There is a problem in XEmacs (21.4) with the current implementation of `slime-complete-symbol*'. On completing a symbol, the text gets truncated to the length of the unambiguous prefix of all matches -- This is entirely an artifact of the XEmacs completion mechanism, and has nothing to do with SLIME's code. I'm appending a patch which works around this issue when the situation is applicable and which I believe is harmless in all other scenarios.
I take it there is no interest in the earlier patch I posted in http://permalink.gmane.org/gmane.lisp.slime.devel/5668 in reviving SLIME's own earlier unambiguous-prefix behaviour.
Who are the active committers on SLIME at this time? Can one conclude that if a patch has not been committed in a few weeks, there is no interest in it and it will likely not be included in slime? I had posted a few bug fixes on this list and would like to know what to expect on those issues, thanks. -- Madhu
--- slime.el.~1.794.~ 2007-06-28 17:57:51.000000000 +0530 +++ slime.el 2007-08-04 18:15:16.000000000 +0530 @@ -6128,10 +6128,12 @@ "Can't find completion for "%s"" prefix) (ding) (slime-complete-restore-window-configuration)) - (goto-char end) - (insert-and-inherit completed-prefix) - (delete-region beg end) - (goto-char (+ beg (length completed-prefix))) + (cond ((> (length completed-prefix) (- end beg)) + (goto-char end) + (insert-and-inherit completed-prefix) + (delete-region beg end) + (goto-char (+ beg (length completed-prefix)))) + (t nil)) (cond ((and (member completed-prefix completion-set) (= (length completion-set) 1)) (slime-minibuffer-respecting-message "Sole completion")
[ARGH] Fixing a wrong URL
* Madhu m31wejfoi0.fsf@robolove.meer.net :
| There is a problem in XEmacs (21.4) with the current implementation of | `slime-complete-symbol*'. On completing a symbol, the text gets | truncated to the length of the unambiguous prefix of all matches -- This | is entirely an artifact of the XEmacs completion mechanism, and has | nothing to do with SLIME's code. I'm appending a patch which works | around this issue when the situation is applicable and which I believe | is harmless in all other scenarios. | | I take it there is no interest in the earlier patch I posted in | http://permalink.gmane.org/gmane.lisp.slime.devel/5668 in reviving | SLIME's own earlier unambiguous-prefix behaviour.
As usual I got the URL wrong. The patch I refer to above is archived here: http://permalink.gmane.org/gmane.lisp.slime.devel/5921
| Who are the active committers on SLIME at this time? Can one conclude | that if a patch has not been committed in a few weeks, there is no | interest in it and it will likely not be included in slime? | I had posted a few bug fixes on this list and would like to know what to | expect on those issues, thanks. -- Madhu | | | --- slime.el.~1.794.~ 2007-06-28 17:57:51.000000000 +0530 | +++ slime.el 2007-08-04 18:15:16.000000000 +0530 | @@ -6128,10 +6128,12 @@ | "Can't find completion for "%s"" prefix) | (ding) | (slime-complete-restore-window-configuration)) | - (goto-char end) | - (insert-and-inherit completed-prefix) | - (delete-region beg end) | - (goto-char (+ beg (length completed-prefix))) | + (cond ((> (length completed-prefix) (- end beg)) | + (goto-char end) | + (insert-and-inherit completed-prefix) | + (delete-region beg end) | + (goto-char (+ beg (length completed-prefix)))) | + (t nil)) | (cond ((and (member completed-prefix completion-set) | (= (length completion-set) 1)) | (slime-minibuffer-respecting-message "Sole completion") |
Updating the patch at:
* Date: Sat, 04 Aug 2007 19:35:15 +0530 Message-ID: m3wswbe6p0.fsf@robolove.meer.net Archived-At: http://permalink.gmane.org/gmane.lisp.slime.devel/6410
| | There is a problem in XEmacs (21.4) with the current implementation | | of `slime-complete-symbol*'. On completing a symbol, the text gets | | truncated to the length of the unambiguous prefix of all matches -- | | This is entirely an artifact of the XEmacs completion mechanism, and | | has nothing to do with SLIME's code. I'm appending a patch which | | works around this issue when the situation is applicable and which I | | believe is harmless in all other scenarios.
[This is not a "consensus issue" but one of fixing an error when using that XEmacs. This patch is NOT recommended to be applied as is to CVS as it prints a debug message to the minibuffer --- the hope is that debug message would convince the committer]
ndex: contrib/slime-c-p-c.el =================================================================== RCS file: /project/slime/cvsroot/slime/contrib/slime-c-p-c.el,v retrieving revision 1.4 diff -u -r1.4 slime-c-p-c.el --- contrib/slime-c-p-c.el 31 Aug 2007 22:16:11 -0000 1.4 +++ contrib/slime-c-p-c.el 2 Sep 2007 07:50:48 -0000 @@ -43,10 +43,13 @@ "Can't find completion for "%s"" prefix) (ding) (slime-complete-restore-window-configuration)) - (goto-char end) - (insert-and-inherit completed-prefix) - (delete-region beg end) - (goto-char (+ beg (length completed-prefix))) + (cond ((> (length completed-prefix) (- end beg)) + (goto-char end) + (insert-and-inherit completed-prefix) + (delete-region beg end) + (goto-char (+ beg (length completed-prefix)))) + (t (message "completed-prefix=%s prefix=%s" + completed-prefix prefix))) (cond ((and (member completed-prefix completion-set) (slime-length= completion-set 1)) (slime-minibuffer-respecting-message "Sole completion")
* Madhu [2007-09-02 10:07+0200] writes:
Updating the patch at:
- Date: Sat, 04 Aug 2007 19:35:15 +0530 Message-ID: m3wswbe6p0.fsf@robolove.meer.net Archived-At: http://permalink.gmane.org/gmane.lisp.slime.devel/6410
[...]
[This is not a "consensus issue" but one of fixing an error when using that XEmacs. This patch is NOT recommended to be applied as is to CVS as it prints a debug message to the minibuffer --- the hope is that debug message would convince the committer]
I applied something similar. Also added a variable for the "consensus issue". Didn't test it, though.
Helmut.
Madhu enometh@meer.net writes:
Who are the active committers on SLIME at this time? Can one conclude that if a patch has not been committed in a few weeks, there is no interest in it and it will likely not be included in slime?
No, it mostly means they can't afford any time on Slime for the moment. But you as patch submitter should make sure that your submitted patches aren't missed in the next round of active commitment.
-T.
-
Helmut Eller -
Madhu -
Tobias C. Rittweiler