bug#23569: thing-at-point, source of slowness

bug#23569: thing-at-point, source of slowness

[Andreas Röhler]

[-- Attachment #1: Type: text/plain, Size: 454 bytes --]

in thingatpt.el:

;; The function bounds-of-thing-at-point finds the beginning and end
;; positions by moving first forward to the end of the "thing", and then
;; backwards to the beginning.

;;;;;;;

Also *bounds-of-thing-at-point jumps to-and-fro, doing some sanity check 
every time.

Suggest to jump ***first *at the beginning, as backward move might be slow.

Avoid to-and-fro, deliver tests instead, so first try will be reliable 
or return nil.
*

[-- Attachment #2: Type: text/html, Size: 832 bytes --]

bug#23569: thing-at-point, source of slowness

[Eli Zaretskii]

> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
> Date: Wed, 18 May 2016 09:59:41 +0200
> 
> Suggest to jump first at the beginning, as backward move might be slow. 

Can you tell why?  AFAIK, moving back is no slower than moving
forward.

bug#23569: thing-at-point, source of slowness

[Andreas Röhler]

On 18.05.2016 19:52, Eli Zaretskii wrote:
>> From: Andreas Röhler <andreas.roehler@easy-emacs.de>
>> Date: Wed, 18 May 2016 09:59:41 +0200
>>
>> Suggest to jump first at the beginning, as backward move might be slow.
> Can you tell why?  AFAIK, moving back is no slower than moving
> forward.

Referring to non-greedy behaviour of re-search-backward. Which requires 
additional checks WRT to a greedy-like match.

After all, it doesn't seem that important, as implemented forms don't 
use re-search.

Remains the to-and-fro issue.

bug#23569: thing-at-point, source of slowness

[Lars Ingebrigtsen]

Andreas Röhler <andreas.roehler@easy-emacs.de> writes:

> in thingatpt.el:
>
> ;; The function bounds-of-thing-at-point finds the beginning and end
> ;; positions by moving first forward to the end of the "thing", and then
> ;; backwards to the beginning.
>
> ;;;;;;;
>
> Also bounds-of-thing-at-point jumps to-and-fro, doing some sanity check every
> time.
>
> Suggest to jump first at the beginning, as backward move might be slow. 
>
> Avoid to-and-fro, deliver tests instead, so first try will be reliable or
> return nil.

(I'm going through old bug reports that unfortunately got little response at
the time.)

I'm not sure I understand the issue, even with the response you gave to
Eli at the time.

Do you have a test case that demonstrates this slowness, and an example
of how you'd change the code?

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

bug#23569: thing-at-point, source of slowness

[Andreas Röhler]

On 08.07.21 02:49, Lars Ingebrigtsen wrote:
> Andreas Röhler <andreas.roehler@easy-emacs.de> writes:
>
>> in thingatpt.el:
>>
>> ;; The function bounds-of-thing-at-point finds the beginning and end
>> ;; positions by moving first forward to the end of the "thing", and then
>> ;; backwards to the beginning.
>>
>> ;;;;;;;
>>
>> Also bounds-of-thing-at-point jumps to-and-fro, doing some sanity check every
>> time.
>>
>> Suggest to jump first at the beginning, as backward move might be slow.
>>
>> Avoid to-and-fro, deliver tests instead, so first try will be reliable or
>> return nil.
> (I'm going through old bug reports that unfortunately got little response at
> the time.)
>
> I'm not sure I understand the issue, even with the response you gave to
> Eli at the time.
>
> Do you have a test case that demonstrates this slowness, and an example
> of how you'd change the code?
>
Thanks for your care. Don't think it's worth digging into that after 
that long time. Suggest to close.

bug#23569: thing-at-point, source of slowness

[Lars Ingebrigtsen]

Andreas Röhler <andreas.roehler@easy-emacs.de> writes:

> Thanks for your care. Don't think it's worth digging into that after
> that long time. Suggest to close.

OK; done.

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