Behavior of the VersionedPutAll

Hello

I am using the VersionedPutAll method, but the results I get are very weird

Given the VersionedPut documents, invoke should return a result if she can't insert (when you call as new VersionedPut(element,true,true)). From I have assumed that VersionedPutAll should return a map with the elements that could not be inserted. So I tried this:

Result = dataMap.invokeAll (map.keySet (), new VersionedPutAll (map, true, true));

Instead, I always get all the elements in the result regardless of whether they have been inserted or not.
What is the expected/current behavior of InvocableMap + VersionedPutAll?

Kind regards
Mikel.

user3211480 wrote:
Hello

I am using the VersionedPutAll method, but the results I get are very weird

Given the VersionedPut documents, invoke should return a result if she can't insert (when you call as new VersionedPut(element,true,true)). From I have assumed that VersionedPutAll should return a map with the elements that could not be inserted. So I tried this:

Result = dataMap.invokeAll (map.keySet (), new VersionedPutAll (map, true, true));

Instead, I always get all the elements in the result regardless of whether they have been inserted or not.
What is the expected/current behavior of InvocableMap + VersionedPutAll?

Kind regards
Mikel.

Hi Mikel,

in fact VersionedPutAll is implemented a little funny that the values returned are concerned. It returns a value of type java.util.Map in sheet result for each entry transformed if the fReturn flag is true. This per-entry-result is empty if the setting has been made the new value in the entry and if it wasn't properly then it contains the old value.

If you actually printed the content of the result of the method invokeAll, you will see the following text:

Result size: ConverterMap {ConverterEntry {key = '2', Value = "{"}, ConverterEntry {key = "1", value = "{}"}}

As you can see, this map contains empty maps indexed by 2 and 1. This means that set the value of these two Scriptures succeeded.

If you have run the invokeAll once more, you will see a different result:

Result size: ConverterMap {ConverterEntry {key = '2', Value = "{"}, ConverterEntry {key = "1", value = "{}"}}
Lead size: ConverterMap {ConverterEntry {key = '2', Value = "{2 = VPA.} MainVersionedPutAll$Test@364641}"}, ConverterEntry {key = value '1' ="{1 = VPA. MainVersionedPutAll$Test@13e754f}"}}

The second line shows that the second call could not update the two entries.

To reproduce this, you would either need a node storage disabled or you test code create new values for each call separately (see my version of your test code), because if you have a single node, then consistency optimizes serialization stage between sending the invocation processing it and the increment of the version attribute is takes place on the MainVersionedPutAll.Test objects initially passed so the second call would succeed Too much.

See my version of the test code:

public class MainVersionedPutAll {
     public static void main(String[] args) {

          NamedCache cardCache = CacheFactory.getCache("dist-test");

          cardCache.clear();

          putValues(cardCache);

          putValues(cardCache);
     }

     private static void putValues(NamedCache cardCache) {
          Map myMap = new HashMap();

          Test test = new Test();
          test.setId(1);
          test.setName("test1");
          myMap.put(test.getId(), test);

          Test test2 = new Test();
          test2.setId(2);
          test2.setName("test2");

          myMap.put(test2.getId(), test2);

          Map result = cardCache.invokeAll(myMap.keySet(), new VersionedPutAll(
                    myMap, true, true));

          System.out.println("Result size: " + result);
     }

     public static class Test implements Versionable, Serializable {
          // ...
     }
}

Best regards

Robert

Tags: Fusion Middleware

Similar Questions

Go back to the previous behavior in the address bar

Just upgraded to Firefox 31.2 and I find that the new behavior of the address bar is in conflict with how I work. I often copy / paste URL, I type directly into the address bar, I use a current to go to about.config tab and none of these work. I found a reference to the evolution of the gfx.xrender.enabled false, but this did not help. Clues on how to revert to the old behavior?

It's about: config with a colon...

Behavior of the filter action "stop filter execution.

What is the intended behavior of the filter action "stop filter execution"? I can't find any recent documentation on this issue. When part of the filters manually execution, "Stop filter execution" is intended to stop all other filters for all messages if the filter that contains both matches? This seems to be the behavior for me. When included in the first several filters filter, only the first filter matches and the log only shows instances of correspondence of the first filter, although several other subsequent filters must correspond to several messages that don't match not the first filter.

I appreciate the functionality of Thunderbird filters, but I try to find a solution to implement more complex logical Boolean to move messages from my Inbox with the help of manually run filters. I have several filters put in place to classify the messages, but I want to implement filters such that they are not applied to tracked messages. Currently I have the logic in most filters do not match a favorite message using the option 'Match all the following', however this prevents the use of the logic of the GOLD in the filter and led to an increase in the number of filters. I want to set up the filters as a first filter messages from games played and prevents other filters running only for messages that match the filter first.

Maybe the behavior I see is a bug specific to my system, but I would like to get a confirmation on the expected behavior and see if others have encountered this problem. I work under the assumption that the behavior of "Stop filter execution" must stop subsequent treatment only for messages on which a filter containing is as indicated in the old post to https://groups.google.com/forum/#! topic/mozilla.feedback.thunderbird.prerelease/rUXaQ0NdZvM, but this assumption may be wrong.

Any help or suggestions are appreciated.

OK, it seems that this is a bug documented, for example to https://bugzilla.mozilla.org/show_bug.cgi?id=552936. I should have searched for bug reports before posting here. I follow things with bug reports. Thank you.

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

FF13 different behavior on the two machines home page

FF gives different results than "Home Page" on both machines.

On the desktop, I get my set (the former google/firefox one) homepage.

On laptop, I have two tabs, the first of about: at home, the second is my fixed home page (the same old google/firefox one).

On the desktop when I press home, page becomes whole-home page.

On laptop when I press home, new tab is launched to give me my whole-home page.

The behavior on the desktop is what I expect and what I want. Which continues with Notepad and how do I do the same thing as the Office?

Check your settings on the laptop home page, you have inadvertently defined two pages as your home page...

See - How to set the home page

change in behavior of the button home and end

Hello
I have a problem with the behavior of the home, end, PG. PREC, PG. next and arrow keys. After that I have upgraded to version 6, they do not work as expected.
I disabled each extension and the behavior is always the same...

They work as intended in IE8.

I hope someone has a solution.

Make sure that NUMLOCK is not engaged, as the keys Home, End, PgUp and PgDn are often superimposed on the numeric keypad. There should be fire of keyboard for Numlock.

Press F7 to activate the navigation with the keyboard on or off, you want it out. Test after changing.

To reduce the problems of caret navigation, see

http://dmcritchie.MVPs.org/Firefox/Firefox-problems.htm#caret

Behavior of the modified mice

I have an Ideapad Y560 with a mouse Bluetooth Lenovo A - 300 M. Normally, when I click on a folder, the folder opens. When I place my cursor in Google, I am allowed to write. I can browse using my mouse.

But today, everything 'glue' to my cursor.

When I click on a folder, the folder stick (as I am dragging the file) to my cursor.

If you go to Google and put my cursor in the search box, the url of the Web site 'glue' to my cursor and 'glue' in the search box.

If I click on a white spot on my screen, the cursor created a blue indicated selection box that changes size when I move my cursor.

I press 'ESC' to take off everything that is glued to my cursor. I have to click several times to close a window or make a selection on a Web page. For the most part, I can't use my mouse because it does not work properly.

I have updated flash and run windows update. I restarted. I put new batteries in my mouse. I ran a virus check. I downloaded the latest driver of Lenovo mouse (from 2010!), but I get a message that my current driver is more current than to Lenovo. I deleted my mouse device and reinstalled. I changed browsers. I'm out of ideas.

Help?

It seems that your mouse broke down. Recently, I threw on a mouse who showed similar behavior. The problem is that the switch within the mouse gets stuck.

Unexpected behavior with the Option "record in the result.

Hello

I have unexpected behavior with the Option "record in the result.

I have a few steps in the subsequence 'X', this subsequence passes a Boolean parameter. According to the value of the parameter I change the "Recorgind results" Option to report it or not. The thing is that if 'result Recorgind' set at race time I modofy by changing the value of Step.ResultRecordingOption to "Enable" and "Disable", the step is not reported until the same sous-suite 'X' is called for the second time (without changing the parameter passed).

For example: (Preconditon: result Recorgind Option of all value sous-suite x are defined as Disable)

1 CallSubsequenceX(Parameter: Enable)

2 CallSubsequenceX(Parameter: Enable)

3 CallSubsequenceX(Parameter: Disable)

4 CallSubsequenceX(Parameter: Disable)

Expected result:

1. measures have been reported.

2. measures have been reported.

3. measures have not been reported.

4. measures have not been reported.

Result:

1. measures would not same value Step.ResultRecordingOption has been changed to 'enable '. (Not Ok)

2. measures have been reported. (Ok)

3. measures reported same value Step.ResultRecordingOption has been changed to 'disable '. (Not Ok)

4. measures have not been reported. (Ok)

I use TestStand 2013 (5.1.0.226)

Thanks in advance.

-Josymar.

Hi josymar_guzman,

I just review the sequence and indeed we´re experience unexpected behavior with the Step.ResultRecordingOption callback. By a reason when you run the callback in the expression before each step section, the statement runs only until the next sequence is called, which is not what we want.

To avoid this, you can place a statement before each step of the sequence, so you can change the State of the Option "record result" for the sequence running (and it is only the following). You can try something like this

where the expression of the statement will be the recall "RunState.NextStep.ResultRecordingOption is YourCondition". With this, we guarantee that the results of the next step will be saved or not. I also remove the expression in the expression prior to each step section, because the condition is now on the statement before each step.

I tried and it works fine. I´ll set the sequence that you share with me, with the changes. I hope this will help you and solve your problem.

Strange behavior of the WHILE loop

Dear experts LabVIEW

I would ask you to give me an explanation of the behavior of the very simple VI as an attachment. This VI contains a WHILE loop with three parameters, input and STEP. The table of RESULTS shows the calculated values. Considering the input parameters OF = 0 TO = 1 LEVEL = 0.1, then the RESULTS table contains the values 0, 0.1, 0.2,..., 0.9, 1. Considering the input settings ON = 1, TO = 2, STEP = 0.1, then RESULT table contains values 1, 1.1, 1.2,..., 1.8, 1.9. The question is, why the number 2 is missing? (I also tried to use different types of data, but without success). Thanks for your thoughts.

Juraj

Because you compare floating point numbers.

For example, 0.1 cannot be represented in exactly in binary, cela repetitive additions probably will not end in an exact integer later.

Since the three entries, you can calculate the number of values exactly and use a loop FOR. Do not use the convoluted code...

Why not just use ramp model?

PCI-6221 behavior off the power

Hello

A card PCI-6221, tension of the PC is out of snap-ins:

When I inject a voltage (5V) input ana on the map, this tension is copied on the other analogue channels.

Room I turn on the PC, and 6s after, injected tension is more copied on the other tracks of ana.

What is the normal behavior of the map?

Thank you pour your answers

Hello Cedric,

The behavior of the card when it is turned off is not defined. What is it you have a problem in your application?

When you area PC, this one seems to provide power to the bus PCI 6 seconds after it starts, which explains why the map then adopts its operation.

The best advice I can give you is to respect the configuration/connection of analog input (manual page 58 Chapter 4-14 M-series cards). About the card, if it works properly when the PC is turned on, and that the connection am well the guide, there is every reason to be reassured.

I wish you a good day,

Marc-Junior

Strange behavior of the browser blocks SQLDEV DM

Hello
I feel some strange behaviors in the browser of Datamodeler.
After the opening of my model
I have doubleclick in the window of the browser to fetch a few properties of the index:
Drawings and models > MyModel > relational models > my RelModel > models > Oracle Database 10g > Tables
So far, everything that happens.
-I can select the tables and
-doubleclicking shows the properties of the table.
-I can scroll vertically
-Select other tables
-open and close the subtree of this physical tables
Now, as soon as I click on any 'more' in the physical tables (for example, the index node or node of the plate) the browser "tri-de gel".
The tree is distorted, scrolling seems erratic and it is impossible to select. Most of the time the main menu is also distorted.
If I make the browser window, most of the time I'm able to hit the "file > save" from the menu. The model seems to be registered properly.
No damage seems to have been done to the template on the reopening
I'm on Windows 8 pro with the 64-bit version of datamodeler version 4.1.2.895 (from datamodeler - x 64 - 4.1.2 - 895.zip)
Any ideas anyone?
any response greatly appreciated.
concerning
Wouter

Hi Wouter,

I'm sorry, this is a bug and we are preparing a new version with this bug fixed.

If I make the browser window, most of the time I'm able to hit the "file > save.

Close the Data Modeler (X - close button in the upper right corner) and it will ask to save the drawing it is changed.

Philippe

Changed the behavior of the combination SHIFT + click in Editor?

In versions cc2015 of Photoshop, the checker behaves differently than older versions when I use 'shift-click' to connect the dots. In PS CS6, for example, when I take the SHIFT key, the clone tool in (or cure) continuously from point to point as I click. In the last two versions of Photoshop, this has changed: a click, then shift + click to a new location translates into a 'line', although it is not until the next point clicked, even by pressing the SHIFT key all the time. Instead, the point is considered as the beginning of a second 'line' of healing.
I hope that makes sense - it's surprisingly complicated to describe such a simple thing. But the effect is really disconcerting - if I'm along a surface of healing, I can't have a continuous 'cure' along a curve, for example.
Tools brush and Clone Stamp don't show it at all - they behave as expected, click connection points without interruption until the SHIFT key is released. I can't find any preference which may have changed, and I'm using a very simple toolset - without wild brushes or anything like that. Very frustrating!
How can I get the old behavior back to the Healing Brush tool? What Miss me?
Thank you
John
-running Mac OS X 10.10.5, Photoshop versions cc2015 and cc2015.5 show that (also another machine at work, OS 10.11, PS cc2015.5.1 shows the same behavior)

I have been informed more in 'PhotoshopFamily' that it is a bug, they sent to engineering to investigate (& difficulty, hopefully!)

«More tests, I found that this problem goes away * if * the source and the destination are on a layer "background" (e.g., on a flattened image).» But try to 'cure' / clone of a source layer to a layer of different destination (or even the same layer, if not a background layer) product failure in the behavior of the combination SHIFT + click. »

Strange behavior of the rectangle tool

I drew a rectangle with the rectangle tool and got a form than of a conical label around the rectangle. Never seen that before and I could not duplicate once I opened the application again. This seems a strange behavior of the app, or is there something that I don't know this tool? Please tell me this is something lacking in my knowledge base. It drives me crazy

Monika, yes it was. He worked with the forward Warp effect. Thank you for your response. Mental health is back thank you for your kindness

Different behavior on the eyes; layer names are correct?

Different behavior on the eyes; scales up and down, left right behaves as expected. All layer names are correct in artificial intelligence. Where to find a solution?

Hmm, the sent JPG seems to be broken.

But here's a place to look: in the Panel of the puppet, with no layers selected, you should see the behavior of face in the properties panel. Twirl in the group "Handles" to see what parts of the face it found. It shows the layers that you expect to find? (ToolTips will tell you what he's looking for.) Twirl down replacements also to see the layers of flashing. Note that if it does not find a layer of Blink for a certain look, it dimensionnera the eye that you close the eye rather than go to the work of replacement blink.

Viruses such as the behavior of the Acrobat Reader software. I have a laptop running WIN 7 and kaspersky AV. yesterday when I started my laptop that I see all the icons on my QuickStart bar browser have been changing drive adobe icon and observe also that

Viruses such as the behavior of the Acrobat Reader software. I have a laptop running WIN 7 and kaspersky AV. yesterday when I started my laptop I see all the icons from the browser on my QuickStart bar have been changed to the adobe reader icon and also observe that many other programs on my desktop have changed to the adobe reader icon. If you click the icon it opens adobe reader. I deleted Adobe reader using Add or remove programs and the icons back to normal. Yet once, I downloaded and installed Adobe reader 11. the problem has resurfaced. then again, I've uninstalled adobe reader software and installed Adobe Reader version 10 with the same question back. I have now uninstalled Adobe Reader. Please advise on how to install a working copy Adobe reader which will not infect or impinge upon other programs in my laptop.

Hi imoorthy,

Please follow the troubleshooting steps mentioned in this document KB Application, file icons change in Acrobat/Reader icon.

Kind regards

Nicos

Behavior of the VersionedPutAll

Similar Questions

Maybe you are looking for