I got a couple of ideas; but the first one is "low stress"...
1. Reduce FatorySPI use to *just* service discovery ... so an
AuthorityFactoryServiceProvider
2. Have create methods on AuthorityFactoryServiceProvider to create the
existing AuthorityFactory instances
3. The AuthorityFactoryServicePro
Hello Jody
Jody Garnett a écrit :
> Martin are you still interested in breaking apart the "service
> discovery" use of factory spi form the management of factory instances?
It is a good time to revisit the factory system since I'm in cleaning stage
right now. I'm sorry that it caused you so mu
Mixed success here Martin :-)
UDig trunk is able to draw using epsg-hsql - and the results look good.
I have been forced to turn off the various global hints as they mess up
our ability to find stuff on the classpath. But for today this is worth
a commit :-)
Jody
> Jody Garnett a écrit :
>> Th
Jody Garnett wrote:
> Please note I am making use of the Hints.FORCE_LONGITUDE_FIRST_AXIS_ORDER
> hint. So it makes sense that the
> EPSG factory is not found on the classpath when requested with this Hint.
>
> So what is the story again on how these hints are supposed to be used?
> Is this a ca
Okay making some progress here -
1. You may not of done significant work in the last year - but we are
running in a new version of Eclipse; and thus a new version of OSGi
which is trying to buckle down even harder on the kind of hacking we
like to do with FactorySPI.
2. FactoryRegistry.scanForPl
Jody Garnett a écrit :
> This email is a starting place; I will comment as I learn more. Martin I
> assume nothing has changed on your end?
No I haven't done any significant work in EPSG factory for a year or so...
I will look if I have idea.
Martin
So I am back to trying out epsg-hsql with udig.
Last time I tried I was able to see the epsg authority but unable to
retrieve a list of known codes. This time I am getting a very deep stack
trace with way more classloader fun than I am used to debugging.
This email is a starting place; I will c
