gEDA-dev: PCB testers needed

[Bernd Jendrissek]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Mon, Apr 30, 2007 at 11:22:30PM -0400, Dan McMahill wrote:
> Thanks for trying it out.

Oh, sorry for not responding with more feedback.  See the .signature!

> >>1) menus/hotkeys which worked in pcb-20070208 with the gtk HID and
> >>are broken now
> >
> >Redraw 'R' seems to have disappeared?
> 
> I can't seem to figure out where that was defined before.  Was it in a
> menu?

I think I may be misremembering 'R' - I don't know if it was ever
defined.  Maybe I'm just confusing it with gschem.

> >>2) any other wierd behaviour relating to menus or hotkeys that is 
> >>different from pcb-20070208.
> >
> >$ pcb nanoweld-alt.pcb
> >ghid_load_menus():  should process Mouse section here
> 
> yep, I need to finish that bit.

It's silent now...

> >unknown action `EditLayerGroups'
> >unknown action `EditLayerGroups'
> >
> >Not sure what that means, but the 'Edit Layer Groups' option in the View
> >menu doesn't seem to do anything (other than spew the warning to
> >stderr).
> 
> fixed.

Thanks.

> >Flipping the view to solder side is broken: I can flip via the View menu
> >(more zooms and view changes), but not with keyboard shortcuts.
> >Pressing Tab makes pcb whine about "ksym 65289 (0xff09) has not been
> >defined", and shift-Tab is the same, except with 65056 (0xfe20).
> 
> fixed.

Hmm, I haven't updated since a while ago, but neither Tab nor shift-Tab
do anything?

> >And enabling both flips at the same time is totally borked - I have
> >arbitrary-angle silkscreen lines across my layout.
> 
> I think it is fixed.  It is at least closer.  I didn't have an example
> with arbitrary-angle silk, but I did have an SOIC where the arc was
> drawn the wrong way and that is fixed.  Let me know if it is better
> for you too.

I don't think I explained very well, but it's moot, as my layout now
flips and spins correctly AFAICT.  What I meant was that AFTER flipping,
my layout would have corrupt, arb-angle lines.  Like I said, it's fixed
now.

> >>3) does this build with gtk >= 2.8.0 and if so, does the "Center()" 
> >>action (bound to 'C') work?
> >
> >
> >$ rpm -q gtk+
> >gtk+-2.8.3-0
> >
> >Yes, 'C' works, it centers the board to the current point, but it seems
> >to warp the cursor to a place that is neither the old or new position of
> >the desired center.
> 
> ok.  Thats what I was afraid of.   It is broken *without* the flip_x or
> flip_y checked? 

Yes, the cursor would warp to a third random-looking point even without
any flips or spins checked.  But again moot, as it's fixed now.

> I found that on my older system 'C' and also where the cursor warps to
> was totally messed up if the board was flipped in either x or y.  I've
> fixed that for me.  Can you see if your problem still persists?  It if
> does then it is surely a case of the wrong arguments in this line of
> src/gtk/gtkhid-main.c:
> 
>  gdk_display_warp_pointer (display, screen, Vx2 (x), Vy2 (y));

A minor nit:

When I press 'C' to center the layout, and I have an xterm obscuring the
center of the view (I'm writing this mail with pointer focus, playing
with PCB behind the xterm), then the center doesn't happen.  The warp
does occur, putting the mouse onto my xterm (giving it focus).  The NEXT
time I pan the layout, it suddenly jumps to where the center should have
put it.  But then when I right-click-drag to pan, there's an odd jump
again.

> I _think_ I have this all fixed.

I agree; it's certainly beyond my 5-minute ability to tell the
difference between fixed and borked!

- -- 
> BTW, sometimes the lack of a specific response indicates *agreement*.
 
Just in case you thought I was agreeing with you.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFGOGXWwyMv24BBd/gRAjbkAKCWHdwESSQaLo5sxzA5NxYuxdiIMgCfa3au
hnIyjtOJrhtzMbj7Ezy7i+g=
=GaLO
-----END PGP SIGNATURE-----