Lisp HUG Maillist Archive

Lispworks bug reports

So, did anyone else file three bug reports the first day they got LWL
or am I special and/or nit-picky?

Lispworks Pro for Linux, 5.0.1.  Debian GNU/Linux, mostly "stable",
some "testing".  Motif library version: 2001 "@(#)GNU/LessTif Version
2.1 Release 0.94.4"

Can anyone else reproduce:

1) DESCRIPTION: Warning in TTY window when clicking on tabs

Clicking on any tab (e.g Listener or Output in a Listener window, or
Text, Output, Buffers, etc in an Editor window) prints a warning in
the TTY window:

  Warning: Couldn't traverse-to #<CAPI::CAPI-GENERIC-TAB-LAYOUT
  #<CAPI:TAB-LAYOUT [2 itmes] 2008D22B> 2009DB23> because widget is
  not navigable.

2) DESCRIPTION: Keyboard traversal of IDE menus is broken

Start LWL
Focus any window
Press Alt-W
The menu comes up as if you'd pressed Alt-W then W again, i.e. with
"Windows" already chosen

or

Start LWL
Focus any window
Press Alt-W
Play with the cursor keys.  You'd expect that the menus would go left,
go right, pop up, pop down, as appropriate, but after fiddling for a
very short time, they stop doing that.

Note that I have left "Alt-key" bound to Alt_L and left "Windows-key"
bound to Meta_L; this may be different from most X users:

> % xmodmap -p
> xmodmap:  up to 3 keys per modifier, (keycodes in parentheses):
>
> shift       Shift_L (0x32),  Shift_R (0x3e)
> lock        Caps_Lock (0x42)
> control     Control_L (0x25),  Control_R (0x6d),  Control_R (0x75)
> mod1        Alt_L (0x40),  Alt_L (0x7d)
> mod2        Num_Lock (0x4d)
> mod3        Meta_L (0x73),  Meta_L (0x9c),  Meta_R (0x74)
> mod4
> mod5        Mode_switch (0x5d),  ISO_Level3_Shift (0x7c)

3) DESCRIPTION: Error caught inside Xt event loop

Start LWL
focus the Podium
Do this several times (sometimes I get the error on the 2nd time,
sometimes the 4th; not sure what the upper bound is):
  Click Works
  Press cursor-up six times
  press cursor-right ("File" menu pops up)
  press Escape

error printed to TTY:

> **********************************************************************
> *                 Error caught inside Xt event loop                  *
> **********************************************************************
[ long error message snipped ]

Thanks for your eyeballs.

-- Larry Clapp


Re: Lispworks bug reports


On Apr 10, 2007, at 5:43 PM, Larry Clapp wrote:

> So, did anyone else file three bug reports the first day they got LWL
> or am I special and/or nit-picky?


I think I filed something close when I first got LWM ;^)

Basically you'll come to realize that LispWorks is like democracy -  
it not perfect, but its way ahead of whatever is in second place.

Naturally we'd all like our pet bugs to be highest priority, but at  
the end of the day, as long as LispWorks Ltd. can keep LispWorks  
running on each new release of the three major OS platforms its a big  
winner.

regards,

Ralph


Raffael Cavallaro, Ph.D.
raffaelcavallaro@mac.com


Re: Lispworks bug reports

Larry Clapp wrote:
> So, did anyone else file three bug reports the first day they got LWL
> or am I special and/or nit-picky?
>
> Lispworks Pro for Linux, 5.0.1.  Debian GNU/Linux, mostly "stable",
> some "testing".  Motif library version: 2001 "@(#)GNU/LessTif Version
> 2.1 Release 0.94.4"
>
> Can anyone else reproduce:
>
>   

Perhaps the problem is with LessTif.  On FreeBSD LW runs much better with
openmotif.

Wade


Re: Lispworks bug reports

On 10/4/07 22:43, Larry Clapp wrote:

> So, did anyone else file three bug reports the first day they got LWL
> or am I special and/or nit-picky?

Larry,

I licensed the 32-bit Mac OS X release just before Easter, and have reported
two so far. One of the penalties of getting so many copies out (at the
conference and with the special pricing deal, which finally swung me) is
that so many newbies come along and break things!

> Can anyone else reproduce:
> 
> 1) DESCRIPTION: Warning in TTY window when clicking on tabs

Not seen on Mac OS X 'native' (Cocoa IDE).

> 2) DESCRIPTION: Keyboard traversal of IDE menus is broken
> 3) DESCRIPTION: Error caught inside Xt event loop

Neither relevant to Mac OS X 'native' (Cocoa IDE).

My two, which seem OS-specific, have both been reported in detail to
LispWorks:

A) DESCRIPTION: Print failure on PowerPC and crash on Intel Macs:
On G5 DP running Mac OS X 10.4.9, printing to networked PostScript laser
printer or to PDF file consistently results in correct number of pages
printed, but only the first page to have content (which appears completely
correct) - second and subsequent pages are completely blank, although
counted correctly.
Went to check this on an Intel Mac and all attempts to print on that,
whether to same printer or to PDF, result in crash which freezes LispWorks
with 'messy' windows (has to be force-quit to clean up).
Workaround: print source files using another editor (BBEdit, Smultron, etc.)

B) DESCRIPTION: Freeze (local) in sys:get-folder-path on Mac OS X when used
in capi:prompt-for-file, on both PowerPC and Intel Macs running Mac OS X
10.4.9
The following use of sys:get-folder-path consistently (100%) results in an
application freeze when run as a standalone, either launched from the Finder
or from Application Builder, although the code runs fine when run within
LispWorks environment.

(capi:prompt-for-file "URM program:"
;   :pathname "~/Documents" ;; works fine, but
    :pathname (sys:get-folder-path :common-documents) ;; freezes dialog when
           ;; run as standalone app, although passing string or nil are fine
    :ok-check nil
    :filter "*.txt;*.text"
    :filters '("Text Documents" "*.txt;*.text")
    :if-exists :ok
    :if-does-not-exist :error
    :owner nil
    :continuation nil
    :operation :open
)

Workaround: don't use sys:get-folder-path, but insert your own path string.

Those said, I am absolutely delighted with LispWorks on Mac OS X. I come as
a fugitive from the long and sorry death of Mac Common Lisp (and not having
written any Lisp code for many years as a result), and wholeheartedly
recommend LispWorks. I have written from scratch a small app with light CAPI
use and produced a Universal Binary in but a few hours, despite taking some
time to track down the second of those bugs.

Howard.

Dr Howard Oakley
The Works columnist for MacUser magazine (UK)
http://www.macuser.co.uk/
http://www.howardoakley.com/





Updated at: 2020-12-10 08:46 UTC