[mapserver-dev] plugin renderer support status update

[Stephen Woodbridge]

thomas bonfort wrote:
> hi Steve(s)
> 
> On Fri, Dec 26, 2008 at 16:30, Steve Lime <Steve.Lime at dnr.state.mn.us> wrote:
>> A nice Christmas present indeed! You'll have to let devs know how we
>> can help.
> I've got a miriad of questions that would need answering. I'm hoping
> this can be a topic of the code sprint as I feel they would be best
> addressed without having to go through the lag of email.
> For now, it might be a good idea to ask ourselves about the future of
> the pdf/svg output in future releases, as the current drivers seem to
> have fallen behind and lack maintanance. There has been some talking
> about dropping pdflib for another pdf backend: any update on this? I
> also have the feeling that adding a native kml outputformat would be a
> trifle with the plugin approach.

Well regarding futures ... I'm not sure if your changes get us any 
closer to good quality print output, but it might from what I saw on the 
pdf output you generated. Print is an issue that comes up repeatedly on 
the list and most people that gotten anything useful to work have had to 
jump through all kinds of hoops to get there doing things like gaint 
image files, alternate mapfiles designed specifically around print 
output, etc. I understand that there is more to the printed map than 
just the map image, but we certainly need to be able to product a high 
quality map rendering as a starting point.

-Steve W

>> Can the OpenGL implementation use this approach?
> I certainly hope so. I doubt it could be used as-is, so this will be a
> good moment to work on the api the renderers should be presenting to
> mapserver.
> Toby, Jonathan, I'm not sure how we can get in sync on this. I'm
> usually on #mapserver irc, but mostly in european daytime hours.
> 
>> I'm not sure what your time line looks like but I would think an addition
>> like this plus the other things in the queue at the moment hint more at a
>> 6.0 release than 5.4. Curious what others think...
> I definitely think so to.
> This could be the time to break some backwards compatibility with
> respect to type vector symbols:
>  * switch their rotation direction: their angle is currently treated
> clockwise, all other symbols are anticlockwise (or the other way
> around, I don't remember)
>  * I'd also like to have the vector symbols be centered on their (0,0)
> point, which would allow preciser placement. thoughts... ?
> 
> cheers,
> thomas
> _______________________________________________
> mapserver-dev mailing list
> mapserver-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/mapserver-dev