bug#21830: Useless messages from comint-previous-matching-input

bug#21830: Useless messages from comint-previous-matching-input

[Jay Adams]

[-- Attachment #1: Type: text/plain, Size: 503 bytes --]

Can we please delete this line:

      (message "History item: %d" (1+ pos))

form comint-previous-matching-input in comint.el?  It provides no value and
makes searching the history very slow, at least on a Mac, when a large part
of the history has to be searched.

Basically, with these messages in place, one of two things happens:

1. They go by so fast as to be unnoticeable (this happens on my Ubuntu
desktop).
2. They go by slowly and make the search excruciatingly slow (this happens
on my Mac).

[-- Attachment #2: Type: text/html, Size: 694 bytes --]

bug#21830: Useless messages from comint-previous-matching-input

[John Wiegley]

>>>>> Jay Adams <jka@google.com> writes:

> Can we please delete this line:
> (message "History item: %d" (1+ pos))

Since there's no way to make use of that information, I wouldn't mind seeing
it removed.

Would you be willing to submit a patch?

John

bug#21830: Useless messages from comint-previous-matching-input

[Eli Zaretskii]

> From: "John Wiegley" <jwiegley@gmail.com>
> Date: Wed, 04 Nov 2015 13:04:06 -0500
> Cc: 21830@debbugs.gnu.org
> 
> >>>>> Jay Adams <jka@google.com> writes:
> 
> > Can we please delete this line:
> > (message "History item: %d" (1+ pos))
> 
> Since there's no way to make use of that information, I wouldn't mind seeing
> it removed.

It was there since 1993.

bug#21830: Useless messages from comint-previous-matching-input

[John Wiegley]

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

>> Since there's no way to make use of that information, I wouldn't mind seeing
>> it removed.

> It was there since 1993.

Well, hard to argue against that too. I can't say I feel any real passion for
this issue, as there may be some who like the feedback they see upon pressing
the key.

I guess what I'd like to know more, then, is why it's so slow on the
reporter's Mac machine?

John

bug#21830: Useless messages from comint-previous-matching-input

[Eli Zaretskii]

> From: "John Wiegley" <jwiegley@gmail.com>
> Cc: jka@google.com,  21830@debbugs.gnu.org
> Date: Wed, 04 Nov 2015 13:24:57 -0500
> 
> >>>>> Eli Zaretskii <eliz@gnu.org> writes:
> 
> >> Since there's no way to make use of that information, I wouldn't mind seeing
> >> it removed.
> 
> > It was there since 1993.
> 
> Well, hard to argue against that too. I can't say I feel any real passion for
> this issue, as there may be some who like the feedback they see upon pressing
> the key.

How about a defcustom to turn that off?

bug#21830: Useless messages from comint-previous-matching-input

[John Wiegley]

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

>> > It was there since 1993.
>> 
>> Well, hard to argue against that too. I can't say I feel any real passion for
>> this issue, as there may be some who like the feedback they see upon pressing
>> the key.

> How about a defcustom to turn that off?

Well, that was silly of me. Of course a defcustom is the right answer to this
query.

John

bug#21830: Useless messages from comint-previous-matching-input

[Wolfgang Jenkner]

On Wed, Nov 04 2015, Jay Adams wrote:

> 1. They go by so fast as to be unnoticeable (this happens on my Ubuntu
> desktop).
> 2. They go by slowly and make the search excruciatingly slow (this happens
> on my Mac).

I guess the message is meant to remain visible until you use the
minibuffer (again) or some other message pops up from somewhere.

At least, this is what happens for me when I call
comint-previous-matching-input-from-input or
comint-previous-matching-input repeatedly (via some key binding).

Can you reproduce your problem with emacs -Q?

bug#21830: Useless messages from comint-previous-matching-input

[Jay Adams]

[-- Attachment #1: Type: text/plain, Size: 878 bytes --]

Thanks to all who replied.

It looks like this is fixed in 24.5 (I was running 24.3).  That message
call now appears inside of (let ((message-log-max nil)) ...).

I verified that this fixes the problem on the Mac.


On Wed, Nov 4, 2015 at 6:11 PM, Wolfgang Jenkner <wjenkner@inode.at> wrote:

> On Wed, Nov 04 2015, Jay Adams wrote:
>
> > 1. They go by so fast as to be unnoticeable (this happens on my Ubuntu
> > desktop).
> > 2. They go by slowly and make the search excruciatingly slow (this
> happens
> > on my Mac).
>
> I guess the message is meant to remain visible until you use the
> minibuffer (again) or some other message pops up from somewhere.
>
> At least, this is what happens for me when I call
> comint-previous-matching-input-from-input or
> comint-previous-matching-input repeatedly (via some key binding).
>
> Can you reproduce your problem with emacs -Q?
>

[-- Attachment #2: Type: text/html, Size: 1290 bytes --]

bug#21830: Useless messages from comint-previous-matching-input

[John Wiegley]

>>>>> Jay Adams <jka@google.com> writes:

> It looks like this is fixed in 24.5 (I was running 24.3). That message call
> now appears inside of (let ((message-log-max nil)) ...).

> I verified that this fixes the problem on the Mac.

Does this mean you don't need a customization variable to disable the message
anymore?

John

bug#21830: Useless messages from comint-previous-matching-input

[Jay Adams]

[-- Attachment #1: Type: text/plain, Size: 451 bytes --]

Yes. No customization variable.
On Nov 4, 2015 9:15 PM, "John Wiegley" <jwiegley@gmail.com> wrote:

> >>>>> Jay Adams <jka@google.com> writes:
>
> > It looks like this is fixed in 24.5 (I was running 24.3). That message
> call
> > now appears inside of (let ((message-log-max nil)) ...).
>
> > I verified that this fixes the problem on the Mac.
>
> Does this mean you don't need a customization variable to disable the
> message
> anymore?
>
> John
>

[-- Attachment #2: Type: text/html, Size: 796 bytes --]