Bug when changing buffer/window and a not finished isearch is active

[klaus.berndl at sdm.de]

Forgot to mention: I'm using Xemacs 21.4.14...

klaus.berndl at sdm.de wrote:
> Hi,
> 
> please try the following:
> 
> 1) Start XEmacs with -vanilla
> 2) Devide *scratch*-buffer into two windows
> 3) insert (setq debug-on-error t) and hit C-x C-e
> 4) Do C-s <a word in the scratch-buffer>. Do not finish the isearch,
>    i.e. do not RET or C-g so ensure the found word appears highlighted
> 5) Click with middle mouse-button in the other window ==> You get an
> error 6) Go with C-x b to the *Backtrace*-buffer
> 8) Now hit q
> 9) Now do M-x and then hit TAB for example. then you should get:
> 
> Signaling: (args-out-of-range #<buffer " *Minibuf-0*"> 8 178)
> set-extent-endpoints(#<extent [detached) 0x1d551c4 from buffer
> *scratch*> 8 178 #<buffer " *Minibuf-0*">)
> isearch-highlight(8 178)
> isearch-update()
> isearch-search-and-update()
> isearch-process-search-string(" " "^I")
> isearch-process-search-char(#<keypress-event tab>)
> #<compiled-function nil "...(21)" [this-command last-command-event
> event isearch-whitespace-chars event-to-character character-to-event
> ?\ isearch-process-search-char] 2 365487 nil>()
> call-interactively(isearch-printing-char)
> read-minibuffer-internal("Find file: ")
> byte-code("..." [standard-output standard-input prompt recursion-depth
> minibuffer-depth t read-minibuffer-internal] 2)
> read-from-minibuffer("Find file: " "c:\\tmp\\" #<keymap
> read-file-name-map size 2 0xcec> nil file-name-history nil nil)
> read-file-name-2(file-name-history "Find file: " nil nil nil nil
> read-file-name-internal)
> read-file-name-1(file file-name-history "Find file: " nil nil nil nil
> read-file-name-internal)
> read-file-name("Find file: " nil nil nil)
> call-interactively(find-file)
> 
> 
> For me this looks like an not wanted behavior - it seems that
> changing window rsp. buffer (either with mouse or not - i have not
> tested this exactly) when 
> a not finished isearch is "open" cause some errors.... IMHO XEmacs
> should be able to handle this error free...
> 
> Ciao,
> Klaus
> 
> Klaus Berndl                     mailto: klaus.berndl at sdm.de
> sd&m AG                         http://www.sdm.de
> software design & management
> Carl-Wery-Str. 42, 81739 Muenchen, Germany
> Tel +49 89 63812-392, Fax -220