               (not (get-buffer display-time-world-buffer-name)))
>      (run-at-time t display-time-world-timer-second 'display-time-world-timer))
> -  (with-current-buffer (get-buffer-create display-time-world-buffer-name)
> -    (display-time-world-display (time--display-world-list))
> -    (display-buffer display-time-world-buffer-name
> -		    (cons nil '((window-height . fit-window-to-buffer))))
> -    (display-time-world-mode)))
> +  (pop-to-buffer display-time-world-buffer-name)
> +  (display-time-world-display (time--display-world-list))
> +  (display-time-world-mode))

You should call fit-window-to-buffer after populating the buffer to
preserve the old display-buffer action semantics.  (pop-to-buffer takes
the same action argument, but buffer might still be empty then.)

>  (defun display-time-world-timer ()
> +  (interactive)

This shouldn't be necessary, as mentioned above.

>    (if (get-buffer display-time-world-buffer-name)
>        (with-current-buffer (get-buffer display-time-world-buffer-name)
>          (display-time-world-display (time--display-world-list)))
> -- 
> 2.26.2
>
>
> From cd4c79a5758f0b6b8f5cc8acde62f806a1452b64 Mon Sep 17 00:00:00 2001
> From: Stefan Kangas <stefankangas <at> gmail.com>
> Date: Sat, 2 May 2020 16:08:33 +0200
> Subject: [PATCH 2/4] Make 'display-time-world' into an alias for 'world-clock'
>  (Bug#40863)
>
> * lisp/time.el (world-clock): Rename from 'display-time-world'.
> (display-time-world): Redefine as alias for 'world-clock'.
> * etc/NEWS: Announce the above change.
>
> (top-level, zoneinfo-style-world-list, legacy-style-world-list)
> (display-time-world-list, display-time-world-time-format)
> (display-time-world-buffer-name)
> (display-time-world-timer-enable)
> (display-time-world-timer-second, display-time-world-label)
> (display-time-world-timer): Update documentation to match the above
> rename.

These last entries should appear under lisp/time.el rather than
etc/NEWS, and you can either leave out top-level (which is the name of
an unrelated symbol) or write it as follows:

* lisp/time.el: Lorem ipsum...
(foo, bar): Curabitur pretium...

Yet another way you could probably get away with (others may correct
me), is:

* lisp/time.el (world-clock): Rename from 'display-time-world'.  Update
all documentation to mention the new name.
(display-time-world): Redefine as alias for 'world-clock'.
* etc/NEWS: Announce the above change.

[...]

> From 043650acccd2528c5459c3c854cbd886acae9642 Mon Sep 17 00:00:00 2001
> From: Stefan Kangas <stefankangas <at> gmail.com>
> Date: Sat, 2 May 2020 17:26:14 +0200
> Subject: [PATCH 3/4] Rename 'display-time-world' to 'world-clock' (Bug#40863)

[...]

> diff --git a/etc/NEWS b/etc/NEWS
> index d2b745c579..d97b148cc8 100644
> --- a/etc/NEWS
> +++ b/etc/NEWS
> @@ -291,9 +291,26 @@ These new navigation commands are bound to 'n' and 'p' in
>  ** Time
>  
>  ---
> -*** 'display-time-world' is now an alias for 'world-clock'.
> +*** 'display-time-world' has been renamed to 'world-clock'.
>  'world-clock' creates a buffer with an updating time display using
> -several time zones.  The new name is hoped to be more discoverable.
> +several time zones.

I think you can keep "it is hoped that the new names are more
discoverable" as a justification for this mass rename.

> +The following functions have been renamed:
> +
> +  'display-time-world'         to 'world-clock'
> +  'display-time-world-mode'    to 'world-clock-mode'
> +  'display-time-world-display' to 'world-clock-display'
> +  'display-time-world-timer'   to 'world-clock-update'
> +
> +The following options have been renamed:
> +
> +  'display-time-world-list'         to 'world-clock-list'
> +  'display-time-world-time-format'  to 'world-clock-time-format'
> +  'display-time-world-buffer-name'  to 'world-clock-buffer-namt'
> +  'display-time-world-timer-enable' to 'world-clock-timer-enablt'
> +  'display-time-world-timer-second' to 'world-clock-timer-secont'

The last three world-clock-* vars are misspelt.

> +The old names are now obsolete.
>  
>  
>  * New Modes and Packages in Emacs 28.1
> diff --git a/lisp/time.el b/lisp/time.el
> index fe290ff71f..3f1880b708 100644
> --- a/lisp/time.el
> +++ b/lisp/time.el
> @@ -122,6 +122,10 @@ display-time-server-down-time
>     "Time when mail file's file system was recorded to be down.
>  If that file system seems to be up, the value is nil.")
>  
> +(defgroup world-clock nil
> +  "Show a world clock."

Nit: s/show/display/

Rings better in my ears and is consistent with the description of the
display-time group.

> +  :group 'applications)

It should continue to be under :group 'display-time as well.

[...]

> +;;; Obsolete names
> +
> +(define-obsolete-variable-alias 'display-time-world-list
> +  'world-clock-list "28.1")
> +(define-obsolete-variable-alias 'display-time-world-time-format
> +  'world-clock-time-format "28.1")
> +(define-obsolete-variable-alias 'display-time-world-buffer-name
> +  'world-clock-buffer-name "28.1")
> +(define-obsolete-variable-alias 'display-time-world-timer-enable
> +  'world-clock-timer-enable "28.1")
> +(define-obsolete-variable-alias 'display-time-world-timer-second
> +  'world-clock-timer-second "28.1")

These varaliases need to be declared before the variables they refer to.

> +(define-obsolete-function-alias 'display-time-world-mode
> +  #'world-clock-mode "28.1")
> +(define-obsolete-function-alias 'display-time-world-display
> +  #'world-clock-display "28.1")
> +(define-obsolete-function-alias 'display-time-world
> +  #'world-clock "28.1")
> +(define-obsolete-function-alias 'display-time-world-timer
> +  #'world-clock-update "28.1")
> +
> +
> +;;; World clock
> +
> +(defface world-clock-label
>    '((t :inherit font-lock-variable-name-face))
>    "Face for time zone label in `world-clock' buffer.")
>  
> -(defvar display-time-world-mode-map
> +(defvar world-clock-mode-map
>    (let ((map (make-sparse-keymap)))
> -    (define-key map "g" #'display-time-world-timer)
> +    (define-key map "g" #'world-clock-update)

Ditto re: revert-buffer-function.

>      map)
> -  "Keymap for `display-time-world-mode'.")
> +  "Keymap for `world-clock-mode'.")

[...]

> @@ -552,32 +580,30 @@ display-time-world-display
>  ;;;###autoload
>  (defun world-clock ()
>    "Show world clock buffer with times in various time zones.
> -`display-time-world-list' specifies the zones.
> +`world-clock-list' specifies the zones.
>  To turn off the world time display, go to that window and type `\\[quit-window]'."
>    (interactive)
> -  (when (and display-time-world-timer-enable
> -             (not (get-buffer display-time-world-buffer-name)))
> -    (run-at-time t display-time-world-timer-second 'display-time-world-timer))
> -  (pop-to-buffer display-time-world-buffer-name)
> -  (display-time-world-display (time--display-world-list))
> -  (display-time-world-mode))
> -
> -(defun display-time-world-timer ()
> +  (when (and world-clock-timer-enable
> +             (not (get-buffer world-clock-buffer-name)))
> +    (run-at-time t world-clock-timer-second 'world-clock-update))
                                               ^
Nit: Quote with #'.

> +  (pop-to-buffer world-clock-buffer-name)
> +  (world-clock-display (time--display-world-list))
> +  (world-clock-mode))

[...]

> From 137c920f5b1ab34120060af0d4e3adada0f3a8a3 Mon Sep 17 00:00:00 2001
> From: Stefan Kangas <stefankangas <at> gmail.com>
> Date: Sat, 2 May 2020 17:37:06 +0200
> Subject: [PATCH 4/4] Rearrange and cleanup code in time.el (Bug#40863)

Nit: The verb is to 'clean up'.

> * lisp/time.el (world-clock, zoneinfo-style-world-list)
> (legacy-style-world-list, world-clock-list)
> (time--display-world-list, world-clock-time-format)
> (world-clock-timer-enable, world-clock-timer-second): Move definitions
> closer to 'world-clock' code.  Remove redundant :group args.
>
> (display-time-mail-file, display-time-mail-directory)
> (display-time-mail-function)
> (display-time-default-load-average)
> (display-time-load-average-threshold, display-time-day-and-date)
> (display-time-interval, display-time-24hr-format)
> (display-time-hook, display-time-mail-face)
> (display-time-use-mail-icon, display-time-mail-string)
> (display-time-format, display-time-string-forms): Remove redundant
> :group args.

[...]

Thanks,

-- 
Basil

Message #70 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: "Basil L. Contovounesios" <contovob <at> tcd.ie>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sat, 23 May 2020 14:44:00 +0100

Stefan Kangas <stefan <at> marxist.se> writes:

>> Just curious: when do we use 'face' and when 'font-lock-face'?
>
> I think this is explained in (info "(elisp) Special Properties").

I'm aware of the difference between the two properties.  My question is,
when do we prefer one or the other?  My impression is that
font-lock-face should be preferred wherever it makes sense, as its
display can be toggled.

Not that I'm arguing for switching to font-lock-face in this case; that
could be done another time.  I was just wondering out loud.

-- 
Basil

Message #73 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: "Basil L. Contovounesios" <contovob <at> tcd.ie>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 7 Aug 2020 10:23:01 -0700

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

"Basil L. Contovounesios" <contovob <at> tcd.ie> writes:

> Thanks for the ping and sorry about not getting back sooner.

Thank you for the detailed review!  I believe I have fixed all your
comments in the attached patches.  (Note that I squashed patch 2 and 3
into one patch.)

Please let me know if there are any further comments.

Best regards,
Stefan Kangas

[0001-Improve-display-time-world-UI-Bug-40863.patch (text/x-diff, attachment)]

[0002-Rename-display-time-world-to-world-clock-Bug-40863.patch (text/x-diff, attachment)]

[0003-Rearrange-and-clean-up-code-in-time.el-Bug-40863.patch (text/x-diff, attachment)]

Message #76 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: "Basil L. Contovounesios" <contovob <at> tcd.ie>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Fri, 07 Aug 2020 18:51:35 +0100

Stefan Kangas <stefan <at> marxist.se> writes:

> Thank you for the detailed review!  I believe I have fixed all your
> comments in the attached patches.  (Note that I squashed patch 2 and 3
> into one patch.)
>
> Please let me know if there are any further comments.

Thanks, we're close. ;)  See below.

> Subject: [PATCH 1/3] Improve display-time-world UI (Bug#40863)

[...]

>  (define-derived-mode display-time-world-mode special-mode "World clock"
>    "Major mode for buffer that displays times in various time zones.
>  See `display-time-world'."
> +  (setq revert-buffer-function 'display-time-world-timer)
                                  ^
                            Nit: #'

More importantly, doesn't revert-buffer-function need two parameters?

[...]

> Subject: [PATCH 3/3] Rearrange and clean up code in time.el (Bug#40863)
>
> * lisp/time.el (world-clock, zoneinfo-style-world-list)
> (legacy-style-world-list, world-clock-list)
> (time--display-world-list, world-clock-time-format)
> (world-clock-timer-enable, world-clock-timer-second): Move definitions
> closer to 'world-clock' code.

Unless I'm misreading it, the patch you attached seems to be adding
duplicate definitions rather than moving them.

-- 
Basil

Message #79 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: "Basil L. Contovounesios" <contovob <at> tcd.ie>
Cc: 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Sun, 9 Aug 2020 15:11:06 -0700

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

"Basil L. Contovounesios" <contovob <at> tcd.ie> writes:

>>  (define-derived-mode display-time-world-mode special-mode "World clock"
>>    "Major mode for buffer that displays times in various time zones.
>>  See `display-time-world'."
>> +  (setq revert-buffer-function 'display-time-world-timer)
>                                   ^
>                             Nit: #'
>
> More importantly, doesn't revert-buffer-function need two parameters?

Fixed.

>> Subject: [PATCH 3/3] Rearrange and clean up code in time.el (Bug#40863)
>>
>> * lisp/time.el (world-clock, zoneinfo-style-world-list)
>> (legacy-style-world-list, world-clock-list)
>> (time--display-world-list, world-clock-time-format)
>> (world-clock-timer-enable, world-clock-timer-second): Move definitions
>> closer to 'world-clock' code.
>
> Unless I'm misreading it, the patch you attached seems to be adding
> duplicate definitions rather than moving them.

Uhm, right... Fixed in the attached.

Thanks again for the review.

Best regards,
Stefan Kangas

[0001-Improve-display-time-world-UI-Bug-40863.patch (text/x-diff, attachment)]

[0002-Rename-display-time-world-to-world-clock-Bug-40863.patch (text/x-diff, attachment)]

[0003-Rearrange-and-clean-up-code-in-time.el-Bug-40863.patch (text/x-diff, attachment)]

Message #82 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: "Basil L. Contovounesios" <contovob <at> tcd.ie>, 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Tue, 18 Aug 2020 15:49:35 +0200

Stefan Kangas <stefan <at> marxist.se> writes:

> +*** 'display-time-world' has been renamed to 'world-clock'.
> +'world-clock' creates a buffer with an updating time display using
> +several time zones.  It is hoped that the new names are more
> +discoverable.

Looks like this was supposed to be applied, but it doesn't look like it
has been yet?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Message #85 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: "Basil L. Contovounesios" <contovob <at> tcd.ie>, 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Tue, 18 Aug 2020 14:37:48 +0000

Lars Ingebrigtsen <larsi <at> gnus.org> writes:

> Looks like this was supposed to be applied, but it doesn't look like it
> has been yet?

I wanted to give Basil some time to comment, since he had some feedback
on the previous patches.  But I guess 9 days is plenty in this case, so
I'll look this all over one final time and push it when I find some
time.

Best regards,
Stefan Kangas

Message #88 received at 40863 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: "Basil L. Contovounesios" <contovob <at> tcd.ie>, 40863 <at> debbugs.gnu.org
Subject: Re: bug#40863: [PATCH] Improve the display-time-world UI
Date: Tue, 18 Aug 2020 18:23:24 +0000

close 40863 28.1
thanks

Stefan Kangas <stefan <at> marxist.se> writes:

> Lars Ingebrigtsen <larsi <at> gnus.org> writes:
>
>> Looks like this was supposed to be applied, but it doesn't look like it
>> has been yet?
>
> I wanted to give Basil some time to comment, since he had some feedback
> on the previous patches.  But I guess 9 days is plenty in this case, so
> I'll look this all over one final time and push it when I find some
> time.

Now pushed to master as commit c70b5e8ad0.  Closing.

Best regards,
Stefan Kangas

Previous Next

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