You are viewing a plain text version of this content. The canonical link for it is here.
Posted to batik-users@xmlgraphics.apache.org by Archie Cobbs <ar...@dellroad.org> on 2004/11/04 16:18:35 UTC

Strange rendering glitch

Hi all,

We have been using Batik for a while and it's been working beautifully.

The other day however we saw a strange new problem. It hasn't happened
again but I'm wondering if anyone else has seen this kind of thing before.

Basically we have a floor plan and we dynamically (via Java) fill in
the room areas with various colors, by setting the "fill" attribute
on the room's <polygon> to e.g. "url(#roomColor.Green)", where in the
<defs> section we've defined the various colors, e.g.:

  <linearGradient id="roomColor.Green">
    <stop offset="0%" stop-color="#33ff33"/>
  </linearGradient>

What we saw the other day was these two anomalies on the screen:

 1. A couple of rooms (call them X and Y) that were supposed to be grey
    were 1/2 the right color (top half) but the bottom half was painted
    with a "snapshot" of another room (call it Z). That is, in the bottom
    half of X and Y you could see Z's room number along with it's green
    coloring: a snapshot of Z rendered within the boundaries of X and Y.

    It's as if the rendering engine forgot to reset it's buffer after
    painting Z before painting X and Y, or something like that.

  2. Next to, but outside of, room Z was a little rectangle of green
     pixels that were completely out of place (nothing should have
     been drawn there).

This happened on two different machines that happened to be displaying
the same map, so it doesn't seem to be a "random glitch".

I am pretty certain that we are performing all updates within the
update manager thread.

Thanks for any insights,

-Archie

P.S. Should I file a bug even though I don't have a test case (yet)?

__________________________________________________________________________
Archie Cobbs      *        CTO, Awarix        *      http://www.awarix.com


*
Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended
recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
*


---------------------------------------------------------------------
To unsubscribe, e-mail: batik-users-unsubscribe@xml.apache.org
For additional commands, e-mail: batik-users-help@xml.apache.org


Re: Strange rendering glitch

Posted by Archie Cobbs <ar...@dellroad.org>.
Thomas DeWeese wrote:
>  > The other day however we saw a strange new problem. It hasn't
>  > happened again but I'm wondering if anyone else has seen this kind
>  > of thing before.
> 
>  > A couple of rooms (call them X and Y) that were supposed to be grey
>  > were 1/2 the right color (top half) but the bottom half was painted
>  > with a "snapshot" of another room (call it Z)
> 
>     Were you using JDK 1.5?  I've very occasionally seen redraw
> glitches with 1.5.  I haven't looked at them in any depth as they
> haven't been anywhere near reproducible, and it was in beta at the
> time.

Sorry, I should have provided that info. This is with JDK 1.4.2p5 on
Debian Linux.

>  > P.S. Should I file a bug even though I don't have a test case (yet)?
> 
>     Feel free to, it can be a good place to collect information if
> nothing else.

Will do, thanks.

-Archie

__________________________________________________________________________
Archie Cobbs      *        CTO, Awarix        *      http://www.awarix.com

---------------------------------------------------------------------
To unsubscribe, e-mail: batik-users-unsubscribe@xml.apache.org
For additional commands, e-mail: batik-users-help@xml.apache.org


Re: Strange rendering glitch

Posted by Thomas DeWeese <Th...@Kodak.com>.
Hi Archie,

 > We have been using Batik for a while and it's been working
 > beautifully.

    Glad to hear it.

 > The other day however we saw a strange new problem. It hasn't
 > happened again but I'm wondering if anyone else has seen this kind
 > of thing before.

 > A couple of rooms (call them X and Y) that were supposed to be grey
 > were 1/2 the right color (top half) but the bottom half was painted
 > with a "snapshot" of another room (call it Z)

    Were you using JDK 1.5?  I've very occasionally seen redraw
glitches with 1.5.  I haven't looked at them in any depth as they
haven't been anywhere near reproducible, and it was in beta at the
time.

 > P.S. Should I file a bug even though I don't have a test case (yet)?

    Feel free to, it can be a good place to collect information if
nothing else.

Archie Cobbs wrote:

> Hi all,
> 
> We have been using Batik for a while and it's been working beautifully.
> 
> The other day however we saw a strange new problem. It hasn't happened
> again but I'm wondering if anyone else has seen this kind of thing before.
> 
> Basically we have a floor plan and we dynamically (via Java) fill in
> the room areas with various colors, by setting the "fill" attribute
> on the room's <polygon> to e.g. "url(#roomColor.Green)", where in the
> <defs> section we've defined the various colors, e.g.:
> 
>   <linearGradient id="roomColor.Green">
>     <stop offset="0%" stop-color="#33ff33"/>
>   </linearGradient>
> 
> What we saw the other day was these two anomalies on the screen:
> 
>  1. A couple of rooms (call them X and Y) that were supposed to be grey
>     were 1/2 the right color (top half) but the bottom half was painted
>     with a "snapshot" of another room (call it Z). That is, in the bottom
>     half of X and Y you could see Z's room number along with it's green
>     coloring: a snapshot of Z rendered within the boundaries of X and Y.
> 
>     It's as if the rendering engine forgot to reset it's buffer after
>     painting Z before painting X and Y, or something like that.
> 
>   2. Next to, but outside of, room Z was a little rectangle of green
>      pixels that were completely out of place (nothing should have
>      been drawn there).
> 
> This happened on two different machines that happened to be displaying
> the same map, so it doesn't seem to be a "random glitch".
> 
> I am pretty certain that we are performing all updates within the
> update manager thread.
> 
> Thanks for any insights,
> 
> -Archie
> 
> P.S. Should I file a bug even though I don't have a test case (yet)?

    Feel

---------------------------------------------------------------------
To unsubscribe, e-mail: batik-users-unsubscribe@xml.apache.org
For additional commands, e-mail: batik-users-help@xml.apache.org