[ESS] forming completions ... doesn't complete

Martin Maechler maechler at stat.math.ethz.ch
Wed Oct 19 10:39:40 CEST 2005


>>>>> "tony" == A J Rossini <blindglobe at gmail.com>
>>>>>     on Tue, 18 Oct 2005 17:57:32 +0200 writes:

    tony> Just let us know when it happens again.  I can't think
    tony> of any reason why 2.2.0 vs. 2.1.0 would stop it from
    tony> happening (i.e. changes affecting things).

I strongly support Tony's point:

Dan, it's hardly imaginable that the differences between 2.1.0
and 2.2.0 would be the reason for the bug disappearing.

Rather I can imagine that you only *thought* you had used 
ESS 5.2.10, but rather you were still using ESS 5.2.7.
And IIRC there *were* completion problems in 5.2.7 that have
been fixed subsequently, particularly this one (from ESS.../lisp/ChangeLog):

>> 2005-09-07  Martin Maechler  <maechler at stat.math.ethz.ch>
>> 
>> 	* ess-inf.el (ess-complete-filename): work around Xemacs bug:
>> 	(comint-dynamic-complete-filename) fails in Xemacs 21.4.17

and you report that you have used Xemacs 21.4.17

Martin

    tony> On 10/18/05, Dan Davison <davison at uchicago.edu> wrote:
    >> I would like to provisionally withdraw this bug
    >> report. My impression is that the problem does not occur
    >> with R 2.2.0.
    >> 
    >> I hope I haven't wasted anyone's time.
    >> 
    >> Thanks,
    >> 
    >> Dan
    >> 
    >> 
    >> On Fri, 14 Oct 2005, Dan Davison wrote:
    >> 
    >> >> I've not seen this.  I wish you could get a backtrace!
    >> >
    >> > A screen's worth of backtrace is below (tell me if I
    >> need to capture more > than this).
    >> >
    >> >> I'm assuming under R and Linux.  What version of Emacs
    >> or XEmacs?  > Sorry, XEmacs 21.4.17-1 (R and linux
    >> version information in original email > below).
    >> >
    >> > Thanks for any help, and let me know if I can do
    >> anything else to > characterise the problem.
    >> >
    >> > Dan
    >> >
    >> > # bind (newalist posn index searchlist) > (let*
    >> ((searchlist ...) (index 1) posn newalist) (while
    >> searchlist (setq > posn ...) (setq newalist ...) (setq
    >> index ...) (setq searchlist ...))  >
    >> (ess-write-to-dribble-buffer (format
    >> "ess-get-modtime-list: new alist of > length %d\n" ...))
    >> (setq ess-sl-modtime-alist newalist)) >
    >> ess-get-modtime-list() > (progn
    >> (ess-write-to-dribble-buffer "-->
    >> (ess-get-modtime-list)\n") > (ess-get-modtime-list)) >
    >> (if (or (not ess-sl-modtime-alist) ess-sp-change) (progn
    >> > (ess-write-to-dribble-buffer "-->
    >> (ess-get-modtime-list)\n") > (ess-get-modtime-list))
    >> (ess-write-to-dribble-buffer " using existing >
    >> ess-*-alist\n")) > # (unwind-protect ...)  >
    >> (save-excursion (set-buffer (process-buffer ...))
    >> (ess-make-buffer-current) > (ess-write-to-dribble-buffer
    >> (format "(get-object-list %s) .." name)) (if (or >
    >> ... ess-sp-change) (progn ... ...)
    >> (ess-write-to-dribble-buffer " using > existing
    >> ess-*-alist\n")) (let* (... ... ... result) (setq result
    >> ...) (while > ... ... ...) (setq ess-object-list ...)))
    >> > (or ess-object-list (save-excursion (set-buffer ...)  >
    >> (ess-make-buffer-current) (ess-write-to-dribble-buffer
    >> ...) (if ... ... ...)  > (let* ... ... ... ...)))  > #
    >> bind (name) > ess-get-object-list("R") > (if classname
    >> (ess-slot-names classname) (ess-get-object-list >
    >> ess-current-process-name)) > (if listname
    >> (ess-object-names listname) (if classname (ess-slot-names
    >> > classname) (ess-get-object-list
    >> ess-current-process-name))) > # bind (classname listname
    >> pattern full-prefix beg buffer-syntax end >
    >> comint-completion-addsuffix) > (let*
    >> ((comint-completion-addsuffix nil) (end ...)
    >> (buffer-syntax ...) (beg > ...) (full-prefix ...)
    >> (pattern full-prefix) (listname ...) (classname ...)  >
    >> (components ...)) (or (comint-dynamic-simple-complete
    >> pattern components) > (quote none))) > (if (memq
    >> (char-syntax ...) (quote ...)) (let*
    >> (... ... ... ... ... ... ...  > ... ...) (or ... ...)))
    >> > # bind (listcomp) > ess-complete-object-name() >
    >> run-hook-with-args-until-success(ess-complete-object-name)
    >> > #<compiled-function nil "...(4)"
    >> [run-hook-with-args-until-success >
    >> comint-dynamic-complete-functions] 2 >
    >> ("/usr/share/xemacs21/xemacs-packages/lisp/xemacs-base/comint.elc"
    >> . 52958) > nil>() >
    >> call-interactively(comint-dynamic-complete) > #
    >> (condition-case ... . error) > # (catch top-level ...)
    >> >
    >> >
    >> > On 10/9/05, Dan Davison <davison at uchicago.edu>
    >> wrote: >>> Dear ESS-help,
    >> >>>
    >> >>> For some time, we have noticed a problem with
    >> tab-completion in an R >>> process buffer (presumably
    >> many people are also aware of this):
    >> >>>
    >> >>> Sometimes, when performing the first tab-completion
    >> of a session, it >>> freezes with a message in the
    >> minibuffer saying "forming completions for >>>
    >> package:methods". Frequently, this does not occur
    >> however. The only >>> solution we have found is to kill
    >> the emacs process.
    >> >>>
    >> >>> Is there a solution/workaround?
    >> >>>
    >> >>> Thanks a lot,
    >> >>>
    >> >>> Dan
    >> >>>
    >> >>> ESS 5.2.10 (and previously with 5.2.7) >>> Debian
    >> sarge (and Fedora Core 4.0) linux >>> > version >>> _ >>>
    >> platform i386-pc-linux-gnu >>> arch i386 >>> os linux-gnu
    >> >>> system i386, linux-gnu >>> status >>> major 2 >>>
    >> minor 1.0 >>> year 2005 >>> month 04 >>> day 18 >>>
    >> language R
    >> >>>
    >> >>> ______________________________________________ >>>
    >> ESS-help at stat.math.ethz.ch mailing list >>>
    >> https://stat.ethz.ch/mailman/listinfo/ess-help
    >> >>>
    >> >
    >> >
    >> >> --
    >> >> best, >> -tony
    >> >
    >> >> blindglobe at gmail.com Muttenz, Switzerland. "Commit
    >> early,commit often, >> and commit in a repository from
    >> which we can easily roll-back your >> mistakes" (AJR,
    >> 4Jan05).
    >> >
    >> >
    >> >
    >> 


    tony> -- best, -tony

    tony> blindglobe at gmail.com Muttenz, Switzerland.  "Commit
    tony> early,commit often, and commit in a repository from
    tony> which we can easily roll-back your mistakes" (AJR,
    tony> 4Jan05).

    tony> ______________________________________________
    tony> ESS-help at stat.math.ethz.ch mailing list
    tony> https://stat.ethz.ch/mailman/listinfo/ess-help




More information about the ESS-help mailing list