Strange behavior of CSS with IE9 and JDeveloper 11.1.1.6.0?

Similar Questions

• I get a strange behavior of the bar tabs and the address bar in Firefox 29,0 for Mac.

  I just installed Firefox 29,0 for Mac.
  I have a strange behavior of the bar tabs and the address bar with this new version.
  Instead of the address bar, I get two rows of symbols. And it is impossible to write anything in the address bar.
  (I add a screenshot, but I can't find a way to do it).

  Firefox has an RSS icon in the palette of the toolbar that you can drag a toolbar (but not on the location bar container).

  • Firefox menu button > Options/Preferences > toolbars
  • View > toolbars > customize

• Issues coming out of "sleep mode" and strange behavior when restarting with GPU

  I use a desktop computer HP P7-1240 completely stock except for a GPU EVGA GeForce GTX 750 TI SC. The GPU works well when the computer is running (which is strange because the forums have several other discussions mentioning the 700 series must be incompatible with my BIOS).

  However, there are two main strange problems that have arisen since the installation of the GPU. One is that, every time that the computer enters mode "sleep", he refuses to wake back up correctly. It seems to me that the computer active but sends all the signals back panel (at the very least, no video because my monitors remain blank). On rare occasions, he wakes back correctly, but at least 90% of the time, it must be restarted. Upon reboot, the same problem sometimes (not), and the computer must be rebooted a second time until it will start again.

  This leads to the second problem. During startup, the computer goes to a black screen, listing the specifications of basic computer skills (see attached photo. In this screen, the computer does not meet any entry.  He proceeds to beep 3 times, all the same length and height spaced about 30 seconds of gap. After the third beep, the screen goes black and then the computer starts normally with no other strange behavior.

  Without the GPU, none of these problems occur. At the very least, not regularly.

  I saw another thread mentioned that, while the HP support for my office page lists not her, there is an update of the BIOS for my motherboard which has fixed issues related to the GPU more modern. By nature it is an update of the BIOS that does not (seem to) officially support Windows 7, I am reluctant to go forward with this potential solution.

  Although I intend to build myself a new computer from scratch at a certain point, I was hoping to get at least another year out of this machine, if not 2.  While this question does not seem to be immediately threatening the stability of the computer, I can't help but think that it is a question should I go quickly to avoid future problems (not to mention it would save me a good amount of electricity to be able to enter the mode 'sleep' on a regular basis).

  Any suggestions would be greatly appreciated!

  

  Problem has been solved!

  For those of you who have trouble sleeping with the p7-1240 or computers with a card mother identical (MSI Jasmine Ms 7778 ver 1.0), I suggest that you follow this thread:

  http://h30434.www3.HP.com/T5/desktop-hardware/unable-to-boot-with-new-graphics-card/TD-p/3930544/highlight/true

  It is to use an update for your BIOS that is intended only for PC under Windows 8, but that seems to work perfectly well with Windows 7 PC. It solved my problems of sleep and makes the computer to behave at least semi normally while restarting.

  However, as indicated in the wire, an update of the BIOS should be a last resort because it can turn your machine into a very heavy and expensive paperweight if something does not work. Do not forget that you have the worm 1.0 card mother MSI Ms 7778 Jasmine and remember that you update the BIOS version 8.18.

  As I am looking for 2aeo on HP.com currently (according to the thread leading up to), I'm not find the same links I found last week when I went through the update. However, you can find elsewhere.

• mail. Live with IE9 and magnifying glass

  Hello

  I installed IE9 and see,.

  If I read my emails with mail.live.com;
  the text disappears when you turn on the Magnifier
  using the Microsoft Mouse 3000.

  This nature is adjustable?

  Thank you
  Erhy

  Found the solution by seeking to magnify in sites of responses:

  http://answers.Microsoft.com/en-us/IE/Forum/IE9-windows_vista/Magnifier-is-odd-in-IE9-the-wrong-tab-is-displayed/579de9c1-183b-42b3-9B6C-8cb6326776d5

  ____________________________________________

  Front of:

  Thanks for your thoughts on the subject!

  In addition, I've seen,
  the problem stops in each content displayed with IE9.

  Sometimes the spotlight seems to work;
  also, in these rare cases,.
  the magnifying glass shows only one
  snapshot of the content displayed.
  For example, during scrolling, the magnified area does not work.

  It was no problem before using IE8.

  Maybe somewhere there is a button to change this behavior?

  Erhy

• Strange behavior. 2590 relay switch and Get position vi.

  I'm reading a PXI2590 1 x 4 MUX switch relays that I fires only when necessary. I know the names of the relays are MUX01, MUX23 and MUX0123. If I switch, niConnect channels INIT, wait Debounce and can get the Position of the relay, I get a correct reading (closed or open). I then called niSwitch close .vi and the switch remains in this position.

  However if I just called INIT, niSwitch getting the Position of relay and close vi seems the selector to a different value (default?). I confirmed this behavior using a HF generator and Analyzer of spec.

  Is what I do with these evil vi? Can I use DAQmx driver for this switch instead?

  LabVIEW 8.5.1

  OR-Switch 3.70.49152

  NOR-DAQmx 8.7.1f2

  NOR-PXI-2590

  Hello GovBob,

  Correct me if I'm wrong.  First you call OR-SWITCH boot (not initialize with topology) or initialize options with the 'reset device' Boolean is unwired? (which means reset = true)

  Then, in a case: call OR-SWITCH connect channels, wait OR-SWITCH Debounce, OR-SWITCH get relay post, then NOR-SWITCH close (close a session, not close relay).  Then physically measure opens or shorts.

  In the other case: appeal OR-SWITCH get relay post, then NOR-SWITCH close.  Then physically measure opens or shorts.

  The results are the second case has openings (ch0 is connected to the com).  This is expected behavior since the indicator 'reset the device' in one of the screws initialize is TRUE if left thread continues.  However, if you physically measure different values compared to what OR-SWITCH get post is back, so please submit your screws and a list of the steps and the values measured for each stage.  I hope this helps!

  -Jordan

• No site Ichain will now load with IE9 and Norton360 since April

  I have Norton 360 and IE9 and since late April, I can't connect to my Citrix using IE9 works, I can also not connect to my web mail works them both use Novell Ichain, sites just don't load up, so I don't not even for login pages, all other internet job sites.

  I tried other websites who use the Ichian and also may not load.

  I can connect using Google chrome, but not with IE9

  Some kind of update on IE or norton stopped what I think?

  Thank you for answer you, I had already tried the above, but last night I found the fix as follows.

  In IE8 and IE9 under internet options and advanced, use TLS 1.1 and 1.2 was checked off by unchecking 1.1 and 1.2 and leaving just TLS 1.0 (which, with IE7, I had was) Citrix is now charging and all my other sites work using Ichain

  I don't know if it was an update but everything was working fine until this end April and it took me a long time to find out why.

  Again thanks for taking the time to answer and you can now close this.

• Inconsistent behavior of focusLost with JTable and JTree

  Swing has known problems for a long time with the cells and the loss of the thematic demonstrations in the JTables. Generally, users expect loss of focus to commit their changes rather than cancel the change. Then edit controls (for example, JTextFields) located in CellEditors should generally be listeners of update and manage focusLost() by calling cellEditor.stopCellEditing (). It is very clear.
  
  JTables to have an additional, special problem associated with a generation (or not) of focus events when moving from one cell to another cell in the same table. To do this, the code that creates the JTable needs to call table.putClientProperty ("terminateEditOnFocusLost", Boolean.TRUE). Setting this property to 'magic' on the table allows expected behaviors Assembly when moving from cell to cell.
  
  These issues are well described here: http://tips4java.wordpress.com/2008/12/12/table-stop-editing/
  
  I've set up the property of the client and the listener to focus, and they perform as expected in most cases. But I seems to have discovered a third case that is not managed by one of these two known workarounds.
  
  We have a design of the user interface master/detail where the master is a tree control and retail is the JTable of editable cells. The (master) JTree and JTable (detail) are each cashed in the JPanels which are in turn greeted in an enclosing JPanel.
  
  If I am editing within a cell and click completely outside this hierarchy - for example, in the menus of the main application bar - the focusLost() is managed and the change is committed, as you wish. And if I click anywhere in the JTable, also commits the change.
  
  But if I'm editing a cell and choose the JTree specifically, the change is aborted.
  
  I added logging for focusLost() Manager. I found that in the majority of cases, for example when clicking on menus or other completely different control bar, table.isEditing () returns TRUE for the JTable both focusLost() Manager is running. Because the change is running, the Manager is capable of broadcasting of stopCellEditing() and validate the user change.
  
  But when I click on the outside for the JTree specifically, then in focusLost() Manager I find isEditing() is WRONG - that is, the change has already been cancelled and there is no way to recover already cancelled user input. The text field focusLost Manager pointed out that 'the opposite element' (wins the focus) is a JTree control as expected.
  
  So it seems that something either the JTree or the locking hierarchy Panel made a subtle change to the behavior of the focusLost(). And therefore bypasses of modification of standard tables do not fully work in our application. And unfortunately, clicking on cells change directly in the tree turns out to be a common case in our design of the user interface. Failing to validate the edition in this case with bad consequences that should be fixed.
  
  Any suggestion would be appreciated. Note: the question is similar to related to this issue:
  
  JTree does not call focusLost on headphone
  
  but the tail wire. I concede that I'm not provide code either and he can't promise right now.
  - - -
  Jeff

  Rainmaker says:
  Right now seems to be a lot to a precise definition of the terms so I ask this question with regard to the docs.

  good start :-)

  >

  You wrote:

  invokesStopCellEditing and terminateEditOnFocusLost set up different behaviors, without overlap.

  Yes, I did, it meant and still means he

  >

  The setInvokesStopCellEditing() docs say:

  + Determines what happens when the edition is interrupted by selecting another node in the tree, a change of data of the tree, or by other means. Setting this property a value of true causes changes to be automatically saved when editing is stopped. +

  Well, your fat is your old problem: assumptions yet ;-) My focus is

  + Determines what happens when the change is interrupted by selecting another node in the tree, a change of data of the tree, or by some other means.

  And which is supported by reading on:

  
       * @param newValue true means that stopCellEditing is invoked
       *        when editing is interrupted, and data is saved; false means that
       *        cancelCellEditing is invoked, and changes are lost
  

  in the total sense: he's interrupted (whatever that means, no conclusive specification except this model or the selection change), this propterty determines whether editing is stopped or cancelled. Conversely: if it is not interrupted, the property has no effect.

  Do you mean that the part of the sentence by other means does not include loses focus? I have to disagree. Of course loses focus is a "way" to interrupt the editing.

  Nothing to accept or not: is done, it's the 'interrupt' that matters. Obviously (although without doubt) lose focus don't consider an interruption - nothing happens when moving the cursor somewhere else, neither commit nor cancel.

  So if invokesStopCellEditing is defined, and the tree loses focus, it should save your work. And that's why the invokesStopCellEditing overlap with terminateEditOnFocusLost.

  just to drive home ;-) - the item no.

  >

  Note that I stick to the definition in the docs.

  No, you interpret the doc to fit your expectations.

  I did an example code and checked the validation works when the tree loses focus.

  You should have (it's a two-liner to add to your example) - see the divergence of expectations and reality, rethink your accession ;-)

  In which case it's bugged, because the docs say it should.

  In fact, I don't think it is (as you may have guessed so read up to here ) the text doc of course is not as easy as it could be, as is unfortunately a habit running in the swing. Assuming that means it that you wait that means is a trap. My general advice is to read the documentation of the api carefully and until its end and then check against the actual behavior expectations. The advice to take it or leave it - your choice, of course.

  CU
  Jeanette

• 3rd party Cookies will disappear with IE9 and Firefox 4? Impact on Marketing Automation?

  People, was curious to know how other people are considering (or have thoughts) all about losing third-party tracking with more recent browsers cookie?

  
  Looks like this could be very disruptive for all providers of marketing automation, or at least some sort of a switch to a party cookie should be first tracking system. If I read correctly, those systems it will be progressive, profiling, some stats CTR and certainly notification Web site visitor would be affected.
  
  More info here: http://paidcontent.org/article/419-mozilla-ceo-do-not-track-option-will-be-in-firefox-4/
  
  Thoughts?

  With the Firefox solution, we do nothing, since it requires their header and sites to work together. The challenge with Firefox and their solution is that the addition as a header-not-track, is that it requires browsers and sites to implement it to be fully effective. In other words, you can set the-no-track disengage all you like to feature in your Firefox browser, but if Web sites that track chooses to ignore the header-no-trail of Mozilla it will not do anything good. We are not ignoring the issue of Firefox, but believe that their solution is not one that will provide the appropriate consumer control.

  Regarding IE, they will be using a local list that the user must fill with sites that they do not want to be tracked on. There is not much we can do about that the consumer has the choice already today to block ALL cookies regardless of the type of part. This feature gives just more detailed control to the consumer at the level of the web address vs just block all third party cookies.

  We tested 'keep my Op-out Google' and it does not affect email and web click followed. It is only monitoring online ads.

  Finally, we are looking at or not stay with third-party cookie tracking or move to the first party. You can imagine my choice but there will be a huge change in how customers work must pass if we choose one over the other. More to come as soon as this.

  -Dennis

• strange behavior of link with OC assigned annotations

  I'm generating PDFs using a PDFL application I wrote. Link annotations are Act strangely when they are assigned to an o with the key of the OC. The behavior differs between Reader and Acrobat.

  In Reader, the hand tool slider meets the State of OC as expected. When the o is on the tool shows the link as being active. When you click on the link is highlighted, and the link to the action is executed. When the o is off the cursor indicates the inactive link. However, if the user clicks, highlights the link and the link to the action is executed.

  In Acrobat, the link is still active, regardless of the State of the o.

  I have a PDF file created with a much older version of this application that has links that behave normally. I got a chance to try to understand the difference between PDF files or code. The fact that the State of the cursor is at least correct in Reader and correct behavior of crime generally argues that optional content must be built correctly. I see no significant difference in the dictionaries of the annotation.

  Where should I look? We never seen anything like this? Any ideas would be very appreciated.

  TIA,

  Alan

  Workaround: Add a link annotation that is not asked to avoid the bug.

• Strange behavior of validation with the fields 'user entered - required '.

  I have a situation where I have a form with several 'user entered - required"text fields  Some required fields are pre-populated by a server process before being presented to a user in the workspace.  When the user fills the remaining fields and submits the form (by clicking the button "Done" in the workspace), the of standard validation 'required' message, even if all required fields contain data.

  The shape where I discovered this problem is moderately complex, so I built a form simple and process to try to understand what is happening.  The simple form contains exactly two text fields, both are user entered - required.  A field is filled by a server process when the task of the associated workspace is called, B field must be filled in by the user.  There is no script customized beyond the bridge of form and a Submit button.  It's about as simple as it gets.

  If the field is filled, and then the user fills B field and click on 'Finish', the message appears saying 'at least one required field is empty... "even if the two fields contain data.

  If I return to the form and remove the B field (and redeploy the form and remember the process), to field is pre-filled and the user clicks on 'Finish', no message is displayed and the form is sent.

  As far as I know, the built-in for required fields validation works correctly only if all fields are pre-filled or none of them are pre-filled.

  Can someone explain what is happening?

  EDIT: I forgot to mention, I also created and configured an extract of the data for the simple form.  The data preview only contains a value for the to field.  When displayed in the preview of the Designer tab, everything works as expected.

  After a little work I have a better understanding of what is happening and a solution.  I have scripted several calls to xfa.host.messageBox (...) on different events.  It turns out that the pre-filled fields are not the culprit - they have values as expected.  The problem is with the non-pre-filled fields, and and a subtle behavior of the event model that causes a problem when using the workspace INTERFACE (at least on SS3).  Field values may not be resolved until a field loses focus (for example when a user moves to another field, or click a button).  Form used outside of the workspace, by clicking a "submit" button causes the focus to move to the button, which causes the values to be bound to the fields.

  Solution: Use an indirect Submit on the form button and submit indirect click handler used (in addition to your normal code):

  xfa.host.setFocus(null);
  xfa.resolveNode("[[fully qualified name of the real submit button]]").execEvent("click");
  

  Working space should detect and then make an Indirect button and run the associated click Manager, claiming in turn the form submit true button.

  This is where begins the next bit of fun: there is a bug in the form bridge (at least in the ES3 version) code prevents him from indirect detection submit buttons when you use JavaScript as a language Manager click of the button and

  Here is my hack to solve this issue:

  1. Open the 'ContainerFoundation_JS' script in the Script of Designer Editor
  2. Ignore the warning at the top of the file
  3. Find the named function: isIndirectSubmitEvent (it's on the line 1512 or something)
  4. Change the line that reads:

     • if (script.indexOf(button.name + ".execEvent") != -1) {
       

       TO:

       if (script.indexOf(button.name + ".execEvent") != -1 || script.indexOf(button.name + "\").execEvent") != -1) {
       

  5. Save and deploy the form

  If there is a better way to achieve this, please share.

• Strange behavior of sorting with Collator

  Hi all, I'm sort of a list of Icelandic words using an assembler, and I get results quite specific (locale is is_IS, using jdk1.6.0_20).

  List<String> list = Arrays.asList(
          "Veiðifélag Hörðudalsár",
          "Veiðifélagið Búrfell (Hólmkelsá)",
          "Veiðifélag Hvammsár",
          "Veiðifélag Jökulsár á Dal",
          "Veiðifélagið Skrauma",
          "Veiðifélag Kiðafellsár",
          "Veiðifélagið Kolka");

  Collections.sort(list, Collator.getInstance());

  Resulting list (poorly sorted, the three points of "Veidifelagid...) ("which appears in the middle):
  
  Veidifelag Hvammsar
  Veidifelag Hordudalsar
  Veidifelagid Búrfell (Holmkelsa)
  Veidifelagid Kolka
  Veidifelagid Skrauma
  At Veidifelag Jokulsar Dal
  Veidifelag Kidafellsar

  Collections.sort(list);

  Resulting (sorted correctly) list:
  
  Veidifelag Hvammsar
  Veidifelag Hordudalsar
  At Veidifelag Jokulsar Dal
  Veidifelag Kidafellsar
  Veidifelagid Búrfell (Holmkelsa)
  Veidifelagid Kolka
  Veidifelagid Skrauma
  
  I'm having a hard time to see something logical on this subject. I tried to play with (as opposed to learning on / understanding) properties of strength/decomposition of the instance of the assembler, but it doesn't seem to have any effect.
  
  I feel like I'm totally wrong understand something here, someone can help me to clarify what it is?

  YoungWinston wrote:

  DrClap wrote:
  The assembler is remove the spaces before it does the scheduling. He does this in English too.

  Y at - it everywhere where that is documented DrClap?

  I have no idea. When I saw what happens to my data, it was immediately obvious to me, so I didn't feel the need to see if it is documented behavior. Neither fact I look to see if it was optional because I liked the behavior and did not want to change it. And you can see this is what happens with the Icelandic published data too.

• Generated code with edge of adobe for my site does not work with fireox, but done with IE9 and chrome. Why?

  The first time I opened Firefox with the code in my site, it has worked. Since then, it has not worked. I tried it on another computer and it worked perfectly. This computer is running under Windows XP.

  Reload Web pages, and ignore the cache.

  • Hold SHIFT and click reload.
  • Press 'Ctrl + F5' or 'Ctrl + Shift + R' (Windows, Linux)
  • Press 'Cmd + Shift + R' (MAC)

  Clear the cache and cookies from sites that cause problems.

  "Clear the Cache":

  • Tools > Options > advanced > network > storage (Cache) offline: 'clear now '.

  'Delete Cookies' sites causing problems:

  • Tools > Options > privacy > Cookies: "show the Cookies".

  With the help of "Forget this Site" remove all of this area like bookmarks, cookies, passwords, cache, history, and exceptions, so be careful if you have to keep the passwords or other data of this site.
  
  You can't recover from which "forget" unless you have a backup of the affected files.

  It has no lasting effect, so if come back you on such a 'forgotten' site, then the data of this Web site will be saved once more.

• Strange behavior on WLS in autonomous, and not on WLS integrated

  Hi all

  This project is in jdev 11.1.1.3.0. The stand-alone server version is 10.3.5.0 and the console/em 11 g.

  When I run the project on integrated WLS, inserting rows behaves as expected. When I deploy on a stand-alone server, the inserted rows suddenly have the default values that appear because it's attributes. The values exactly copy the other already existing lines to the table.

  Anyone experienced something like that or know the possible causes?

  Problem solved. It's the 11.1.1.5 insert with immediate = true bug button. We consider using the fix for this problem or live with immediate = false property.

• Strange behavior on DataGrid with ArrayCollection as a DataProvider collection

  I have a Datagrid with an ArrayCollection as a DataProvider collection, the arrayCollection collection is partially generated by a remoteObject call, the dataprovider seems to work at least until I try to change the field...

  By the RemoteObject, I get only a collection ArrayCollection with the 'ip' field, but the data grid seeks ip, check and recordfields...

  If I add/edit this field new it works, but only under specific condition

  The DataGrid control:

   <s:DataGrid id="datagrid" left="10" right="10" top="136" 
         dataProvider="{listaIPCheck}" bottom="10" requestedRowCount="4">
            <s:columns>
                  <s:ArrayList>
                       <s:GridColumn dataField="ip" headerText="Asset"/>
                       <s:GridColumn dataField="check" headerText="Inventory"/>
                       <s:GridColumn dataField="save" headerText="Salvataggio"/>
                  </s:ArrayList>
             </s:columns>
   </s:DataGrid>
  

  The Script:

     [Bindable]private var listaIPCheck:ArrayCollection; 
  
      private function ro_resultHandler(event:Event=null):void
      {
        listaIPCheck = new ArrayCollection();
        listaIPCheck = ro.getListUpdate.lastResult;
        heap = 0; 
        // Read Below {POINT #1}
        init3(); 
      }
  
  
      private function init3():void
      {
       // Read Below {POINT #2}
       if (heap<listaIPCheck.length)
       {
          // omitted the initialization of the process p
          p.addEventListener(NativeProcessExitEvent.EXIT, onExit);
          try{ 
            p.start(startupInfo);
          }catch(e:Error){}
       }
      }
  
  
  
      private function onExit(e:NativeProcessExitEvent):void { 
          // Read below {POINT #3}
      }
  

  Here is my code, now, as you can see there are 3 lines where I wrote to read below...

  Suppose that to put this simple 'for' instead of the commented (once both) line

      for (var k:Number=0;k<listaIPCheck.length;k++)
      {
        listaIPCheck.getItemAt(k).check = "checkVal";
        listaIPCheck.getItemAt(k).save = "saveVal";
      }
  

  This code still works in 3 points, so at the end of the call, the ArrayCollection collection is always filled with new values, but the datagrid control refresh the items only in the POINT #1 and #2 of POINT

  Why not the Point #3?

  When you get a large number of items in your list, you will probably also find it resets the selectedIndex property and jumps on the first line of data.  It is not automatically updated because the ArrayCollection collection look inside objects to see if they change.  When they change, you must indicate the ArrayCollection collection they changed, call itemUpdated.

• CSS with LI and a class

  I'm working on a site where a new page has been added. The existing stylesheet redefined the tag LI to show a green ball. There are tons of these bullets.

  Now the new page that they would like a section of chips with a different color.  I tried to do an inline style and a style for a new class, but it does not work.  I don't want to change all the other styles of pages.  Anyway around this?

  LI

  {

  list-style-image: url(images/bullet_green.jpg);

  Color: #000000;

  make-style: normal;

  text-decoration: none;

  list-style-position: inside;

  }

  What I want is

  subbox LI

  {

  list-style-image: url(images/bullet_orange.jpg);

  Color: #000000;

  make-style: normal;

  text-decoration: none;

  list-style-position: inside;

  }

  You are on the right track.

  sub-frame needs a point (or period) in front is a class.

  Li .subbox

  Then replace it

  You have a link?