RE: Proposed fix for 687421; page cleared in x11alpha

Raph,

>It does not. However, that's not all that surprising, because I don't
>think the forwarding device gets used much (if at all) in the standard
>devices we use for testing. I guess that, in order to really validate
>this patch, we should find out under exactly what circumstances a
>forwarding device gets installed.

A quick grep for bbox_device and device_bbox shows that besides gdevxini.c
and gdevx.c, the major users seem to be gdevps.c, gdevvec.c, and gximag3x.c.
I do not know what other forwarding devices that we have (pdf14, clipping,
overprinting, what else?).

The regression testing does include several examples of pdf14 transparency,
clipping, and overprinting.


>Alternatively, I could just check in the patch and see if we get any squeals

Obviously the x11 device is important on *nix and this problem is a major
flaw in the x11 device.

The biggest worry is issue that you found about Peter's change back in 4.8.


Dan


-----Original Message-----
From: Raph Levien [mailto:[email protected]]
Sent: Saturday, September 18, 2004 8:50 PM
To: [email protected]
Subject: Re: [gs-code-review] Proposed fix for 687421; page cleared in
x11alpha (xefitra)


On Fri, 17 Sep 2004 00:03:32 -0700, Dan Coby <[email protected]> wrote:
> 
> Raph,
> 
> I also have no idea about why gx_forward_get_page_device had logic
> for returning the forwarding device if the target device was the
> page device.  It does look a very strange piece of logic.
> 
> Does this change make any differences in the regression testing?

It does not. However, that's not all that surprising, because I don't
think the forwarding device gets used much (if at all) in the standard
devices we use for testing. I guess that, in order to really validate
this patch, we should find out under exactly what circumstances a
forwarding device gets installed.

Alternatively, I could just check in the patch and see if we get any squeals :)

Raph