Strange Flex 3 Tree behaviour. A bug?

Hello Everyone!
I use the flex 3 Tree component with a custom data
descriptor.
The data-structure, which is processed by the data
descriptor,
allows that one node may have several parent-nodes.
In this case the node is being shown in tree several times -
once for every
parent-node.
The node itself may also have several children. (Some sort of
a simplified Graph structure.)
This is rendered well, but when I open the node in the tree,
the same node is automatically
opened by other parent-nodes. When i move the mouse-pointer
over the nodes-children. they are not highlighted, and the
respective children of some other parent-node are.
Any Ideas?
Thanks!

"newflexquest" <[email protected]> wrote in
message
news:gcfaoi$5i$[email protected]..
> Thats right, this behaviour takes plase only while using
the same object
> instance several times in the tree.
> The bad stuff is that i need exactly this :)
> I cannot destroy data-integrity by duclicating nodes and
all of their
> children, and their children and so on...
>
> I made a debug in Tree, List and ListBase classes. The
behaviour is as
> follows:
> 1. DisplayObjects are created for each appearance of the
data-object in
> the
> tree.
> 2. On mouse-over event the display-object is taken to
find corresponding
> data-object. Data-objects are stored in a map as
key-value pairs, where
> the key
> is object-id and the value is the object itself (plus
the coordinates on
> the
> screen).
>
> Therefore it is clear: if an item must appear twice in
the tree, related
> data-object may not be placed twice to the map (key
violation). So at the
> end
> we have at the background only ONE data-object for all
of the
> display-objects
> which represent duplication of the same item. And this
one object contains
> only
> one set of coordinates (of one related display object).
> It is not important on which of the duplicated display
objects you fire
> the
> mouse-over event. You will see the highlicht only on the
display-object,
> whichs
> coordinates are stored in the map.
>
> I hope the explanation is not heavy to read.
Write your own tree...?

Similar Messages

ISE 1.1.2 - strange GUI behaviour? bug?

Something has changed in GUI in the latest 1.1.2 version cause its not working properly.
In the first place I encountered an issue while adding new endpoint group.
I added some - then it refused to add new, rename old. I tried to add authorization profile and received such a notification (after 2 min of processing)
I patched the system (v3) and this problem disappeared but now I have another problem.
I added some authorization policies (in a number of 9) -> saved it and it show only part of it.
I can't add it againg cause it says its already created.
After restarting all the services I see only one policy (not even the defaults)
(tried different browsers - the same)
Has anyone had similar issues?
Should I downgrade to 1.1.1 ?
Its a pair of 3315.
After promoting the secondary to Administration role, I see 4 of the 10 policies and what is the most interesting the one (default) I deleted after installing the server.
I will deregister the pair and try them independently with different software version. (1.1.2patch3 vs 1.1.1patch5)
regards
Przemek

thx for fast reply
- resizing - did solve the problem
I tried:
- opera - didn't work at all
- FF (18.0.1) - issue as described earlier
- IE 9 - the same
- IE 8 - the same
- chrome - the same
I have my lab NFR ISE 1.1.2 (standalone) and have no such dificulties with any browser. I'm starting to think it concerns deployment scenario with 2xISE
What is more I noticed sth strange after promoting ISE (only in Chrome).
In more details:
- I have 2xISE (gatemaster.domain.local, gatemaster2.domain.local)
After promoting gatemaster2 and accessing the GUI my main view was as the following
It seems to me that even though Im using gatemaster2, internal scripts refers to gatemaster,
I observed it only in chrome probably cause I installed Google Ch. after the promotion and it didn't have it cached.
Is your environment standalone?
regards

CS3 Strange Behaviour or Bug?

Create a shape.
Fill the shape with a color no stroke.
Copy that shape and paste in front. Make this new shape on top a gradient fill no stroke. Set the gradient fill to multiply 100%.
Draw another radom shape on top your existing shapes. Now copy one of your original shapes and paste in front in order to mask this drawn shape.
You should now have 3 layers of geometry. A solid color, a gradient on top and another shape on top that is masked with the a duplicate of the original shape.
Save the file and reopen it. I now only have two sets of geometry. The file combines the mask with the gradient shape below it. Where did my mask go?
thx
paul

If you have any attributes on your object that you're using as your clipping path (fill, stroke, drop shadow, etc...), they will get removed when you make it into a path. You have to apply them after the fact. To help select the mask, you can either grab it with the direct select tool or you can select your whole masked group, mask and all, and use these two buttons up in the control bar (up above your window) to help select it:
Clicking the left circle will select only the masks within your selection, clicking the right one will give you just the contents.

Strange measurement input fields behaviour

Hello,
I have a Czech version of Illustrator CC on Windows 7 Pro.
Some measurement input fields are not working as expected, for example:
When creating a new document:
Decimal marks in document dimensions have disappeared (instead 210.00 mm it shows 210 00 mm). It also disappears after I add it. Another strange thing is when I click in the input box, do nothing and click out the value changes to 5779 55 mm
(Something similar is happening in my InDesign: I can't change an object's dimension expressed with a number with a decimal mark to another number with a decimal mark - only to an integer. I'm not working in Web mode.)
Is this behaviour controlled by some settings I can't find or is this a bug?
Every note leading to a solution much appreciated!

maara,
The decimal point is basically governed by your Regional/Language settings, in the Control Panel I believe.
There may be some issues between . and , for decimal point (and reversely for thousands).
But, especially if the behaviour has not always been there, you may try the list below.
In any case you may try a chat or a support call, here or here,
Creative Cloud support (all Creative Cloud customer service issues, chat open between 5AM and 7PM PST/PDT on workdays)
http://helpx.adobe.com/x-productkb/global/service-ccm.html
Adobe Support (phone),
http://helpx.adobe.com/adobe-connect/adobe-connect-phone-numbers.html
The following is a general list of things you may try when the issue is not in a specific file (you may have tried/done some of them already); 1) and 2) are the easy ones for temporary strangenesses, and 3) and 4) are specifically aimed at possibly corrupt preferences); 5) is a list in itself, and 6) is the last resort.
If possible/applicable, you should save curent artwork first, of course.
1) Close down Illy and open again;
2) Restart the computer (you may do that up to 3 times);
3) Close down Illy and press Ctrl+Alt+Shift/Cmd+Option+Shift during startup (easy but irreversible);
4) Move the folder (follow the link with that name) with Illy closed (more tedious but also more thorough and reversible);
5) Look through and try out the relevant among the Other options (follow the link with that name, Item 7) is a list of usual suspects among other applications that may disturb and confuse Illy, Item 15) applies to CC, CS6, and maybe CS5);
Even more seriously, you may:
6) Uninstall, run the Cleaner Tool (if you have CS3/CS4/CS5/CS6/CC), and reinstall.
http://www.adobe.com/support/contact/cscleanertool.html

Anyone understand this strange LR3 side panel behaviour?

In the last couple of days my laptop has developed a piece of strange behaviour related to the side panels. I'm pretty sure it didn't start when I installed the LR3.5 update, but this may have been a contributing factor.
Until recently I always worked in "solo" mode. A couple of days ago, LR's behaviour changed so that opening or closing any of the side panels would cause the relevant side (left or right) to become totally unresponsive. The rest of the program continued to work fine, and the equivalent side panel in other modules continued to work properly until a panel was opened or closed.
In an attempt to understand what was going on, I turned "solo" mode off. This led to some very interesting behaviour indeed:
If I used the triangle to open or close a side panel, everything worked exactly as expected.
If I clicked on the dark grey bar to open or close a side panel, it became stuck in one of the two states.
When stuck in "open" state, clicking in the grey area simply caused the panel to close and then re-open, very rapidly.
When stuck in "closed" state, something similar happened in that the panel opened and then re-closed, very rapidly.
When in "stuck" state, clicking on the triangle had no effect on the displayed panel but it seemed to toggle the internal state between opened and closed.
Once toggled, clicking on the grey bar opened the panel where it used to close it, and vice versa.
When collapsed, the panel displays as pale grey rather than the more usual dark grey. See below for an illustration.
I have a desktop machine as well which doesn't show the behaviour. I've tried reinstalling LR3.5, to no avail. My guess is that we're dealing with inconsistent panel state information in the "preferences" file, but I don't know for sure.
If anyone knows what is happening, or (more importantly) how to stop it happening, I'd be enormously grateful for some help.
Many thanks in advance,
Ian Wilson
Cambridge UK

Further investigation yielded yet more interesting behaviour - and, it turns out, a pointer to the problem.
I tried renaming the old "preferences" file, causing LR to create a new one. The problem was unchanged, apart from losing my registration information...
I tried creating a new, virgin catalogue. This behaved normally, so I thought I was onto something!
I tried importing a folder of pictures into the new catalogue. The old behaviour returned.
I created a new virgin catalogue, which reverted to good behaviour. I then imported one picture into the catalogue. Behaviour BAD.
I then tried removing the single picture from the catalogue and restarted LR. Behaviour GOOD!
Conclusion: Lightroom works just fine for me providing I never put any pictures into my catalogue!!
This got me thinking that, perhaps, the issue was related to the rendering of images. I tried installing the latest nVidia drivers for my laptop, but this had no effect whatsoever. Given the issue clearly related to how images are rendered, however, I wondered whether the problem could relate to colour profiles. BINGO!! I'd recently upgraded my "i1xtreme" to X-Rite's new "i1publish". The profiles created with the latest software are ".icm" rather than ".icc", although this shouldn't matter (and doesn't matter to PS and Bridge). It seems to matter to LR, however, as reverting to my earlier ".icc" profiles made everything work again just fine.
There is obviously a bug somewhere, but it's not clear whether I should be chasing Adobe or X-Rite. I'd welcome any comments or advice...
Many thanks for the suggestions,
Ian.

Strange 10gR1 Opatch Logging Behaviour

Hi,
just to see if someone else has experencied that:
10gR1 opatch needs ORACLE_HOME/.patch_storage directory
for logging opatch lsinventory command messages, that is LsInventory.pm perl module.
The point is that the module doesn't create that directory and it will show up following error message:
Problems with creating the log file:
Couldn't create file for logging. Error is: A file or directory in the path name does not exist.
So until you use opatch apply command (to apply a patch to the ORACLE_HOME) or until you manual create that directory in your ORACLE_HOME, you
will be unable to log opatch messages.
Thats' all,
Greetings-

Hi!
I have recently encountered the same behaviour when running Forms from the Builder through my local OC4J container.
The error was not same everytime.
I would start debugging and the problem would disappear.
I found that if I did a "Compile All" on the Form everytime i run, I can avoid these problems.and I wasted many hours on this. not in forms but in reports also, everytime i have to compile all the rdf before running it from web.
I was running Forms Builder 9i (9.0.2) on Windows XP and all the forms were build in this version. as I had not used this developer version I was saying this is 9i bug.
I have used developer 10g version and face no such problems but tim is facing this in 10g. so it looks problem is somewhere........
Frank this indeed sounds strange
reg

Flex 3 locale swf loading bug on firefox

Hi all
I REALLY need your help here
if anyone knows this bug or have any idea how to deal with it, it would really mean a lot to me
i've been tring everything
description:
i'm building an application using flex 3, where all texts are being localized via the locale loaded at runtime
i suspect that for some reason the locale swf i'm loading isn't updating elements of the page
as you can see in the image below, texts appear cut off.
this happens to me only in firefox, no matter which page of the app is being displayed
any ideas?
appreciate your time in answering
Yariv Gilad

Hello and thank you for your quick answer
the text is indeed right, and what you say make sense to me
how can i proceed from here though?
how can i fix the "sizing and positioning problem" ?
the UI components are there and are being populated dynamically through the locale
is there a way to force them to update?
is this the right direction at all?
any other ideas?
just to note that if i hard-code the text into the components they are rendered perfectly
so i don't see a reason to change their position or size...
thank you for your time
Yariv Gilad

Tree slow and bug with Add Multiple Items to End

Hello,
I use "Add Multiple Items to End" to fill a tree quickly in Labview 8.5.
In my test example, It's possible to choise an array of 100, 255 or 1000. parents.
Each parent has 10 children
100 is correctly, 255 is slowly an 1000 is big bug !
Y does a solution, or it has Labview major weaknesses with the function tree.
Thanks for you help.
A. Carbonnel
AC
Attachments:
Load Tree.vi ‏37 KB

Here is the response of France rupport labview
Sir,
The behavior you observed beyond 1000 elements vi Tree is a known bug of R & D LabVIEW. Pending that it is corrected in a future version, I invite you to use the ActiveX Treeview. He should be able to meet your expectations in terms of number of elements.
You can see an example of the use of the ActiveX in attachments.
Sincerely,
AC
Attachments:
Tree view.zip ‏48 KB

Module imports incorrect behaviour. BUG?

Hi,
I was reading the forum and didn't find a solution on this problem, although I saw quite some problems with modules.
I have a problem with module imports in the library module. DB XML 2.5.16/ MacOS 10.6.2. Has been tested using DB XML shell and Java API
I have two modules: mod_slave.xq that is imported by mod_main.xq which in turn imported by the main query:
mod_slave.xq:
module "http://tntbase.mathweb.org/ns";
declare function test_slave() as xs:string {
"test_slave"
mod_main.xq :
module namespace tnt = "http://tntbase.mathweb.org/ns";
import module namespace foo='http://tntbase.mathweb.org/ns' at 'file:/path/to/mods/mod_slave.xq';
declare function tnt:test_main() as xs:string {
foo:test_slave()
main query:
import module namespace tnt = 'http://tntbase.mathweb.org/ns' at '/path/to/mods/mod_main.xq';
tnt:test_main()Executing the main query gives me an error:
query failed, Error: The graph of module imports contains a cycle for namespace 'http://tntbase.mathweb.org/ns' [err:XQST0073], <query>:1:1But there is no cycles: main query imports mod_main and mod_main imports mod_slave which imports nothing.
Moreover XQuery spec says: "it is a static error [err:XQST0073] if the graph of module imports contains a cycle (that is, if there exists a sequence of modules M1 ... Mn such that each Mi imports Mi+1 and Mn imports M1), *unless all the modules in the cycle share a common namespace*.". So even is there were a cycle, the error shouldn't be raised because my modules share a common namespace.
The next observation:
if I change the prefix foo to tnt in the mod_main, then I get the error:
query failed, Error: Namespace prefix has already been bound to a namespace [err:XQST0033], file:///path/to/mods/mod_main.xq:2:25But the spec says: "The namespace prefix specified in a module import must not be xml or xmlns [err:XQST0070], and must not be the same as any namespace prefix bound in the same module by another module import, a schema import, a namespace declaration, or *a module declaration with a different target namespace*". But I have the same namespace in the module declaration and a module import, so the error should not be raised.
The workaround for me for now is to put those two mods into different namespaces or unite both modules.
I would appreciate if anybody could confirm or point me to the real error in my modules. Thanks
Vyacheslav

Hi Vyacheslav,
I think you're analysis of both the XQuery spec and Berkeley DB XML is correct. The behaviour you're seeing is almost certainly a bug, but it's unlikely to be of a high enough priority to get attention in the near future. If fixing this is a priority to you, I expect it would be simple enough to point you in the right direction for the code that implements this.
John

Flex 4 tree component - how to populate using an array?

It's unclear how to populate a tree control using an array as the dataprovider like you could in Flex 3.
Here is a Flex 3 example. Even when I replace the mx:Object tags with fx:Object tags things don't work ... there seems to be no mx:children tag ...
http://blog.flexexamples.com/2007/12/03/using-an-array-as-a-data-provider-in-a-flex-tree-c ontrol/
Please advise.

This should do it:
<?xml version="1.0" encoding="utf-8"?>
<s:Application xmlns:fx="http://ns.adobe.com/mxml/2009"
        xmlns:s="library://ns.adobe.com/flex/spark"
        xmlns:mx="library://ns.adobe.com/flex/halo">
    <fx:Script>
        <![CDATA[
            private function tree_labelFunc(item:Object):String {
                var suffix:String = "";
                if (tree.dataDescriptor.hasChildren(item)) {
                    suffix = " (" + item.children.length + ")";
                return item.name.toUpperCase() + suffix;
        ]]>
    </fx:Script>
    <fx:Declarations>
        <fx:Array id="arr">
            <fx:Object name="One">
                <fx:children>
                    <fx:Array>
                        <fx:Object name="Two" />
                        <fx:Object name="Three" />
                        <fx:Object name="Four">
                            <fx:children>
                                <fx:Array>
                                    <fx:Object name="Five" />
                                    <fx:Object name="Six" />
                                </fx:Array>
                            </fx:children>
                        </fx:Object>
                        <fx:Object name="Seven">
                            <fx:children>
                                <fx:Array>
                                    <fx:Object name="Eight" />
                                </fx:Array>
                            </fx:children>
                        </fx:Object>
                        <fx:Object name="Nine" />
                    </fx:Array>
                </fx:children>
            </fx:Object>
        </fx:Array>
    </fx:Declarations>
    <mx:Tree id="tree"
            dataProvider="{arr}"
            labelFunction="tree_labelFunc"
            width="200"
            horizontalCenter="0"
            verticalCenter="0" />
</s:Application>
Peter
Update: I also posted the updated Spark example at http://blog.flexexamples.com/2007/12/03/using-an-array-as-a-data-provider-in-a-flex-tree-c ontrol/.

Flex Using Tree Control in item renderer(Url Navigate)?

HI All.,
Iam Using flex 3 using tree control in item renderer to click haschildren label to be navigate url is possible ?.
Any one help me.....
Thanks in Advance......

You can do this by writing the item's tag to the ActiveItemTag property, the column number you're intersted in to the ActiveColNum property and reading the Cell String Property.
Mike....
PS: If anybody at NI is listening, that interface really, really, REALLY needs to get rewritten...
Certified Professional Instructor
Certified LabVIEW Architect
LabVIEW Champion
"... after all, He's not a tame lion..."
Be thinking ahead and mark your dance card for NI Week 2015 now: TS 6139 - Object Oriented First Steps
Attachments:
tree_properties.vi ‏9 KB

Flex 4 Modules Creation Cycle bug ?

http://filip.nedyalkov.net/creationcycle/ - view source is enabled, download and see the traces
Ok so here is what happens in this application:
Some elements are created in the main application and also a module is loaded. As soon as it is ready the button in the main app goes visible. When you click it the module is added to the main application.
There is where it gets weird. If the module loader is not added to the main application before the module is ready, everything I've set into the module gets created kinda randomly. For example if I have a button in the module, it should fire button created and then module created, but here it doesn't sometimes. Check out the application I've provided to see what I mean.
I think that it shouldn't matter if the module loader is added or not, and the module should fire the creation complete events in the normal order, from bottom to top (top is the module itself).
So I hope someone familiar with modules life cycle behaviour could say if what I see is a bug or not and how it can be fixed. Of course I could add the module loader before the module is loaded but make the moduleloader visible=false, and when I need it visible=true, but that's not the point here.
Thanks

Go to this thread: http://forums.adobe.com/thread/764956 - this is with better example.

Flex - 3 Tree Control

Hi,
I have to display Tree items in clickable link item as simillar to LinkButton.
Can anyone help me plz.
Thanks!

Thanks buddy for the answer.
Unfortunately the answer came after quite long time of posting the message. Anyway I was able to open a tree on demand using HttpService and due to my new requirement I changed it to RemoteObject.
I my latest change I am able to populate tree nodes on demand and also the same solution if getting update from server via JMS using Consumer object.
I kind a like this solution because it took me good amount of effort to find the right solution.
If any one is intersted the he/she can reply to the post and I can provide code here or may at some location so that it can be easily downloaded.
The solution is Flex-Grails combination.
Thanks everybody.

Flex profiler weird behaviour

I'm experiencing a weird problem with the profiler.
Thing is, it won't show objects instantiated after a certain event in the application life, specifically after loading an xml file.
If i create some 1000 bogus objects in the method that loads the xml, i can see them showing up in the live objects panel, while moving their creation in, for example, the method fired by Event.COMPLETE of the xml URLLoader doesn't change at all the live objects report.
I couldn't manage to find a solution googling and i'm wondering, is it a known problem?
p.s.
By the way, profiling still works with the listed objects so it's not some connection problem. Also, i tried importing the same project (which works perfectly, just needs a bit of resource optimizing) on another flex builder 3 and on a flash builder beta 2, obtaining the same results.

Objects created in an event handler will be collected if not stored somewhere permanent. Make sure you've got your code correctly then file a bug if there is still an issue
Alex Harui
Flex SDK Developer
Adobe Systems Inc.
Blog: http://blogs.adobe.com/aharui

Strange Custom Template Sections Behaviour

I've created some custom template sections so I can insert my own various page types for a document I'm producing. The strange thing is Pages will insert an extra blank page (similar to the master but not exactly the same it's missing a text box) between sections when I add a section—but only some of the time!
It will not do it when I initially add the section.
It will only add in a 2nd page between when I add additional sections, be they the same type of section or another type of section.
I cannot delete this 2nd page without deleting the 1st page of the section as well.
When I try to capture just the 1st page to make another version of the custom template it still exhibits the above behaviours, inserting this "buffer" 2nd page in between sections.
All the other custom sections I have created do not exhibit this behaviour.
I've turned on invisibles and can't see anything that makes this section special or different.
Any guesses as to why it inserts this extra page in between sections to make a 2 page section instead of staying as a 1 page section?
Message was edited by: Nathan Muirhead

Nathan
You have some text being forced over to the next page, either because you hammered away at the spacebar/return key/tabs, or have some object wrap pushing returns over, or both:
+Menu > View > Show Invisibles/ Show Layout+
Once you have cleaned up your pages recapture them as sections and resave your template over the old one.
Peter

Strange Flex 3 Tree behaviour. A bug?

Similar Messages

Maybe you are looking for