Hi Paul,
I'd rather not open the thread can of worms right now. Once we
start going down the road of investigating if every method in JUMP is
thread safe, progress will probably come to a screeching halt. My
policy is that if I get concurrent modification exceptions, then I fix
them. I spent most of a month testing the new ThreadQueue and implied
changes to the RenderingManager and other code, and in the end, who
can tell the difference? It worked before. It works now.
Failing to disable event firing is a completely separate issue and can
cause unexpected side effects even without thread safety problems.
What Malte is pointing out, and I am agreeing with is the lack of
consistency in the API.
regards,
Larry
On 9/12/07, Paul Austin <[EMAIL PROTECTED]> wrote:
> Larry,
>
> As the list of styles on a layer is not thread safe you should disable
> firing events if you are changing the list of styles for a layer in any
> other thread than the rendering thread, otherwise you could get
> concurrent modification execptions when reading the list of styles.
>
> Paul
>
> Larry Becker wrote:
> >> But why should I turn off firing events for a CTS and not for a
> >> BasicStyle ?
> >>
> >
> > You're right. You shouldn't have to do that. Apparently the Layer
> > Tree needs a ColorThemingStyle, and the code was structured so that if
> > one is not present, it is always added. We should probably add this
> > issue to the bug tracker. The fix would be to simply always add a CTS
> > when BasicStyle and LabelStyle are added and to add support in Layer
> > via a getColorThemingStyle() method. There are getBasicStyle(),
> > getLabelStyle(), and getVertexSTyle() methods, but no getCTS method.
> > Why???
> >
> > regards,
> > Larry Becker
> >
> >
> > On 9/12/07, Malte Weller <[EMAIL PROTECTED]> wrote:
> >
> >> Hello Paul, hello Larry,
> >>
> >> sorry for my late response, but I'm working from 9 to 7 in Germany .....
> >> :-)
> >>
> >> But it is good to know that I get a response in the early evening on the
> >> same day if I have a question in the morning. :-D
> >>
> >> Paul,
> >> after I tried it with turned off firing events on the layer manager I
> >> could remove the first CTS without any problems.
> >> So thank's for that.
> >> But why should I turn off firing events for a CTS and not for a
> >> BasicStyle ?
> >>
> >> Greetings from Hannover, Germany on a fresh but not rainy wednesday,
> >> Malte
> >>
> >> Paul Austin schrieb:
> >>
> >>> Malte,
> >>>
> >>> Did you turn off firing events on the layer manager before starting to
> >>> make your changes?
> >>>
> >>> Paul
> >>>
> >>> Larry Becker wrote:
> >>>
> >>>
> >>>> Oh, I figured the first answer was a little too obvious. 8-)
> >>>>
> >>>> As to where the CTS object is coming from, I think it is
> >>>> LayerTreeModel.getChildren():
> >>>>
> >>>> if (parent instanceof Layer
> >>>> && ColorThemingStyle.get((Layer) parent).isEnabled()) {
> >>>> ...
> >>>>
> >>>> ColorThemingStyle.get() is:
> >>>>
> >>>> public static ColorThemingStyle get(Layer layer) {
> >>>> if ((ColorThemingStyle)
> >>>> layer.getStyle(ColorThemingStyle.class) == null) {
> >>>> ColorThemingStyle colorThemingStyle = new
> >>>> ColorThemingStyle(
> >>>> pickNonSpatialAttributeName(layer
> >>>>
> >>>> .getFeatureCollectionWrapper().getFeatureSchema()),
> >>>> new HashMap(), new
> >>>> BasicStyle(Color.lightGray));
> >>>> layer.addStyle(colorThemingStyle);
> >>>> }
> >>>> return (ColorThemingStyle)
> >>>> layer.getStyle(ColorThemingStyle.class);
> >>>> }
> >>>>
> >>>> It seems to always add a CTS if one is not found. A very strange side
> >>>> effect.
> >>>>
> >>>> regards,
> >>>> Larry
> >>>> On 9/11/07, Malte Weller <[EMAIL PROTECTED]> wrote:
> >>>>
> >>>>
> >>>>
> >>>>> Hi Larry,
> >>>>>
> >>>>> thanks for your reply.
> >>>>>
> >>>>> I know that the CTS added first should be removed first. ;-)
> >>>>> But I was only able to remove that after I added the second one.
> >>>>>
> >>>>> In my Thread I look at the Style-List of my new Layer if it holds a CTS
> >>>>> Object.
> >>>>> If so, I try to remove it because my new CTS Object should be used.
> >>>>> Surprisingly I could not remove it. It will added immediately at the
> >>>>> end of the List again.
> >>>>> For example at the beginning it was in the middle of the List and now I
> >>>>> find it at the end.
> >>>>> I can remove a BasicStyle Object without any problems.
> >>>>> It is only possible to remove the first CTS Object from the List after I
> >>>>> added my new one.
> >>>>>
> >>>>> I think, I should be able to remove the first CTS from the list without
> >>>>> any trouble like the BasicStyle and not only after I added a second one.
> >>>>> Maybe, the use of a Thread causes this problem ?
> >>>>>
> >>>>> Greetings from Hannover,
> >>>>> Malte
> >>>>>
> >>>>> Larry Becker schrieb:
> >>>>>
> >>>>>
> >>>>>
> >>>>>> Hi Malte,
> >>>>>>
> >>>>>> If I understand you correctly, I guess I'm not surprised at the
> >>>>>> fact that a Style added first would be removed first since the
> >>>>>> underlying data structure is an ArrayList. The javadoc for ArrayList
> >>>>>> says that list.remove(o) will delete the first occurance of o in list.
> >>>>>>
> >>>>>> regards,
> >>>>>> Larry Becker
> >>>>>>
> >>>>>> On 9/10/07, Malte Weller <[EMAIL PROTECTED]> wrote:
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>> Hello List,
> >>>>>>>
> >>>>>>> I use a Thread to create several Layers from a Database.
> >>>>>>> My run-Method calls two other Methods.
> >>>>>>> First one to create and add the layers.
> >>>>>>> Second one to create a ColorThemingStyle-Object to add this to the
> >>>>>>> created Layer.
> >>>>>>>
> >>>>>>> If the Layer holds a ColorThemingStyle-Object it should be removed and
> >>>>>>> the new one should be added.
> >>>>>>> By default my new Layers are holding a ColorThemingStyle-Object (Does
> >>>>>>> anybody know where this one will be added to the Layers Style List ?).
> >>>>>>>
> >>>>>>> Immediately after I removed the "default" CTS-Object from the List it
> >>>>>>> will be added again at the end.
> >>>>>>> So I cannot remove this "default" CTS-Object.
> >>>>>>> After I added my new CTS-Object it looks like that the "default"
> >>>>>>> CTS-Object is still be used instead.
> >>>>>>>
> >>>>>>> I found a workaround for that.
> >>>>>>> I add my new CTS-Object first and then I remove a CTS-Object with
> >>>>>>>
> >>>>>>> "layer.removeStyle(ColorThemingStyle.get(layer));
> >>>>>>>
> >>>>>>> Surprisingly the "default" CTS-Object has been removed and the new CTS
> >>>>>>> is shown.
> >>>>>>>
> >>>>>>> Can anybody varify this behaviour and has a explanation for it ?
> >>>>>>>
> >>>>>>> Greetings from Hannover,
> >>>>>>> Malte
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>> -------------------------------------------------------------------------
> >>>>>>> This SF.net email is sponsored by: Microsoft
> >>>>>>> Defy all challenges. Microsoft(R) Visual Studio 2005.
> >>>>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> >>>>>>> _______________________________________________
> >>>>>>> Jump-pilot-devel mailing list
> >>>>>>> Jump-pilot-devel@lists.sourceforge.net
> >>>>>>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>> -------------------------------------------------------------------------
> >>>>> This SF.net email is sponsored by: Microsoft
> >>>>> Defy all challenges. Microsoft(R) Visual Studio 2005.
> >>>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> >>>>> _______________________________________________
> >>>>> Jump-pilot-devel mailing list
> >>>>> Jump-pilot-devel@lists.sourceforge.net
> >>>>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>
> >>> -------------------------------------------------------------------------
> >>> This SF.net email is sponsored by: Microsoft
> >>> Defy all challenges. Microsoft(R) Visual Studio 2005.
> >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> >>> _______________________________________________
> >>> Jump-pilot-devel mailing list
> >>> Jump-pilot-devel@lists.sourceforge.net
> >>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
> >>>
> >>>
> >>>
> >> -------------------------------------------------------------------------
> >> This SF.net email is sponsored by: Microsoft
> >> Defy all challenges. Microsoft(R) Visual Studio 2005.
> >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> >> _______________________________________________
> >> Jump-pilot-devel mailing list
> >> Jump-pilot-devel@lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
> >>
> >>
> >>
> >>
> >
> >
> >
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Jump-pilot-devel mailing list
> Jump-pilot-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
>
--
http://amusingprogrammer.blogspot.com/
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Jump-pilot-devel mailing list
Jump-pilot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
