GNU bug report logs - #35708
[27.0.50]: thingatpt.el, thing-at-point-looking-at redundant

Previous Next

Package: emacs;

Reported by: Andreas Röhler <andreas.roehler <at> easy-emacs.de>

Date: Mon, 13 May 2019 07:19:01 UTC

Severity: wishlist

Tags: notabug, wontfix

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

Bug is archived. No further changes may be made.

Full log


View this message in rfc822 format

From: Andreas Röhler <andreas.roehler <at> easy-emacs.de>
To: npostavs <at> gmail.com
Cc: 35708 <at> debbugs.gnu.org
Subject: bug#35708: [27.0.50]: thingatpt.el, thing-at-point-looking-at redundant
Date: Wed, 15 May 2019 12:11:05 +0200
[Message part 1 (text/plain, inline)]
Am 15.05.19 um 08:59 schrieb Andreas Röhler:
>
> Am 14.05.19 um 16:34 schrieb npostavs <at> gmail.com:
>> Andreas Röhler <andreas.roehler <at> easy-emacs.de> writes:
>>
>>>> Hmm, current thing-at-point-looking-at might be slow with large
>>>> buffers. The slightly modified test should reveal it:
>>>>
>>>> (ert-deftest thing-at-point-looking-at-2 ()
>>>>    (with-temp-buffer
>>>>      (insert "1abcd 222abcd")
>>>>      (dotimes (_ 99999) (insert " asdf "))
>>>>      (goto-char (point-min))
>>>>        (search-forward "2ab")
>>>>        (should (thing-at-point-looking-at "2abcd"))
>> Yes, in this case, since the loop over looking-at only needs to iterate
>> twice, so it will be faster.  But what about when there is no match?
>> E.g.,
>>
>> (with-temp-buffer
>>    (dotimes (_ 99999) (insert " asdf "))
>>    (goto-char (point-max))
>>    (list :ar-regexp-atpt (benchmark-run (ar-regexp-atpt "foo"))
>>          :thing-at-point-looking-at (benchmark-run 
>> (thing-at-point-looking-at "foo"))))
>>
>>> Another fix, as a bug showed up when testing (ar-regexp-atpt "[a-z]+"):
>>>
>>> (defun ar-regexp-atpt (regexp)
>>>    "Return t if REGEXP matches at or before point, nil otherwise.
>>>
>>> Changes match-data"
>>>    (save-excursion
>>>      (if (looking-at regexp)
>>>      (while
>>>          (and (not (bobp))
>>>           (or (progn (backward-char) (looking-at regexp))
>>>               (forward-char 1))))
>>>        (while (not (or (bobp) (backward-char) (looking-at regexp))))
>>>        (ar-regexp-atpt regexp))
>> What's this recursive call for?  It triggers (error "Lisp nesting
>> exceeds ‘max-lisp-eval-depth’") in the benchmark above.
>>
>>>      (looking-at regexp)))
>
>
> The recursive call needed a guard:     (unless (bobp)
>
> It is called after function went backward while not looking-at matches,
>
> Now the result for the 99999 is
>
> (:ar-regexp-atpt (0.774574453 0 0.0) :thing-at-point-looking-at 
> (0.000798669 0 0.0))
>
>
> The fixed form:
>
>
Make sure match pos includes cursor pos:


(defun ar-regexp-atpt (regexp)

  "Return t if REGEXP matches at or before point, nil otherwise.
Changes match-data"
  (save-excursion
    (let ((orig (point)))
      (if (looking-at regexp)
      (while
          (and (not (bobp))
           (or (progn (backward-char) (looking-at regexp))
               (forward-char 1))))
    (while (not (or (bobp) (backward-char) (looking-at regexp))))
    (unless (bobp) (ar-regexp-atpt regexp)))
      (and
       (looking-at regexp)
       (<= (match-beginning 0) orig)
       (>= (match-end 0) orig)))))


[Message part 2 (text/html, inline)]

This bug report was last modified 5 years and 362 days ago.

Previous Next


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