Strange Behavior In Defining A Template Interface -- Bug/RFE

Hello.
This is a bit of a bug. I wanted to define a "template" interface. If you follow this sequence of steps, you may see this odd behavior:
In a class diagram, insert a new Interface.
Add a template parameter to that interface. Now, the display of the object has not changed.
Right-click the interface, and under "Show", select "As Icon".
Right-click again and under "Show" select "As Non-Iconic Class".
Now the interface is displayed as a "Template Class" on the diagram, with its parameters visible. This comes about since the Modeling Palette only contains a tool to draw a "Template Class", which creates a class only and not an interface.
Is this behavior somehow intentional or expected?
Is there any bug reporting tool for such reports? Or is posting on this forum sufficient to constitute feedback on a bug?
Thanks in advance.
- Steven Coco.

Hi.
I am not aware of another notation: the display given by JSE (after making it display the parameters) is all I would expect -- the UML superstructure just says that a templatable element has the dashed-outline box listing its parameters. Now, the display of the parameters can be suppressed, but I would have expected in that case a more direct "Hide or Show Template Parameters" option for the element. And in fact, I'd say that a display option like this is really what this boils down to: there's no option to hint JSE to display the parameters; and you'd have to go a round-about fashon to make them appear.
And my display does still show <<interface>> after geting the parameters to appear.
Thanks!
- Steev.

Similar Messages

Local adjustement strange behavior. Bug ?

Hallo,
     Whilte retouching picture from a wedding I got some strange behavior. I wanted to hightlight the couple using local adjustement / Brightness.
But the result get "noisy". Look at the screenshot.
More about the picture:
Very low light, EOS 5D Mark II, 3200 ISO, RAW/CR2
     Any opinion ?
     Regards

If it doesnt help, can you provide a different photo that has the same behavior happening. It's hard to tell what's going on with the one that you provided

Adding custom navigation rules results in strange behavior

Hello,
We'd like to add navigation rules to our application. To avoid post-JHeadstart-generation-steps we created an extra faces-config-custom.xml file which contains the navigation rules. When adding this file to the web.xml and run the aplication we encounter strange behavior
- Errors are shown in duplicate
- 'Transaction completed' messages are not shown
Try adding the underneath faces-config-custom.xml to a standard HR demo project and you will get the same behavior.
(1) What is the reason of this strange behavior?
(2) How can we add custom navigation rules, without having to do post creation steps?
Regards Leon
[faces-config-custom.xml]
<?xml version="1.0" encoding="windows-1252" ?>
<!DOCTYPE faces-config PUBLIC
"-//Sun Microsystems, Inc.//DTD JavaServer Faces Config 1.1//EN"
"http://java.sun.com/dtd/web-facesconfig_1_1.dtd">
<faces-config xmlns="http://java.sun.com/JSF/Configuration">
<navigation-rule>
<from-view-id>*</from-view-id>
<navigation-case>
<from-outcome>BezwaarVerzoeken</from-outcome>
<to-view-id>/pages/inboeken/BezwaarVerzoeken.jspx</to-view-id>
</navigation-case>
<navigation-case>
<from-outcome>LosseOpdrachten</from-outcome>
<to-view-id>/pages/inboeken/LosseOpdrachten.jspx</to-view-id>
</navigation-case>
</navigation-rule>
</faces-config>
[Add faces-config-custom.xml to web.xml]
<context-param>
<param-name>javax.faces.CONFIG_FILES</param-name>
<param-value>/WEB-INF/faces-config-custom.xml,...

Leon,
When you perform a drag and drop operation, JDeveloper adds the following lines to the faces-config.xml:
<lifecycle>
<phase-listener>
Oracle.adf.controller.faces.lifecycle.ADFPhaselistener
</phase-listener>
</lifecycle>
However, JHeadstart uses its own subclass of ADFPhaselistener, and defines the lifecycle element in JhsCommon-beans.xml. Due to a bug in ADF, ADF does not look for the lifecycle element in other files than faces-config, and adds its own element in faces-config.xml If you remove these lines, everything works fine again.
To prevent this from happening again, you can move the following entry from JhsCommon-beans.xml to faces-config.xml:
<lifecycle> <phase-listener>oracle.jheadstart.controller.jsf.lifecycle.JhsADFPhaseListener</phase-listener>
</lifecycle>
And then make a custom template for JhsCommon-beans.vm where you remove this entry.
Steven Davelaar,
JHeadstart team.

Config Files Backup to Avoid Interface Bug

I have recently upgraded to Adobe Creative Suite Master
Collection. Although I am impressed with the collection for the
most part, I can't believe that Adobe has not addressed a glaring
bug in the user interface that will contsantly corrupt the sites
list and configuration files.
The bug stems from having a large website portfolio. As a web
designer, as well as a webmaster, I have just finished configuring
most of the sites and directories I need immediate access to.
Dreamweaver CS3 could not import the previous site list and
configuration because of the interface bug which is constantly
corrupting that file. What happens, is that I select the website to
work on (say, dnsite.com) from the drop down list, and am satisied
that I am there. Unfortunately, my attention is diverted or I
simply forget to click somewhere else in the window. If I would
click somewhere else in the window, things would be fine. But that
fact that I need to is a true interface bug, because when I don't
click somewhere else (for whatever reason), I look at my list of
files and immediately start to scroll down through the file list to
get to the file I want to work on, BY USING THE MOUSE WHEEL, like I
always do. However, because I have not clicked outside of that
original site list drop down box, that site list selector scrolls.
The result is that Dw loads and scans each site over and over until
it comes to the end of the list, OR CORRUPTS THE SITE LIST AND
CONFIG FILE. This happens all the time with me, and it is extremely
infuriating. In order to combat this interface design bug, and Dw's
inability to protect its own configuration files by handling them
correctly, or even producing a backup file that it can revert to in
the event that it has corrupted that last site list and config
file, this interface bug has forced me to manually backup the site
list and config file. But apparently, this file has moved since
Dreamweaver 2004 MX?
By the way, it would be simple to fix this bug with a simple
pop-up dialogue window that would disappear once the correct site
was selected (duh... I thought that Adobe's noted interface design
excellence would have fixed this strange idiotic flaw immediately,
but I see they haven't addresses anything us web designers want or
need). I know Macromedia was working on the interface, but they
never did get it right and I was expecting BIG improvements from
Adobe RIGHT AWAY (because they are so good at it and Macromedia was
so poor at UI design).
Because of this this bug, and because I just configured over
60 websites that I need to maintain immediate access to, I need to
backup my website list and configuration file, but I see that it
has changed or moved. It used to be in the application data folder
in the folder up one directory from the My Documents for my
username on the computer, but I can't find it anymore. I checked
All Users as well. Where has this file gone, and what is it named,
now?
By the way, as a default, you would be saving me and other
web designers a great deal of time (re)configuring my list of sites
if you would have included .psd .ai and .eps as file extensions to
cloak. Macromedia didn't do this because they didn't want to
promote you, but you own this software now, update it.
So little thought has gone into this update that I am truly
dissapointed. Dw used to improve with each update. You can't tell
me anyone is happy with it and keep a straight face.
-Doug
Symbiotic Design

> to the file I want to work on, BY USING THE MOUSE WHEEL,
like I always do.
> However, because I have not clicked outside of that
original site list drop
> down box, that site list selector scrolls.
Not what you'll want to hear probably- but the site list
scrolls again
because the site list drop down is what has focus. It has
focus until you
click something else. That's the GUI behavior i would expect
instead of a
warning dialog box getting in the way.
click in the files panel if you want to scroll the list of
files for the
chosen site.
location of Configuration folder:
Dreamweaver CS3 on Windows Vista:
C:\Users\[username]\AppData\Roaming\Adobe\Dreamweaver
9\Configuration
* Dreamweaver 8 on Windows Vista:
C:\Users\[username]\AppData\Roaming\Macromedia\Dreamweaver
8\Configuration
* Dreamweaver CS3 on Windows XP:
C:\Documents and Settings\[username]\Application
Data\Adobe\Dreamweaver
9\Configuration
* Dreamweaver 8 on Windows XP & 2000:
C:\Documents and Settings\[username]\Application
Data\Macromedia\Dreamweaver
8\Configuration
* Dreamweaver CS3 on Mac OS X:
Mac HD:Users:[username]:Library:Application
Support:Adobe:Dreamweaver
9:Configuration
* Dreamweaver 8 on Mac OS X:
Mac HD:Users:[username]:Library:Application
Support:Macromedia:Dreamweaver
8:Configuration
The site definition information is stored in a Registry Key
on windows.
mm-exporter.joexx.de/ is a free third-party utility that can
handle this and
other tasks.
The site defs on macintosh- probaby easiest to use the Export
button in the
manage sites dialog box. Can't lay hands on the file location
offhand.
Alan
Adobe Community Expert, dreamweaver
http://www.adobe.com/communities/experts/

Strange behavior - icon thumbnail instead of image - CC@OSX10.10

Hi,
my CC (@OS X 10.10.3) showing a strange behavior using vanishing point filter.
After defining grid and try to paste the image OS X paste the icon thumbnail instead of the image,
Many thanks in advance!!
br
Joachim

It is probably related to the fact that you are setting the tables row height in the renderer this will fire a render message when the height changes. One height is applied to all rows of a table. This means that is you have images of different heights you are continually changing the tables row height and therefore continually rendereing.
My approach would be to have all the images fit into a set size ( say 64*64 ( with scaling if neccesary ) you could then have a popup show the full size image if required.

Strange behavior of textareas in BPEL Worklist (10.1.2)

I have a very strange behavior with linebreaks in all my <textarea>s in the BPEL-Worklist-jsps: (they work fine as <INPUT>s
Example:
Default-text:
line1
I update the textarea to:
line1
line2
-> hit update
line1
*** (second line is completly missing)
-> hitting update again (no change in text)
line2line1
*** (all on one line with the new line in front)
As Value I use:
<%=PayloadFormGenerator.selectNodeValue(payload, "/ns0:task/ns0:payload/ns1:abstract", form.getNamespaceMap(),"string", context.getLocale())%>
and as mentioned: It works fine for <INPUT>.
What do I do wrong? Is this a bug?
Daniel
Message was edited by:
user552073

Had the same problem, could not get it to be consistent. I would say this is a bug, but decided to edit payloads via the database in the end...

Strange behavior with Zoom and Image control

HELP - I have a strange behavior (bug?) with using Zoom
effect on an Image that has been placed on a Canvas. I am using
dynamically instantiated images which are placed on a canvas inside
a panel. I then assign a Zoom IN and Zoom Out behavior to the
image, triggered by ROLL_OVER and ROLL_OUT effect triggers. THE BUG
is that the image jumps around on the Zoom OUT and lands on a
random place on the canvas instead of coming back to the original
spot. This is especially true if the mouse goes in and out of the
image very quickly. HELP -- what am I doing wrong? Computer = Mac
OS X 10.4.9 Flex 2.0.1
Here's a simple demo of the bug -- be sure to move the mouse
in and out rapidly:
<?xml version="1.0" encoding="utf-8"?>
<mx:Application xmlns:mx="
http://www.adobe.com/2006/mxml"
layout="absolute" creationComplete="setUp();">
<mx:Script><![CDATA[
import mx.events.EffectEvent;
import mx.effects.Fade;
import mx.effects.Zoom;
import mx.rpc.events.ResultEvent;
import flash.display.Sprite;
import mx.core.UIComponent;
import mx.controls.Image;
private var zoomIn:Zoom;
private var zoomOut:Zoom;
private function setUp():void {
var image:Image = new Image();
image.id = "album_1_1";
image.x = 200;
image.y = 200;
image.width = 64;
image.height = 64;
image.source = "
http://s3.amazonaws.com/davidmccallie/album-128.jpg";
image.addEventListener(MouseEvent.ROLL_OVER, doZoom);
image.addEventListener(MouseEvent.ROLL_OUT, doZoom);
myCanvas.addChild(image);
zoomIn = new Zoom();
zoomIn.zoomHeightTo = 2.0;
zoomIn.zoomWidthTo = 2.0;
zoomIn.captureRollEvents = true;
zoomIn.suspendBackgroundProcessing = true;
zoomOut = new Zoom();
zoomOut.zoomHeightTo = 1.0;
zoomOut.zoomWidthTo = 1.0;
zoomOut.captureRollEvents = true;
zoomOut.suspendBackgroundProcessing = true;
private function doZoom(event:MouseEvent):void {
var image:Image = Image(event.currentTarget);
if (event.type == MouseEvent.ROLL_OVER) {
zoomIn.target = event.currentTarget;
zoomIn.play();
} else if (event.type == MouseEvent.ROLL_OUT) {
zoomOut.target = event.currentTarget;
zoomOut.play();
]]>
</mx:Script>
<mx:Panel width="100%" height="100%"
layout="absolute">
<mx:Canvas id="myCanvas" width="100%" height="100%">
</mx:Canvas>
</mx:Panel>
</mx:Application>

There must be bugs in the Zoom effect code -- I changed the
Zoom to Resize in the above code, and it works perfectly. Of
course, Resize is not as nice as Zoom because you can't set the
resize to be around the center of the image, but at least it works.
Does anyone know about bugs in the Zoom effect?

Strange behavior of "Partial message encryption " in OWSM

Strange behavior of “Partial message encryption “ in OWSM
If message format is like this
<cban:transferFund xmlns:cban="http://cbank.com">
<cban:arg0>string</cban:arg0>
<cban:arg1>string</cban:arg1>
</cban:transferFund>
Then body element configuration in message encryption setting
Name space http://cbank.com
Element arg0
is working fine
But my requirement is for below message
<cban:transferFund xmlns:cban="http://cbank.com">     <
<arg0>string</arg0>
<arg1>string</arg1 >
</cban:transferFund>
For this
Name space : http://cbank.com
Element : I tried all possible combination but none is working
I tried these
:arg0
//:arg0
//cban:arg0
//cban//arg0
cban:arg0
With Regards
Siddharth

Had the same problem, could not get it to be consistent. I would say this is a bug, but decided to edit payloads via the database in the end...

Strange behavior in entity bean : get Timestamp

Hello:
I'm working with SUNONE 7 AppServer , over SunOS 5.9
I've a strange behavior with entity's get methods which return Timestamp value.
For example, I've got
Timestamp date;
If I do
entity.setF(date) , ( date is a Timestamp with value "12/12/2005 12:30:00" )
all works right, and in database is wrote right ( "12/12/2005 12:30:00" )
But , if I do
date = entity.getF()
the, date variable has the value "12/12/2005 00:00:00"
So, in get method is lost the time value of a Timestamp data
Could be a code bug in my source , but if I use Jboss AS over Windows XP , all work right ( set and get methods ). The database is the same one ( Oracle 9i )

Well, I found the solution.
The problem was the ojdbc14.jar driver, which made wrong schema files.
Exactly, with the bad ojdbc14.jar, generated this entry
<_type>91</_type>
when the right one for date types ( Timestamp ) is
<_type>93</_type>
I dont know why the new ojdbc14.jar works fine, but I paste its size
good ojdbc14.jar : 1181679 bytes

Strange behavior in deleting images - works on second but not first try

In Library mode, Grid view, right after importing, after I select several images with an "x" to be deleted, when I bring up the "Delete Rejected Photos" dialog, I find this strange behavior. I get the message that there are not photos marked for deletion. I then invoke the dialog again, and my photos that were marked show up, with the choices of deleting from disk or removing from the catalog. What am I likely doing wrong? I'm on WinXP Pro with LR 2.2. Thanks!
Dilip

Dilip,
I too have seen this behaviour in all versions of LR. Even when deleting a single image in the develop module. If LR is writting the preview you can't delete the file until the preview is rendered. I think it's more to do with the OS not allowing the deletion of a file being accessed than a LR bug as such.
Gordon

Strange behavior in Lyx

Hi,
I've stumbled to an strange behavior in Lyx. When I write a document (I have this problem in Finnish at least) and convert it to pdf, in every word that has "fi" in it has the "fi" syllable been replaced with white square. It also happens with "--". It doesn't happen for "FI" or "Fi" or any other combination that I've noticed.
I'm a bit puzzled here, could somebody try it and tell me if it is just me or is there a problem with the pdf converter?
greets,
.murkus

Snowman wrote:
This is a ligature problem (latex/dvips bug).
Solutions:
- If you use dvips, use the -G0 option: dvips -Ppdf -G0 -o output_file.ps input_file.dvi
- Or, in your text, replace "fi" by"f{}i". This will prevent latex to use ligatures.
Many thanks! :-D
.murkus

Strange behavior with Firefox

Hi Team,
I am using HTMLDB V2.0 and Firefox V1.5.0.1. I used wizard to create a report/form application. However, when I press a button to update the form all the buttons line up in a vertical row and I then have to press the same button again.
My application is 30412 at htmldb.oracle.com. The problem is occurring at page 4. Go to page 3 (2nd tab) and select any record. You will be directed to page 4. Then when you press any button (cancel, create, delete, apply changes) on page 4, page 4 is refreshed and the buttons strangely line up in a vertical column. I then have to press the same button again before the update occurs.
I have the same report/form setup on pages 1 and 2 and it is working fine on those pages.
This strange behavior is not occurring in IE. I also notice that the font size is bigger in IE .
I would appreciate any help you might offer.
I really enjoy working with HTMLDB.
Thanks, Andy

Sorry about that. I removed all the authorization schemes from the application. I also changed the authentication scheme to HTMLDB. Let me know if I need to do something more for you to get access.
Thanks for looking at it.
I think it may have something to do with the "button, alternative 3" template from theme 3 that I am using. When I switch to the "button" template from theme 3 then the strange behavior goes away in Firefox.
Andy

802.1x "MachineorUser" Auth Mode strange behavior in 2950 & 3750 Switches

Good Day Support Team around the world,
Having started recently tests with 802.1x in a lab environment, I noticed a strange behavior related to authentication. First let me provide you with the network components I used.
supplicant:                    domain-joined laptop with Windows XP SP3 802.1x embedded client
authenticator1:              Cisco 2950-24
authenticator2:              Cisco 3750-24
authentication server:     MS NPS Windows Server 2008
1.     In the first scenario with 3750 switch when I connect the laptop to relevant port the machine authentication is successful. Then I try to login with a domain account and again the authentication is completed without any problem. Then I log off and user authentication is revoked and the machine authentication is used again without any issue. When I try to login again as local user the authentication fails as expected but the port remains disabled (port blinking amber) regardless the fact that port is configured for Auth-Fail Vlan. When I log off then the machine authentication is used again and the access is granted.
2.     In the second scenario with 2950 switch as authenticator, I follow the same steps as before and when I try to login as local user the authentication is failed and the port is assigned the Auth-Fail Vlan (as expected based on configuration). However when I log off it seems that the 2950 switch still use the Auth-Fail Vlan for that port and never authenticates again for machine authentication.
Could you please let me someone know if this is normal ( I suppose no). Please find attached the relevant debug output from the second scenario.
Thank you!!!

Hi,
basically what happens is that the maximum EAP packet size for communication between client and RADIUS server is negotiated. Therefore, in your case the switch notifies NPS that the client is capable of handling packets up to 9000 bytes in size.
EAP messages, especially those containing the server certificate, are usually bigger than 1500 bytes and arrive at the switch in multiple fragments:
Mar 6 15:50:11.881: RADIUS(0000002C): Received from id 1645/41
Mar 6 15:50:11.881: RADIUS/DECODE: EAP-Message fragments, 253+253+253+253+253+253+253+253+20, total 2044 bytes
Having learned that 2044 bytes is acceptable for the client, the switch forwards the full message in one chunk, but since your client is likely to have set the interface MTU to 1500, the packet is oversized and never reaches its destination.
And yes, I think changing the System Jumbo MTU to 1500 bytes would lead to the same result. If my memory serves me right, a new setting takes effect only after a reboot, so I'd suggest giving it a go in your lab first.
Best regards,
Josef

Strange behavior of Grapher 2.0

Hello everyone,
I have a problem with vector fields in Grapher.
My Grapher 2.0 (running on Mac OS X 10.5.2 ) shows the following strange behavior:
When I create a new 2D-vector field in any document, it is shown correctly. All vector arrows point in the direction I expect.
When I now save the document and open it again, first the vector field is shown correctly, but one or two seconds later the field collapses and all the vectors are aligned parallel to the x- axis. After reentering the formula describing the field, I get the correct result again.
In the Grapher examples are two items including vector fields. One of them has the same behavior like my own documents, the other has not, but it‘s working well.
I have already deleted the preferences (com.apple.grapher.plist) , but without any success. The program behavior is unchanged.
Does anyone know the cause of this problem, and what I can do to avoid it?
regards woopecker

This is bug nr. 4/22 in my personal Grapher's bugs list. Exists in 2D & 3D.
To restaure a saved .gcx vector field, make a minor change in its equation (e.g. cancel one letter), Return touch, repair the equation, Return touch, your vector field becomes OK !

Strange behavior of StuffIt-Expander

Hi there,
having heard of the "new viri" for Mac OSX I turned over to MacFixIt and similar sites to check what's to do.
The first advice I read was to "un-check" the "open 'safe' files automatically in Safari" option in the Safari preferences. So I did and all files I was downloading ended unchanged (= un-expanded) in my download-folder. So far, so good.
Now I wanted to expand one <filename>.dmg.gz with double-clicking it. StuffItExpander started, the the message: "StuffItExpander unexpectedly quits" poped up.
I couldn't expand my d/l files, none! Even with dragging/dropping the file onto the StuffIt icon I had no results.
Made some brainstorming and resulted in the one point: The only thing I changed in the last time, was indeed the "un-checking" of the open-option in the Safari preferences.
So I went to the Safari preferences again and "checked" this option, turning it on.
And you are right: Now I could expand my d/l files like I always could.
So, how comes?? Is it only me, where StuffIt shows such a strange behavior? What can I do? Perhaps a re-installation of StuffIt?
Any help is appreciated.
AndreasB
iMac G5 1.8 17" Mac OS X (10.4.4) Ext. FireWire Drive, Brother DCP-110C Printer

Andreas, see if this article helps. A lot of folk were plagued with this wee bug, including me. The linked article solved my problem

Strange Behavior In Defining A Template Interface -- Bug/RFE

Similar Messages

Maybe you are looking for