Flex limit

Similar Questions

• I can catch a competitor receive a limit event?

  Hello

  11.1.1.7 BPM

  -J' have a process that has a correlation receive the event.

  -Under certain circumstances, I get a contradictory reception and I want to catch it and ignore it (rather than show as a flaw in Enterprise Manager).

  -J' have an error event on the receive activity limit catch a contradictory reception of SOAP (I also tried captures all exceptions) and the transition at the end.

  -L' error is never caught. Only, it shall terminate the proceedings and report the error in Enterprise Manager

  Exception is illustrated below.

  see you soon

  Tony

  < auditQueryPayload auditId = '5491' ciKey = "40130" >

  < dataState >

  < name of the DataObject = "FaultMessage" isBusinessIndicator = "false" >

  < Value > ORABPEL-03810 conflicting receive. Such a receive activity is declared in the same process.

  Another receive activity or equivalent (currently, onMessage branch a pick activity) has already been activated

  the partnerLink 'CorrelatedProcess.service', 'checkExists' operation and correlation set name

  "{{http://xmlns.oracle.com/bpmn/bpmnProcess/CorrelationTest} = theId {http://xmlns.oracle.com/bpmn/bpmnProcess/CorrelationTest} theId = 2}" (or conversation ID).

  Appendix A - standard flaws in the specification BPEL 1.1 Specifies a fault should be thrown under these conditions.

  Redeploy the process after removing the competitor receive activities. < / value >

  < / dataObject >

  < / dataState >

  < / auditQueryPayload >

  Hi Tony - good to hear from you.

  I think I know a few things related to correlations that you can run in (at least a couple that I hit in the past).

  If you create an instance of a process that uses the same correlation ID that you already used and it is still active in the process, you get undetectable except that I think you're getting.  Wish it was understandable and we were able to test at least to see if the correlation already exists before sending to the instance in a process and have it fail as I think you see.  If that's what you're running, have you considered adding a service activity that calls the API to query the process to see if a specific id has already been used?  If you have a simple scalar variable in your process and you have set to the id of a field of flex, you can use the queryTasks () API method to see if a value exists in all cases (I completely agree with you that there should be a way to have done this without the API).

  Here's another thing that bit me.  Do not know how your process is implemented, but assuming you have a receive activity, you might also have a problem of hydration.  Know this sounds nuts, but if you don't get the results that you think you should be in your return of the process called answers, try putting an event in the sequence flow immediately before your Receive activity, and set the timer to 1 second.

  Hope this helps, you get

  Dan

• Flex mobile app for iPhone too big?

  Hi, I got this message after I presented my Flex for iphone application to apple for approval.

  Is there anything I can do to help?

  It's a very simple app with 6 views.

  Dear developer,

  The binary app below was 7 MB when you submit it, but will be 20 MB after treatment for the App Store.

  This exceeds the size of cellular network download limit and would require your application to download on Wi - Fi:

  Since the iOS is based are DURATION captive all builds, they include the Air/Flex from Adobe code needed to run your code.  This adds some overhead to the size of the executable.

  The explosion of 7 MB to 20 MB is caused by Apple decompress the executable, encrypt to the App Store and recompression.  Encrypted data cannot be compressed as well as data unencrypted, which causes the size of the app up.

• Cannot change limit property of the TextInput via ActionScript

  Please tell me why this doesn't work, or suggest a work around.

  This example of debugging must change the TextInput to allow a dash when the box is checked, Yes?

  <? XML version = "1.0" encoding = "utf-8"? >
  " < = xmlns:fx s:Application ' http://ns.Adobe.com/MXML/2009 "
  xmlns:s = "library://ns.adobe.com/flex/spark".
  xmlns:MX = "library://ns.adobe.com/flex/mx" minWidth = "955" = "600" minHeight >

  < fx:Script >
  <! [CDATA]
  public function chkClicked (): void {}
  {if (CHK1. Selected)}
  trace ("box checked, allow-");
  TI1. Restrict = "0-9. \- » ;
  }
  else {}
  trace ("box unchecked, reject-'");
  TI1. Restrict = "0-9";
  }
  }
  []] >
  < / fx:Script >
  
  < mx:VBox >
  < s:CheckBox id = "chk1" click = "chkClicked ()" / > "
  < s:TextInput id = 'ti1' width = "60" limit = "0-9" / >
  < / mx:VBox >

  < / s:Application >

  Each of the restrict strings will work when it is applied directly as an mxml attribute.

  Thank you

  Al

  You must ESCAPE from the dashboard.

  Try.

  {if (CHK1. Selected)}
  trace ("box checked, allow-");
  TI1. Restrict = "0-9. \\- » ;
  }

  If this post was helpful please mark it as such.

  Post edited by: UbuntuPenguin

• Flex 4.5 RSL using modules and scheme

  Using RSL with modules and scheme

  RSL is designed to work effectively with the modules and the scheme. When you use RSL with modules and scheme, note that:

  • Main applications only load RSL that is necessary, but to create placeholders for the Framework RSL rest. This allows the sub-application or module load other Framework RSL into the main application when they need.

  • Scheme and modules share default RSL frame.

  • Scheme and modules can share custom RSL if you configure the RSLS loaded in the application domain.

  RSL using placeholder

  When you compile an application with Flex, the default behavior is to compile the application against the framework RSL (or by default). When the application runs, it only loads the Framework RSL, it actually uses and creates placeholders for the Framework RSL rest. When the application loads a module or a sub-application requiring one of the RSL setting for which there is a placeholder, the sub-application or module and then load that RSL into the main application.

  If a module or a sub-application requires a framework RSL already loaded, it will not load the RSL again. Instead, it will use the existing RSL.

  To ensure use of the placeholders for Framework RSL, set remove-unused-rsls the compiler argument is true when compiling the main application, scheme and modules. It is the default value.

  The default Framework RSL is loaded in the highest area of application that contains a placeholder for this RSL. As a result, modules or scheme can share RSL which had been loaded by other modules or the scheme, as long as you don't limit the area in which the RSL has been loaded.

  Could someone tell me if I understand:

  
  Seen this scenario:

  I have an RSL to a custom component. I use this custom only in a module of my application component. The result will be that the application does not load the RSL and it will load only when the module is loaded, but before the module.

  If that is correct, I have another question:

  The main application can follow the progress cargo module and display it with a for example progress bar. How to track the loading progress of the RSLS loaded before the module?

  OR

  The main application is the module loading, and when the module finishes loading I need to use a preloader class as of the main application for the module to track the loading of the RSL?

  Modules do not have a preloader as a demand. You can follow the status of the module being loaded, but not the status of the RSL. If you would like this feature please create an enhancement request in the JIRA bug database. Thank you.

  -Darrell

• Load an external Php variable in Actionscript in flex.

  Hello

  I have be trying to load an external variable from a php file in an actionscript variable.i know how to bind the php mysql data to datagrid but not what I'm looking for, example say a php variable has a value = 5, this value will be passed to a variable limit actionscript, so now limit will have the value 5.

  I use flex 3 and I have tried many examples listed on google, I went through all the tutorials, codes, etc... all of them were made using cs3 or cs4. And when I ran get them I have strength any output at all. Either I myself am thatproperty charger was not defined or nothing in my textarea.

  If possible please guide me on the next or even want to give a small sample that works on flex3.i am also unsure of what I shud write in my php file.

  I added the code to my mxml file for what does not work and shows me a blank text box.

  Any help would be appreciated... Thanks in advance

  <? XML version = "1.0" encoding = "utf-8"? >
  < mx:Application xmlns:mx = "httw://www.adobe.com/2006/mxml" xmlns:custom = "" test.external. * "layout ="absolute"initialize =" load() ">"
  < mx:Script >
  <! [CDATA]
  
  private function load(): void {}
  var loader: URLLoader = new URLLoader();
  loader.addEventListener (Event.COMPLETE, onComplete);
  Loader.Load (new URLRequest (https://localhost/external.php));
  }
  
  private void onComplete(e:Event):void {}
  Ext. Text = URLLoader (e.target) .data;
  }

  
  
  []] >
  < / mx:Script >
  < mx:TextArea id = "ext" x = "296" y = "199" >

  
  < / mx:TextArea >
  < mx:Button click = "load()" x = "343,5" y = "274" > "
  
  < / mx:Button >
  < / mx:Application >

  
  

  Your service works when she is called in the browser?

  C

• THEY Flash Remoting request limit introduced with Coldfusion 9?

  Hi all

  We've just been our Coldfusion servers since 8.01 CF CF 9. They run Win 2008. On servers that provide the Flash remoting backend for a Flex application, we provide, we ran into trouble.

  After upgrading CF 9 noticed that during busy when many customers Flex have been completed, we connect, we seemed to have a strict limit of 25 Flash Remoting running queries, despite the much higher limits (in fact 150) in CF Admin.

  We first thought that it was a problem with the fact that Blaze DS now ships with CF 9 (rather than a separate installation) we decided to roll back the installation of CF 9. This, unfortunately, has not worked and we were always stuck with out strict limit of 25 applications for Flash Remoting.

  Then looking at IIS we noticed the CF9 ISAPI filter was installed yet (after we ran the part of installing Web Service Configuration). Who took the CF 8 were re-run and all of a sudden the strict limit Flash Remoting has disappeared. So it seems that it might have had something to do with the wsconfig CF 9 (C:\ColdFusion9\runtime\bin\wsconfig.exe)

  If anyone has had this problem or does anyone know where these hard limits are configured in IIS?

  Any help appreciated!

  @ciaranarcher (and other interested parties in the present), I posted what, in my view, is the solution on the stackoverflow website where it had also posted the question:

  http://StackOverflow.com/questions/3037390/IIS-Flash-Remoting-request-limit-introduced-wit h-coldfusion-9

  Let us know (here and there) if it is the solution for you (and others). Thank you.

  /Charlie

• Flex Application Interaction with load Flash SWF

  Hi all

  First of all, the context:

  I'm creating a Flex application with a component that displays several SWF loaded dynamically, one at a time (a bit like a kiosk). These loaded SWF files is created in the Flash IDE, Flex No. They will not be created by us (we will provide a .fla but model it is), so I am loading them in a separate SecurityDomain.

  My Questions:

  1. I want the application to be able to call methods inside the loaded SWF file. To do this, I would just call Security.allowDomain ("flex application domain") in the boot process for the loaded SWF file?

  2. I want the SWF loaded to throw events that are taken by the application. Can I achieve this through the SWFLoader.swfBridge (http://www.adobe.com/livedocs/flex/3/langref/mx/controls/SWFLoader.html#swfBridge) property?

  3. I'm defining an interface for methods inside the loaded SWF file. But I don't want to leave the implementation up to the customer. Instead, I want to provide another SWF that contains the implementation of the interface and have a constraint that the customer includes it in their SWF file. I'm not very familiar with the Flash programming environment, how would that work so that the SWF interface implementation exposes his methods so as to be accessible to the Flex? Is it possible to sign the SWF file so that the Flex application knows that it is dealing with the original, not a parody? How would the Flex code pass in arguments, since it is through SecurityDomain limit - is there marshaling involved (for example if I want to pass in a dictionary or some kind of object graph)?

  4. The Flex application can load hundreds of these small other sovereign wealth funds during his life. How can ensure me that it is not continue to use more memory? I intend using SWFLoader.unloadAndStop () and erase all references to the Flex object that refers to it. Is this enough? Will be the AppDomain for the loaded SWF file to tear down automatically so that the class definitions are no longer in memory?

  Thank you

  Knockaert

  3 sounds good, but that's all theory.

  I think that there are other ways to package code to Flash other than "components".  I think that there are some libraries of data/network, but I don't know how they are packed these days.

  4. in theory, it will try to match fields that match and ignore those that do not fit as well.  With the help of readObject/writeObject should work, but you should be able to call the IExternalizable APIs directly.

  Alex Harui

  Flex SDK Developer

  Adobe Systems Inc..

  Blog: http://blogs.adobe.com/aharui

• How to restrict user access mutiple clicks with asynchrounous behavior of FLEX

  I developed a flex application and it has a datagrid control, and double click on the DataGrid, I show a pop-up window (the code in the pop-up window to call httpService.send () to get the data and display). Before the popup is open, the user is double clicking again on the data grid, which inturn is trying to open the pop-up window (with a new call to the web service). These calls are piling up. How can I limit using httpService.send () when the user double clicked for the second time on the datagrid control. Since then, call web service async by nature, I'm not able to find a way to do this. Can someone please help.

  set

  This.Enabled = false;

  After the service.

  and in the event of failure/result.

  This.enabled = true;

  Yet, all things BUSYCURSOR, because referring the good user.

• Multiple Flex applications on a single Page

  Now that I was laid off from Adobe (with 700 other people), I would ask you Flex consulting opportunities, and one of the possibilities are web designers/developers wishing to offer their customers a value-added components.

  My question is I've heard you don't want for example three separate applications to Flex on a single web page, as it results in three instances of the generated Flash Player, or something like that.

  So is it necessary to group multiple Flex components in a Flex application to each web page? This may limit the placement of Flex components on a web page, because the web page will be completely done in Flex, that the page should be available.

  No indication in this regard, especially Alex and others on Adobe's Flex product team.

  All new Flex needs help on Flex components/applications, I am available: [email protected]

  Thank you!

  Greg

  IMHO, it's all about memory.  Many pages have several flash banner ads running drives.  This is because these ads are small.  If you go to paste a few applications really small flex on a page, I would expect that most people have enough computer resources to manage it.  Once you get more complex, your apps also probably face more screen real estate and soon you are looking for a portal and probably should have a main shell that loads of apps.

  Alex Harui

  Flex SDK Developer

  Adobe Systems Inc..

  Blog: http://blogs.adobe.com/aharui

• Flex not including ArrayCollection collection passed to PHP

  First of all, I use Flex Builder 3, Wamp with 5.2.9 - 2 PHP and MySQL 5.1.33.  I also use the Zend Framework to retrieve data from MySQL.

  I have Flex using an HTTPService to send a request; the server is running then some PHP that reads a database MySQL, written each row in a table, all rows in a table.  Although I changed the actual data, here's what my returns a call on my MySQL database:

  SELECT  *
FROM  `table1`
LIMIT 0 , 30

  1 video filename.flv

  otherDescription 2 filename2.flv

  thirdDescription 3 filename3.flv

  The names of the columns are 'id', 'title', and 'source', respectively.

  Arrays in php seem to work very well, because in PHP,

  Print($result[1]['title']);

  We will give "otherDescription", and we can see that easily in Flex.

  The problem is I don't want "otherDescription; I want the whole table passed to Flex as a collection of table.

  I am able to pass the array from php, and then it's just a 'table' that I can't do anything with.  No matter what I trace him, he always says "Array", or it just doesn't work... Why is this?  Why can't I get anything out of it?  The thing is useful I can switch from PHP to Flex is a string, and it's awkward for what I do.

  Here is my PHP code:

  <? PHP

  require('Zend/DB/adapter/PDO/MySQL.php');

  $arr = array ('host' = > '127.0.0.1', 'username' = > 'root', 'password' = > ", 'database_name' = > 'videolist');

  $db = new Zend_Db_Adapter_Pdo_Mysql ($arr);

  $db-> getConnection();

  $sql = ' SELECT * FROM 'table1' LIMIT 0, 30';

  $result = $db-> fetchAll ($sql, 2);

  $resultArray = array();

  {foreach ($result as $resultItem)

  {foreach ($resultItem as $itemFragment)}

  [] $resultArrayFragment = $resultItem;

  }

  $resultArray [] = $resultArrayFragment;

  }

  echo $result;  or else, to illustrate how the collection of table, we can put "$result [1] ["title"]; ', in which case, Flex will show «otherDescription»

  to the traces below

  ? >

  Here are my relevant ActionScript:

  private function onResult (): void {//this is drawn, we get a result after calling the HTTPService send()

  var result = userRequest.lastResult;

  trace ("trace 1 =" + String (userRequest.lastResult));     product: 'table '.

  trace ("trace 2 =" + result [1] ["title"]);                           product: 'table '.

  trace ("trace 3 =" + result [1] ["title"] [0] [0] [0] [0]);            Product ALWAYS just: 'table '.

  var someDataProvider:ArrayCollection = new ArrayCollection (userRequest.lastResult.valueOf ());

  trace ("trace 4 =" + result [0]);                                     product: "RangeError: Index '1' specified is out of range...". »

  .

  .

  .

  }

  XML is fine for small amounts of data, you can use the object of the authority which is fast and lightweight. In my opinion, by far the simplest is Jason.

  PHP can encode and decode Jason, you can load a library in flex jason and decode and coded.

  Works perfectly, is now my main way to communicate with PHP.

  Google flex and jason there are tons of projects there.

• Can I limit the number of lines of text entered into a TextArea?

  Hello

  I use Flex Builder 2 to create an application that allows users to enter text in different TextAreas and eventually create a PDF file with their different text. My only problem is that I can't limit the number of lines of text used in each text box. I can limit characters by using maxChar, but this does not take into account a user entering multiple presses "return" - this does not affect the character limit, but can result in the final PDF layout (spitting the text on page 2).

  Is there a way to keep track and limit the number of lines of text in a text box (in Flex 2)?

  (Ideally I won't disable the ability for users to use the 'back' button to create several paragraphs)

  Any help would be appreciated.

  Thank you

  Ben.

  Hello

  Try this code.

  import

   

  flash.events.KeyboardEvent;

  Import flash.events.TextEvent;

  Import mx.controls.TextArea;

  public class LimitedLinesTextArea extends TextArea {}

  private var numLinesAllowed:int = 0;

  private var moreCharsAllowed:Boolean = true;

  public function LimitedLinesTextArea():void

  {

  this.addEventListener (flash.events.TextEvent.TEXT_INPUT, textInputHandler);

  }

  override protected function keyDownHandler(event:KeyboardEvent):void

  {

  if (this. textField.numLines > numLinesAllowed)

  moreCharsAllowed =

  fake;

  on the other

  moreCharsAllowed =

  true;

  Super.keyDownHandler (event);

  }

  private function textInputHandler(event:TextEvent):void

  {

  If

  (! moreCharsAllowed)

  Event.preventDefault ();

  }

  public function the value numberofLines(value:int):void

  {

  numLinesAllowed = value;

  }

  }

• Flex Remoting - deeply nested not strongly typed ArrayCollections

  I have a Flex 3 application that uses remote access. Everything works relatively well except for the following...

  I have a graph of highly nested objects as follows:

  (btw: each VO uses [RemoteClass(alias="com.mypackage.xxx")] meta tags)

  The main class is called 'BrandVO' that contains an ArrayCollection on ProductLineVOs collection.

  each "ProductLineVO" has a collection s ArrayCollection 'ProductVO.

  each "ProductVO' has a collection s ArrayCollection 'CaseVO' and 'UseVO's s 'IngredientVO '.

  Nickname graphically:

  +-BrandVO

  + - ProductLineVO

  + - ProductVO

  -CaseVO

  -UseVO

  -IngredientVO

  + - ProductVO

  -CaseVO

  -UseVO

  -IngredientVO

  + - ProductLineVO

  + - ProductVO

  -CaseVO

  -UseVO

  -IngredientVO

  + - ProductVO

  -CaseVO

  -UseVO

  -IngredientVO

  ... etc...

  When I do the communication distance call (AMF3), I properly repay an ArrayCollection of BrandVOs collection...

  (Debug displays each item in the ArrayCollection collection [0]...) [n] has a value com.mypackage.BrandVO(@1234etc)

  Expand each BrandVO shows correctly typed nested elements of ProductLineVO...

  (Debug displays each item in the ArrayCollection collection [0]...) [n] has a value com.mypackage.ProductLineVO(@1234etc)

  Expand each ProductLineVO shows correctly typed nested elements of ProductVO...

  (Debug displays each item in the ArrayCollection collection [0]...) [n] has a value of com.mypackage.ProductVO (@1234etc)

  The problem is then when expanding each ProductVO, nested CaseVOs, UseVOs and IngredientVOs (even if they contain appropriate data) are generic type.

  (CaseVOs Debug displays each item in each collection ArrayCollection [0]...) [n] has a value object (@1234etc)

  (Debugging UseVOsshows each element in each collection ArrayCollection [0]...) [n] has a value object (@1234etc)

  (IngredientVOs Debug displays each item in each collection ArrayCollection [0]...) [n] has a value object (@1234etc)

  I don't know why all the other VO typing correctly but CaseVOs, UseVOs and IngredientVOs are not strongly typed and default to generic objects.

  I checked the call data remote back (with ServiceCapture) and return objects have the appropriate types. The problem seems to be at the level of Flex.

  Anyone know why that would happen?

  Is there a limit to the object grpah depth beyind what objects get typed generically?

  I thought is maybe was because the ProductVO has 3 ArrayCollections (CaseVOs, UseVOs and IngredientVOs), then I tried to delete all except CaseVO and had the same problem... CaseVO is of type Object.

  Any help/thoughts would be gratly appreciated.

  ~ e

  I feel guilty of not asking you if you did. Yes, you need to do this, because if you don't refer to a class in your application, it does not include the Flex compiler in ABC - SWF.

• Y at - it a maximum size in ColdFusion, Flex or AMF

  I wish there was a forum of the AMF.

  I have CF version 8,0,0,176276, Flex 3.0.214193 and Oracle 10 g.

  I am building an application in Flex RemoteObject access to a ColdFusion .cfc via AMF channel. The application has several comment fields with are allowed to be over 4,000 characters, so in Oracle, they are defined as CLOB. Everything was fine until I started to test it with a few comments very long. Then the updates stopped working, but there was no mistake. I use CFFILE to empty the arguments passed to a file and found that one of the CLOB fields in where clause has been truncated. But then I don't know with certainty the CFFILE wasn't the one making the truncation. As far as I know with certainty is the update is not complete. None of the newspapers to show anything. I can't use a TRY/CATCH in the .cfc as the CFQUERY runs without error.

  Is there a limit of size in Flex RemoteObject? Maybe the limit is the authority for the financial markets. I have the ColdFusion limit set for 1 GB pagesize, but isn't a .cfc considered for pagesize?

  I don't know even where to start on this one. Any help you can give is appreciated.

  Scott

  answers (or at least discussion) continue here: http://forums.adobe.com/thread/414553?tstart=0

  You must close that one :-)

• Dynamic presentation in Flex 2?

  Hello
  
  I'm writing a SWF application that, after loading in the browser, read the data and the page view definition server and inspiring that definition is based on the page. All the meta-data for the page definition is in XML format and is generated dynamically (so cannot be pre-compiled). With the user clicks, the Flex code will need to blank the existing pages, read another definition of dynamic (like XML) page from the server and repaint the page.
  
  It is even possible in Flex, and if yes what is the best way to do this? I probably can do that if I pull the entire page in ActionScript dynamically instead of using the MXML based page layout? But is there a better way, I want to say is possible to send a new definition of the page in MXML to SWF code where actionscript could simply do something like "page.redraw (newDef)" where newDef is the new MXML?
  
  Essentially, if we do this generic page code converter SWF basically behave like an XSLT file only he would do on the browser.
  
  Any help would be greaty appreciated.
  
  Thank you
  
  -allnighter

  To understand our approach to Flex, you must know something of its history. Flex is based on Flash technology. Flash and Flash Player are very popular for its compact size, functions of diffusion and Ubiquity. Flash Player knows that sovereign wealth funds - it just runs the byte-code is given. Something which makes it very effective - and highly desired small in a web application. Because the SWF is a streaming media format, the player can begin to show the SWF as soon as sufficient information comes to market.

  MXML sending the player goes against the philosophy of the Flash platform. Should the Player contains a compiler and all kinds of JAR files. Could turn out to be so big and heavy, person would.

  Flex makes writing easier Flash movies for business developers because it allows them to work in a way that fits their environment: text files, control code source, centralized server, language familiar(-ish) and so on.

  When people see what Flex can do they get all kinds of wonderful ideas of how to apply the technology. But there is a limit today to what extent you can extend that. There is only so much available bandwidth from a server to a client, so the SWF must always be kept to a reasonable size. Data back and forth, too. Flash Player is in the same boat as any client/server application. But at least with Flash, you can send binary data compressed rather than HTML and leave the rendering on the client rather than the server.

  You can partition a large application upward in "application domains" - separate SWF that can be dynamically loaded and unloaded and you can use RSLS (runtime shared libraries) that contain a common INTERFACE elements (buttons, skins, etc.) shared between sovereign wealth funds.

  For example, if your application has reports and graphs, but one person never uses the cards, then the SWF chart never need to download. But when they go for a report, the report of SWF (s) can be loaded on demand. Already having most of their controls (buttons, text entry fields, etc.) in the player following a RSLS, means the report THAT SWF is much, much smaller and so quick to load.

  You might consider an approach like this: place of the common controls in an RSL, so that can be loaded with the main SWF. Generate your SWF as MXML page, then compile it on the server, indicating to the compiler to use the RSL. In this way the resulting SWF is low and can load faster.

  To your specifications, I don't see that it would work otherwise.