Search Results

Search found 7321 results on 293 pages for 'john smith optional'.

Page 36/293 | < Previous Page | 32 33 34 35 36 37 38 39 40 41 42 43  | Next Page >

  • Removing a Preceding Backslash from Apostrophes

    - by John
    Hello, When I add a comment using the variable below, apostrophes are printed with a backslash in front of them. How can I get rid of the backslashes? Thanks in advance, John Example of printed result: My roommate\'s brother\'s ex-girlfriend\'s aunt drive a Toyota. $comment = mysql_real_escape_string($_POST['comment']);

    Read the article

  • How to echo Hyperlink Containing a Variable?

    - by John
    Hello, In the code below, I would like to make the word that prints out as the variable "$submittor" a hyperlink to "http://www...com/.../members/index.php?profile=$submittor" . I can't get it to work; I think I'm doing the formatting wrong. How can I do it? Thanks in advance, John echo '<div class="sitename3name">Submitted by '.$submittor.' on '.$dt->format('F j, Y &\nb\sp &\nb\sp g:i a').'</div>';

    Read the article

  • How does the hash part in hash maps work?

    - by sub
    So there is this nice picture in the hash maps article on Wikipedia: Everything clear so far, except for the hash function in the middle. How can a function generate the right index from any string? Are the indexes integers in reality too? If yes, how can the function output 1 for John Smith, 2 for Lisa Smith, etc.?

    Read the article

  • After opening a mdb in a working copy

    - by John
    Hallo all, In my institute, Tortoise is employed for the purpose of version control. I find, if a mdb file which belongs to the working copy of a project reporitory is opened thru Access, the ordinary will be labelled with an explamation mark. Since the database has not been modified, I don't unterstand why Tortoise regards the opening as a kind of modification. Thanks in advance for any tips. John

    Read the article

  • Variable disappears when I log in

    - by John
    Hello, I have profile page where the profile is retrieved via GET. The index file has this: $profile = $_GET['profile']; When I log in on the profile page, the $profile variable disappears. Here is the form action on the login function: <form name="login-form" id="login-form" method="post" action="./index.php"> (The $profile variable is separate of the login username.) How could I make the page retain the $profile variable? Thanks in advance, John

    Read the article

  • Can't iterate over nestled dict in django

    - by fredrik
    Hi, Im trying to iterate over a nestled dict list. The first level works fine. But the second level is treated like a string not dict. In my template I have this: {% for product in Products %} <li> <p>{{ product }}</p> {% for partType in product.parts %} <p>{{ partType }}</p> {% for part in partType %} <p>{{ part }}</p> {% endfor %} {% endfor %} </li> {% endfor %} It's the {{ part }} that just list 1 char at the time based on partType. And it seams that it's treated like a string. I can however via dot notation reach all dict but not with a for loop. The current output looks like this: Color C o l o r Style S ..... The Products object looks like this in the log: [{'product': <models.Products.Product object at 0x1076ac9d0>, 'parts': {u'Color': {'default': u'Red', 'optional': [u'Red', u'Blue']}, u'Style': {'default': u'Nice', 'optional': [u'Nice']}, u'Size': {'default': u'8', 'optional': [u'8', u'8.5']}}}] What I trying to do is to pair together a dict/list for a product from a number of different SQL queries. The web handler looks like this: typeData = Products.ProductPartTypes.all() productData = Products.Product.all() langCode = 'en' productList = [] for product in productData: typeDict = {} productDict = {} for type in typeData: typeDict[type.typeId] = { 'default' : '', 'optional' : [] } productDict['product'] = product productDict['parts'] = typeDict defaultPartsData = Products.ProductParts.gql('WHERE __key__ IN :key', key = product.defaultParts) optionalPartsData = Products.ProductParts.gql('WHERE __key__ IN :key', key = product.optionalParts) for defaultPart in defaultPartsData: label = Products.ProductPartLabels.gql('WHERE __key__ IN :key AND partLangCode = :langCode', key = defaultPart.partLabelList, langCode = langCode).get() productDict['parts'][defaultPart.type.typeId]['default'] = label.partLangLabel for optionalPart in optionalPartsData: label = Products.ProductPartLabels.gql('WHERE __key__ IN :key AND partLangCode = :langCode', key = optionalPart.partLabelList, langCode = langCode).get() productDict['parts'][optionalPart.type.typeId]['optional'].append(label.partLangLabel) productList.append(productDict) logging.info(productList) templateData = { 'Languages' : Settings.Languges.all().order('langCode'), 'ProductPartTypes' : typeData, 'Products' : productList } I've tried making the dict in a number of different ways. Like first making a list, then a dict, used tulpes anything I could think of. Any help is welcome! Bouns: If someone have an other approach to the SQL quires, that is more then welcome. I feel that it kinda stupid to run that amount of quires. What is happening that each product part has a different label base on langCode. ..fredrik

    Read the article

  • Friendly way to parse XDocument

    - by Oli
    I have a class that various different XML schemes are created from. I create the various dynamic XDocuments via one (Very long) statement using conditional operators for optional elements and attributes. I now need to convert the XDocuments back to the class but as they are coming from different schemes many elements and sub elements may be optional. The only way I know of doing this is to use a lot of if statements. This approach doesn't seem very LINQ and uses a great deal more code than when I create the XDocument so I wondered if there is a better way to do this? An example would be to get <?xml version="1.0"?> <root xmlns="somenamespace"> <object attribute1="This is Optional" attribute2="This is required"> <element1>Required</element1> <element1>Optional</element1> <List1> Optional List Of Elements </List1> <List2> Required List Of Elements </List2> </object> </root> Into public class Object() { public string Attribute1; public string Attribute2; public string Element1; public string Element2; public List<ListItem1> List1; public List<ListItem2> List2; } In a more LINQ friendly way than this: public bool ParseXDocument(string xml) { XNamespace xn = "somenamespace"; XDocument document = XDocument.Parse(xml); XElement elementRoot = description.Element(xn + "root"); if (elementRoot != null) { //Get Object Element XElement elementObject = elementRoot.Element(xn + "object"); if(elementObject != null) { if(elementObject.Attribute(xn + "attribute1") != null) { Attribute1 = elementObject.Attribute(xn + "attribute1"); } if(elementObject.Attribute(xn + "attribute2") != null) { Attribute2 = elementObject.Attribute(xn + "attribute2"); } else { //This is a required Attribute so return false return false; } //If, If/Elses get deeper and deeper for the next elements and lists etc.... } else { //Object is a required element so return false return false; } } else { //Root is a required element so return false return false; } return true; } Update: Just to clarify the ParseXDocument method is inside the "Object" class. Every time an xml document is received the Object class instance has some or all of it's values updated.

    Read the article

  • JMS Step 4 - How to Create an 11g BPEL Process Which Writes a Message Based on an XML Schema to a JMS Queue

    - by John-Brown.Evans
    JMS Step 4 - How to Create an 11g BPEL Process Which Writes a Message Based on an XML Schema to a JMS Queue ol{margin:0;padding:0} .c11_4{vertical-align:top;width:129.8pt;border-style:solid;background-color:#f3f3f3;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c9_4{vertical-align:top;width:207pt;border-style:solid;background-color:#f3f3f3;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt}.c14{vertical-align:top;width:207pt;border-style:solid;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c17_4{vertical-align:top;width:129.8pt;border-style:solid;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c7_4{vertical-align:top;width:130pt;border-style:solid;border-color:#000000;border-width:1pt;padding:0pt 5pt 0pt 5pt} .c19_4{vertical-align:top;width:468pt;border-style:solid;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c22_4{background-color:#ffffff} .c20_4{list-style-type:disc;margin:0;padding:0} .c6_4{font-size:8pt;font-family:"Courier New"} .c24_4{color:inherit;text-decoration:inherit} .c23_4{color:#1155cc;text-decoration:underline} .c0_4{height:11pt;direction:ltr} .c10_4{font-size:10pt;font-family:"Courier New"} .c3_4{padding-left:0pt;margin-left:36pt} .c18_4{font-size:8pt} .c8_4{text-align:center} .c12_4{background-color:#ffff00} .c2_4{font-weight:bold} .c21_4{background-color:#00ff00} .c4_4{line-height:1.0} .c1_4{direction:ltr} .c15_4{background-color:#f3f3f3} .c13_4{font-family:"Courier New"} .c5_4{font-style:italic} .c16_4{border-collapse:collapse} .title{padding-top:24pt;line-height:1.15;text-align:left;color:#000000;font-size:36pt;font-family:"Arial";font-weight:bold;padding-bottom:6pt} .subtitle{padding-top:18pt;line-height:1.15;text-align:left;color:#666666;font-style:italic;font-size:24pt;font-family:"Georgia";padding-bottom:4pt} li{color:#000000;font-size:10pt;font-family:"Arial"} p{color:#000000;font-size:10pt;margin:0;font-family:"Arial"} h1{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:18pt;font-family:"Arial";font-weight:normal;padding-bottom:0pt} h2{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:18pt;font-family:"Arial";font-weight:bold;padding-bottom:0pt} h3{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:14pt;font-family:"Arial";font-weight:normal;padding-bottom:0pt} h4{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-style:italic;font-size:11pt;font-family:"Arial";padding-bottom:0pt} h5{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:10pt;font-family:"Arial";font-weight:normal;padding-bottom:0pt} h6{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-style:italic;font-size:10pt;font-family:"Arial";padding-bottom:0pt} This post continues the series of JMS articles which demonstrate how to use JMS queues in a SOA context. The previous posts were: JMS Step 1 - How to Create a Simple JMS Queue in Weblogic Server 11g JMS Step 2 - Using the QueueSend.java Sample Program to Send a Message to a JMS Queue JMS Step 3 - Using the QueueReceive.java Sample Program to Read a Message from a JMS Queue In this example we will create a BPEL process which will write (enqueue) a message to a JMS queue using a JMS adapter. The JMS adapter will enqueue the full XML payload to the queue. This sample will use the following WebLogic Server objects. The first two, the Connection Factory and JMS Queue, were created as part of the first blog post in this series, JMS Step 1 - How to Create a Simple JMS Queue in Weblogic Server 11g. If you haven't created those objects yet, please see that post for details on how to do so. The Connection Pool will be created as part of this example. Object Name Type JNDI Name TestConnectionFactory Connection Factory jms/TestConnectionFactory TestJMSQueue JMS Queue jms/TestJMSQueue eis/wls/TestQueue Connection Pool eis/wls/TestQueue 1. Verify Connection Factory and JMS Queue As mentioned above, this example uses a WLS Connection Factory called TestConnectionFactory and a JMS queue TestJMSQueue. As these are prerequisites for this example, let us verify they exist. Log in to the WebLogic Server Administration Console. Select Services > JMS Modules > TestJMSModule You should see the following objects: If not, or if the TestJMSModule is missing, please see the abovementioned article and create these objects before continuing. 2. Create a JMS Adapter Connection Pool in WebLogic Server The BPEL process we are about to create uses a JMS adapter to write to the JMS queue. The JMS adapter is deployed to the WebLogic server and needs to be configured to include a connection pool which references the connection factory associated with the JMS queue. In the WebLogic Server Console Go to Deployments > Next and select (click on) the JmsAdapter Select Configuration > Outbound Connection Pools and expand oracle.tip.adapter.jms.IJmsConnectionFactory. This will display the list of connections configured for this adapter. For example, eis/aqjms/Queue, eis/aqjms/Topic etc. These JNDI names are actually quite confusing. We are expecting to configure a connection pool here, but the names refer to queues and topics. One would expect these to be called *ConnectionPool or *_CF or similar, but to conform to this nomenclature, we will call our entry eis/wls/TestQueue . This JNDI name is also the name we will use later, when creating a BPEL process to access this JMS queue! Select New, check the oracle.tip.adapter.jms.IJmsConnectionFactory check box and Next. Enter JNDI Name: eis/wls/TestQueue for the connection instance, then press Finish. Expand oracle.tip.adapter.jms.IJmsConnectionFactory again and select (click on) eis/wls/TestQueue The ConnectionFactoryLocation must point to the JNDI name of the connection factory associated with the JMS queue you will be writing to. In our example, this is the connection factory called TestConnectionFactory, with the JNDI name jms/TestConnectionFactory.( As a reminder, this connection factory is contained in the JMS Module called TestJMSModule, under Services > Messaging > JMS Modules > TestJMSModule which we verified at the beginning of this document. )Enter jms/TestConnectionFactory  into the Property Value field for Connection Factory Location. After entering it, you must press Return/Enter then Save for the value to be accepted. If your WebLogic server is running in Development mode, you should see the message that the changes have been activated and the deployment plan successfully updated. If not, then you will manually need to activate the changes in the WebLogic server console. Although the changes have been activated, the JmsAdapter needs to be redeployed in order for the changes to become effective. This should be confirmed by the message Remember to update your deployment to reflect the new plan when you are finished with your changes as can be seen in the following screen shot: The next step is to redeploy the JmsAdapter.Navigate back to the Deployments screen, either by selecting it in the left-hand navigation tree or by selecting the “Summary of Deployments” link in the breadcrumbs list at the top of the screen. Then select the checkbox next to JmsAdapter and press the Update button On the Update Application Assistant page, select “Redeploy this application using the following deployment files” and press Finish. After a few seconds you should get the message that the selected deployments were updated. The JMS adapter configuration is complete and it can now be used to access the JMS queue. To summarize: we have created a JMS adapter connection pool connector with the JNDI name jms/TestConnectionFactory. This is the JNDI name to be accessed by a process such as a BPEL process, when using the JMS adapter to access the previously created JMS queue with the JNDI name jms/TestJMSQueue. In the following step, we will set up a BPEL process to use this JMS adapter to write to the JMS queue. 3. Create a BPEL Composite with a JMS Adapter Partner Link This step requires that you have a valid Application Server Connection defined in JDeveloper, pointing to the application server on which you created the JMS Queue and Connection Factory. You can create this connection in JDeveloper under the Application Server Navigator. Give it any name and be sure to test the connection before completing it. This sample will use the connection name jbevans-lx-PS5, as that is the name of the connection pointing to my SOA PS5 installation. When using a JMS adapter from within a BPEL process, there are various configuration options, such as the operation type (consume message, produce message etc.), delivery mode and message type. One of these options is the choice of the format of the JMS message payload. This can be structured around an existing XSD, in which case the full XML element and tags are passed, or it can be opaque, meaning that the payload is sent as-is to the JMS adapter. In the case of an XSD-based message, the payload can simply be copied to the input variable of the JMS adapter. In the case of an opaque message, the JMS adapter’s input variable is of type base64binary. So the payload needs to be converted to base64 binary first. I will go into this in more detail in a later blog entry. This sample will pass a simple message to the adapter, based on the following simple XSD file, which consists of a single string element: stringPayload.xsd <?xml version="1.0" encoding="windows-1252" ?> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns="http://www.example.org" targetNamespace="http://www.example.org" elementFormDefault="qualified" <xsd:element name="exampleElement" type="xsd:string"> </xsd:element> </xsd:schema> The following steps are all executed in JDeveloper. The SOA project will be created inside a JDeveloper Application. If you do not already have an application to contain the project, you can create a new one via File > New > General > Generic Application. Give the application any name, for example JMSTests and, when prompted for a project name and type, call the project JmsAdapterWriteWithXsd and select SOA as the project technology type. If you already have an application, continue below. Create a SOA Project Create a new project and choose SOA Tier > SOA Project as its type. Name it JmsAdapterWriteSchema. When prompted for the composite type, choose Composite With BPEL Process. When prompted for the BPEL Process, name it JmsAdapterWriteSchema too and choose Synchronous BPEL Process as the template. This will create a composite with a BPEL process and an exposed SOAP service. Double-click the BPEL process to open and begin editing it. You should see a simple BPEL process with a Receive and Reply activity. As we created a default process without an XML schema, the input and output variables are simple strings. Create an XSD File An XSD file is required later to define the message format to be passed to the JMS adapter. In this step, we create a simple XSD file, containing a string variable and add it to the project. First select the xsd item in the left-hand navigation tree to ensure that the XSD file is created under that item. Select File > New > General > XML and choose XML Schema. Call it stringPayload.xsd and when the editor opens, select the Source view. then replace the contents with the contents of the stringPayload.xsd example above and save the file. You should see it under the xsd item in the navigation tree. Create a JMS Adapter Partner Link We will create the JMS adapter as a service at the composite level. If it is not already open, double-click the composite.xml file in the navigator to open it. From the Component Palette, drag a JMS adapter over onto the right-hand swim lane, under External References. This will start the JMS Adapter Configuration Wizard. Use the following entries: Service Name: JmsAdapterWrite Oracle Enterprise Messaging Service (OEMS): Oracle Weblogic JMS AppServer Connection: Use an existing application server connection pointing to the WebLogic server on which the above JMS queue and connection factory were created. You can use the “+” button to create a connection directly from the wizard, if you do not already have one. This example uses a connection called jbevans-lx-PS5. Adapter Interface > Interface: Define from operation and schema (specified later) Operation Type: Produce Message Operation Name: Produce_message Destination Name: Press the Browse button, select Destination Type: Queues, then press Search. Wait for the list to populate, then select the entry for TestJMSQueue , which is the queue created earlier. JNDI Name: The JNDI name to use for the JMS connection. This is probably the most important step in this exercise and the most common source of error. This is the JNDI name of the JMS adapter’s connection pool created in the WebLogic Server and which points to the connection factory. JDeveloper does not verify the value entered here. If you enter a wrong value, the JMS adapter won’t find the queue and you will get an error message at runtime, which is very difficult to trace. In our example, this is the value eis/wls/TestQueue . (See the earlier step on how to create a JMS Adapter Connection Pool in WebLogic Server for details.) MessagesURL: We will use the XSD file we created earlier, stringPayload.xsd to define the message format for the JMS adapter. Press the magnifying glass icon to search for schema files. Expand Project Schema Files > stringPayload.xsd and select exampleElement: string. Press Next and Finish, which will complete the JMS Adapter configuration. Wire the BPEL Component to the JMS Adapter In this step, we link the BPEL process/component to the JMS adapter. From the composite.xml editor, drag the right-arrow icon from the BPEL process to the JMS adapter’s in-arrow. This completes the steps at the composite level. 4. Complete the BPEL Process Design Invoke the JMS Adapter Open the BPEL component by double-clicking it in the design view of the composite.xml, or open it from the project navigator by selecting the JmsAdapterWriteSchema.bpel file. This will display the BPEL process in the design view. You should see the JmsAdapterWrite partner link under one of the two swim lanes. We want it in the right-hand swim lane. If JDeveloper displays it in the left-hand lane, right-click it and choose Display > Move To Opposite Swim Lane. An Invoke activity is required in order to invoke the JMS adapter. Drag an Invoke activity between the Receive and Reply activities. Drag the right-hand arrow from the Invoke activity to the JMS adapter partner link. This will open the Invoke editor. The correct default values are entered automatically and are fine for our purposes. We only need to define the input variable to use for the JMS adapter. By pressing the green “+” symbol, a variable of the correct type can be auto-generated, for example with the name Invoke1_Produce_Message_InputVariable. Press OK after creating the variable. ( For some reason, while I was testing this, the JMS Adapter moved back to the left-hand swim lane again after this step. There is no harm in leaving it there, but I find it easier to follow if it is in the right-hand lane, because I kind-of think of the message coming in on the left and being routed through the right. But you can follow your personal preference here.) Assign Variables Drag an Assign activity between the Receive and Invoke activities. We will simply copy the input variable to the JMS adapter and, for completion, so the process has an output to print, again to the process’s output variable. Double-click the Assign activity and create two Copy rules: for the first, drag Variables > inputVariable > payload > client:process > client:input_string to Invoke1_Produce_Message_InputVariable > body > ns2:exampleElement for the second, drag the same input variable to outputVariable > payload > client:processResponse > client:result This will create two copy rules, similar to the following: Press OK. This completes the BPEL and Composite design. 5. Compile and Deploy the Composite We won’t go into too much detail on how to compile and deploy. In JDeveloper, compile the process by pressing the Make or Rebuild icons or by right-clicking the project name in the navigator and selecting Make... or Rebuild... If the compilation is successful, deploy it to the SOA server connection defined earlier. (Right-click the project name in the navigator, select Deploy to Application Server, choose the application server connection, choose the partition on the server (usually default) and press Finish. You should see the message ---- Deployment finished. ---- in the Deployment frame, if the deployment was successful. 6. Test the Composite This is the exciting part. Open two tabs in your browser and log in to the WebLogic Administration Console in one tab and the Enterprise Manager 11g Fusion Middleware Control (EM) for your SOA installation in the other. We will use the Console to monitor the messages being written to the queue and the EM to execute the composite. In the Console, go to Services > Messaging > JMS Modules > TestJMSModule > TestJMSQueue > Monitoring. Note the number of messages under Messages Current. In the EM, go to SOA > soa-infra (soa_server1) > default (or wherever you deployed your composite to) and click on JmsAdapterWriteSchema [1.0], then press the Test button. Under Input Arguments, enter any string into the text input field for the payload, for example Test Message then press Test Web Service. If the instance is successful you should see the same text in the Response message, “Test Message”. In the Console, refresh the Monitoring screen to confirm a new message has been written to the queue. Check the checkbox and press Show Messages. Click on the newest message and view its contents. They should include the full XML of the entered payload. 7. Troubleshooting If you get an exception similar to the following at runtime ... BINDING.JCA-12510 JCA Resource Adapter location error. Unable to locate the JCA Resource Adapter via .jca binding file element The JCA Binding Component is unable to startup the Resource Adapter specified in the element: location='eis/wls/QueueTest'. The reason for this is most likely that either 1) the Resource Adapters RAR file has not been deployed successfully to the WebLogic Application server or 2) the '' element in weblogic-ra.xml has not been set to eis/wls/QueueTest. In the last case you will have to add a new WebLogic JCA connection factory (deploy a RAR). Please correct this and then restart the Application Server at oracle.integration.platform.blocks.adapter.fw.AdapterBindingException. createJndiLookupException(AdapterBindingException.java:130) at oracle.integration.platform.blocks.adapter.fw.jca.cci. JCAConnectionManager$JCAConnectionPool.createJCAConnectionFactory (JCAConnectionManager.java:1387) at oracle.integration.platform.blocks.adapter.fw.jca.cci. JCAConnectionManager$JCAConnectionPool.newPoolObject (JCAConnectionManager.java:1285) ... then this is very likely due to an incorrect JNDI name entered for the JMS Connection in the JMS Adapter Wizard. Recheck those steps. The error message prints the name of the JNDI name used. In this example, it was incorrectly entered as eis/wls/QueueTest instead of eis/wls/TestQueue. This concludes this example. Best regards John-Brown Evans Oracle Technology Proactive Support Delivery

    Read the article

  • JMS Step 5 - How to Create an 11g BPEL Process Which Reads a Message Based on an XML Schema from a JMS Queue

    - by John-Brown.Evans
    JMS Step 5 - How to Create an 11g BPEL Process Which Reads a Message Based on an XML Schema from a JMS Queue .jblist{list-style-type:disc;margin:0;padding:0;padding-left:0pt;margin-left:36pt} ol{margin:0;padding:0} .c12_5{vertical-align:top;width:468pt;border-style:solid;background-color:#f3f3f3;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c8_5{vertical-align:top;border-style:solid;border-color:#000000;border-width:1pt;padding:5pt 5pt 0pt 5pt} .c10_5{vertical-align:top;width:207pt;border-style:solid;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c14_5{vertical-align:top;border-style:solid;border-color:#000000;border-width:1pt;padding:0pt 5pt 0pt 5pt} .c21_5{background-color:#ffffff} .c18_5{color:#1155cc;text-decoration:underline} .c16_5{color:#666666;font-size:12pt} .c5_5{background-color:#f3f3f3;font-weight:bold} .c19_5{color:inherit;text-decoration:inherit} .c3_5{height:11pt;text-align:center} .c11_5{font-weight:bold} .c20_5{background-color:#00ff00} .c6_5{font-style:italic} .c4_5{height:11pt} .c17_5{background-color:#ffff00} .c0_5{direction:ltr} .c7_5{font-family:"Courier New"} .c2_5{border-collapse:collapse} .c1_5{line-height:1.0} .c13_5{background-color:#f3f3f3} .c15_5{height:0pt} .c9_5{text-align:center} .title{padding-top:24pt;line-height:1.15;text-align:left;color:#000000;font-size:36pt;font-family:"Arial";font-weight:bold;padding-bottom:6pt} .subtitle{padding-top:18pt;line-height:1.15;text-align:left;color:#666666;font-style:italic;font-size:24pt;font-family:"Georgia";padding-bottom:4pt} li{color:#000000;font-size:10pt;font-family:"Arial"} p{color:#000000;font-size:10pt;margin:0;font-family:"Arial"} h1{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:24pt;font-family:"Arial";font-weight:normal} h2{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:18pt;font-family:"Arial";font-weight:normal} h3{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:14pt;font-family:"Arial";font-weight:normal} h4{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:12pt;font-family:"Arial";font-weight:normal} h5{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:11pt;font-family:"Arial";font-weight:normal} h6{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:10pt;font-family:"Arial";font-weight:normal} Welcome to another post in the series of blogs which demonstrates how to use JMS queues in a SOA context. The previous posts were: JMS Step 1 - How to Create a Simple JMS Queue in Weblogic Server 11g JMS Step 2 - Using the QueueSend.java Sample Program to Send a Message to a JMS Queue JMS Step 3 - Using the QueueReceive.java Sample Program to Read a Message from a JMS Queue JMS Step 4 - How to Create an 11g BPEL Process Which Writes a Message Based on an XML Schema to a JMS Queue Today we will create a BPEL process which will read (dequeue) the message from the JMS queue, which we enqueued in the last example. The JMS adapter will dequeue the full XML payload from the queue. 1. Recap and Prerequisites In the previous examples, we created a JMS Queue, a Connection Factory and a Connection Pool in the WebLogic Server Console. Then we designed and deployed a BPEL composite, which took a simple XML payload and enqueued it to the JMS queue. In this example, we will read that same message from the queue, using a JMS adapter and a BPEL process. As many of the configuration steps required to read from that queue were done in the previous samples, this one will concentrate on the new steps. A summary of the required objects is listed below. To find out how to create them please see the previous samples. They also include instructions on how to verify the objects are set up correctly. WebLogic Server Objects Object Name Type JNDI Name TestConnectionFactory Connection Factory jms/TestConnectionFactory TestJMSQueue JMS Queue jms/TestJMSQueue eis/wls/TestQueue Connection Pool eis/wls/TestQueue Schema XSD File The following XSD file is used for the message format. It was created in the previous example and will be copied to the new process. stringPayload.xsd <?xml version="1.0" encoding="windows-1252" ?> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema"                 xmlns="http://www.example.org"                 targetNamespace="http://www.example.org"                 elementFormDefault="qualified">   <xsd:element name="exampleElement" type="xsd:string">   </xsd:element> </xsd:schema> JMS Message After executing the previous samples, the following XML message should be in the JMS queue located at jms/TestJMSQueue: <?xml version="1.0" encoding="UTF-8" ?><exampleElement xmlns="http://www.example.org">Test Message</exampleElement> JDeveloper Connection You will need a valid Application Server Connection in JDeveloper pointing to the SOA server which the process will be deployed to. 2. Create a BPEL Composite with a JMS Adapter Partner Link In the previous example, we created a composite in JDeveloper called JmsAdapterWriteSchema. In this one, we will create a new composite called JmsAdapterReadSchema. There are probably many ways of incorporating a JMS adapter into a SOA composite for incoming messages. One way is design the process in such a way that the adapter polls for new messages and when it dequeues one, initiates a SOA or BPEL instance. This is possibly the most common use case. Other use cases include mid-flow adapters, which are activated from within the BPEL process. In this example we will use a polling adapter, because it is the most simple to set up and demonstrate. But it has one disadvantage as a demonstrative model. When a polling adapter is active, it will dequeue all messages as soon as they reach the queue. This makes it difficult to monitor messages we are writing to the queue, because they will disappear from the queue as soon as they have been enqueued. To work around this, we will shut down the composite after deploying it and restart it as required. (Another solution for this would be to pause the consumption for the queue and resume consumption again if needed. This can be done in the WLS console JMS-Modules -> queue -> Control -> Consumption -> Pause/Resume.) We will model the composite as a one-way incoming process. Usually, a BPEL process will do something useful with the message after receiving it, such as passing it to a database or file adapter, a human workflow or external web service. But we only want to demonstrate how to dequeue a JMS message using BPEL and a JMS adapter, so we won’t complicate the design with further activities. However, we do want to be able to verify that we have read the message correctly, so the BPEL process will include a small piece of embedded java code, which will print the message to standard output, so we can view it in the SOA server’s log file. Alternatively, you can view the instance in the Enterprise Manager and verify the message. The following steps are all executed in JDeveloper. Create the project in the same JDeveloper application used for the previous examples or create a new one. Create a SOA Project Create a new project and choose SOA Tier > SOA Project as its type. Name it JmsAdapterReadSchema. When prompted for the composite type, choose Empty Composite. Create a JMS Adapter Partner Link In the composite editor, drag a JMS adapter over from the Component Palette to the left-hand swim lane, under Exposed Services. This will start the JMS Adapter Configuration Wizard. Use the following entries: Service Name: JmsAdapterRead Oracle Enterprise Messaging Service (OEMS): Oracle WebLogic JMS AppServer Connection: Use an application server connection pointing to the WebLogic server on which the JMS queue and connection factory mentioned under Prerequisites above are located. Adapter Interface > Interface: Define from operation and schema (specified later) Operation Type: Consume Message Operation Name: Consume_message Consume Operation Parameters Destination Name: Press the Browse button, select Destination Type: Queues, then press Search. Wait for the list to populate, then select the entry for TestJMSQueue , which is the queue created in a previous example. JNDI Name: The JNDI name to use for the JMS connection. As in the previous example, this is probably the most common source of error. This is the JNDI name of the JMS adapter’s connection pool created in the WebLogic Server and which points to the connection factory. JDeveloper does not verify the value entered here. If you enter a wrong value, the JMS adapter won’t find the queue and you will get an error message at runtime, which is very difficult to trace. In our example, this is the value eis/wls/TestQueue . (See the earlier step on how to create a JMS Adapter Connection Pool in WebLogic Server for details.) Messages/Message SchemaURL: We will use the XSD file created during the previous example, in the JmsAdapterWriteSchema project to define the format for the incoming message payload and, at the same time, demonstrate how to import an existing XSD file into a JDeveloper project. Press the magnifying glass icon to search for schema files. In the Type Chooser, press the Import Schema File button. Select the magnifying glass next to URL to search for schema files. Navigate to the location of the JmsAdapterWriteSchema project > xsd and select the stringPayload.xsd file. Check the “Copy to Project” checkbox, press OK and confirm the following Localize Files popup. Now that the XSD file has been copied to the local project, it can be selected from the project’s schema files. Expand Project Schema Files > stringPayload.xsd and select exampleElement: string . Press Next and Finish, which will complete the JMS Adapter configuration.Save the project. Create a BPEL Component Drag a BPEL Process from the Component Palette (Service Components) to the Components section of the composite designer. Name it JmsAdapterReadSchema and select Template: Define Service Later and press OK. Wire the JMS Adapter to the BPEL Component Now wire the JMS adapter to the BPEL process, by dragging the arrow from the adapter to the BPEL process. A Transaction Properties popup will be displayed. Set the delivery mode to async.persist. This completes the steps at the composite level. 3 . Complete the BPEL Process Design Invoke the BPEL Flow via the JMS Adapter Open the BPEL component by double-clicking it in the design view of the composite.xml, or open it from the project navigator by selecting the JmsAdapterReadSchema.bpel file. This will display the BPEL process in the design view. You should see the JmsAdapterRead partner link in the left-hand swim lane. Drag a Receive activity onto the BPEL flow diagram, then drag a wire (left-hand yellow arrow) from it to the JMS adapter. This will open the Receive activity editor. Auto-generate the variable by pressing the green “+” button and check the “Create Instance” checkbox. This will result in a BPEL instance being created when a new JMS message is received. At this point it would actually be OK to compile and deploy the composite and it would pick up any messages from the JMS queue. In fact, you can do that to test it, if you like. But it is very rudimentary and would not be doing anything useful with the message. Also, you could only verify the actual message payload by looking at the instance’s flow in the Enterprise Manager. There are various other possibilities; we could pass the message to another web service, write it to a file using a file adapter or to a database via a database adapter etc. But these will all introduce unnecessary complications to our sample. So, to keep it simple, we will add a small piece of Java code to the BPEL process which will write the payload to standard output. This will be written to the server’s log file, which will be easy to monitor. Add a Java Embedding Activity First get the full name of the process’s input variable, as this will be needed for the Java code. Go to the Structure pane and expand Variables > Process > Variables. Then expand the input variable, for example, "Receive1_Consume_Message_InputVariable > body > ns2:exampleElement”, and note variable’s name and path, if they are different from this one. Drag a Java Embedding activity from the Component Palette (Oracle Extensions) to the BPEL flow, after the Receive activity, then open it to edit. Delete the example code and replace it with the following, replacing the variable parts with those in your sample, if necessary.: System.out.println("JmsAdapterReadSchema process picked up a message"); oracle.xml.parser.v2.XMLElement inputPayload =    (oracle.xml.parser.v2.XMLElement)getVariableData(                           "Receive1_Consume_Message_InputVariable",                           "body",                           "/ns2:exampleElement");   String inputString = inputPayload.getFirstChild().getNodeValue(); System.out.println("Input String is " + inputPayload.getFirstChild().getNodeValue()); Tip. If you are not sure of the exact syntax of the input variable, create an Assign activity in the BPEL process and copy the variable to another, temporary one. Then check the syntax created by the BPEL designer. This completes the BPEL process design in JDeveloper. Save, compile and deploy the process to the SOA server. 3. Test the Composite Shut Down the JmsAdapterReadSchema Composite After deploying the JmsAdapterReadSchema composite to the SOA server it is automatically activated. If there are already any messages in the queue, the adapter will begin polling them. To ease the testing process, we will deactivate the process first Log in to the Enterprise Manager (Fusion Middleware Control) and navigate to SOA > soa-infra (soa_server1) > default (or wherever you deployed your composite to) and click on JmsAdapterReadSchema [1.0] . Press the Shut Down button to disable the composite and confirm the following popup. Monitor Messages in the JMS Queue In a separate browser window, log in to the WebLogic Server Console and navigate to Services > Messaging > JMS Modules > TestJMSModule > TestJMSQueue > Monitoring. This is the location of the JMS queue we created in an earlier sample (see the prerequisites section of this sample). Check whether there are any messages already in the queue. If so, you can dequeue them using the QueueReceive Java program created in an earlier sample. This will ensure that the queue is empty and doesn’t contain any messages in the wrong format, which would cause the JmsAdapterReadSchema to fail. Send a Test Message In the Enterprise Manager, navigate to the JmsAdapterWriteSchema created earlier, press Test and send a test message, for example “Message from JmsAdapterWriteSchema”. Confirm that the message was written correctly to the queue by verifying it via the queue monitor in the WLS Console. Monitor the SOA Server’s Output A program deployed on the SOA server will write its standard output to the terminal window in which the server was started, unless this has been redirected to somewhere else, for example to a file. If it has not been redirected, go to the terminal session in which the server was started, otherwise open and monitor the file to which it was redirected. Re-Enable the JmsAdapterReadSchema Composite In the Enterprise Manager, navigate to the JmsAdapterReadSchema composite again and press Start Up to re-enable it. This should cause the JMS adapter to dequeue the test message and the following output should be written to the server’s standard output: JmsAdapterReadSchema process picked up a message. Input String is Message from JmsAdapterWriteSchema Note that you can also monitor the payload received by the process, by navigating to the the JmsAdapterReadSchema’s Instances tab in the Enterprise Manager. Then select the latest instance and view the flow of the BPEL component. The Receive activity will contain and display the dequeued message too. 4 . Troubleshooting This sample demonstrates how to dequeue an XML JMS message using a BPEL process and no additional functionality. For example, it doesn’t contain any error handling. Therefore, any errors in the payload will result in exceptions being written to the log file or standard output. If you get any errors related to the payload, such as Message handle error ... ORABPEL-09500 ... XPath expression failed to execute. An error occurs while processing the XPath expression; the expression is /ns2:exampleElement. ... etc. check that the variable used in the Java embedding part of the process was entered correctly. Possibly follow the tip mentioned in previous section. If this doesn’t help, you can delete the Java embedding part and simply verify the message via the flow diagram in the Enterprise Manager. Or use a different method, such as writing it to a file via a file adapter. This concludes this example. In the next post, we will begin with an AQ JMS example, which uses JMS to write to an Advanced Queue stored in the database. Best regards John-Brown Evans Oracle Technology Proactive Support Delivery

    Read the article

  • Routes for IIS Classic and Integrated Mode

    - by imran_ku07
         Introduction:             ASP.NET MVC Routing feature makes it very easy to provide clean URLs. You just need to configure routes in global.asax file to create an application with clean URLs. In most cases you define routes works in IIS 6, IIS 7 (or IIS 7.5) Classic and Integrated mode. But in some cases your routes may only works in IIS 7 Integrated mode, like in the case of using extension less URLs in IIS 6 without a wildcard extension map. So in this article I will show you how to create different routes which works in IIS 6 and IIS 7 Classic and Integrated mode.       Description:             Let's say that you need to create an application which must work both in Classic and Integrated mode. Also you have no control to setup a wildcard extension map in IIS. So you need to create two routes. One with extension less URL for Integrated mode and one with a URL with an extension for Classic Mode.   routes.MapRoute( "DefaultClassic", // Route name "{controller}.aspx/{action}/{id}", // URL with parameters new { controller = "Home", action = "Index", id = UrlParameter.Optional } // Parameter defaults ); routes.MapRoute( "DefaultIntegrated", // Route name "{controller}/{action}/{id}", // URL with parameters new { controller = "Home", action = "Index", id = UrlParameter.Optional } // Parameter defaults );               Now you have set up two routes, one for Integrated mode and one for Classic mode. Now you only need to ensure that Integrated mode route should only match if the application is running in Integrated mode and Classic mode route should only match if the application is running in Classic mode. For making this work you need to create two custom constraint for Integrated and Classic mode. So replace the above routes with these routes,     routes.MapRoute( "DefaultClassic", // Route name "{controller}.aspx/{action}/{id}", // URL with parameters new { controller = "Home", action = "Index", id = UrlParameter.Optional }, // Parameter defaults new { mode = new ClassicModeConstraint() }// Constraints ); routes.MapRoute( "DefaultIntegrated", // Route name "{controller}/{action}/{id}", // URL with parameters new { controller = "Home", action = "Index", id = UrlParameter.Optional }, // Parameter defaults new { mode = new IntegratedModeConstraint() }// Constraints );            The first route which is for Classic mode adds a ClassicModeConstraint and second route which is for Integrated mode adds a IntegratedModeConstraint. Next you need to add the implementation of these constraint classes.     public class ClassicModeConstraint : IRouteConstraint { public bool Match(HttpContextBase httpContext, Route route, string parameterName, RouteValueDictionary values, RouteDirection routeDirection) { return !HttpRuntime.UsingIntegratedPipeline; } } public class IntegratedModeConstraint : IRouteConstraint { public bool Match(HttpContextBase httpContext, Route route, string parameterName, RouteValueDictionary values, RouteDirection routeDirection) { return HttpRuntime.UsingIntegratedPipeline; } }             HttpRuntime.UsingIntegratedPipeline returns true if the application is running on Integrated mode; otherwise, it returns false. So routes for Integrated mode only matched when the application is running on Integrated mode and routes for Classic mode only matched when the application is not running on Integrated mode.       Summary:             During developing applications, sometimes developers are not sure that whether this application will be host on IIS 6 or IIS 7 (or IIS 7.5) Integrated mode or Classic mode. So it's a good idea to create separate routes for both Classic and Integrated mode so that your application will use extension less URLs where possible and use URLs with an extension where it is not possible to use extension less URLs. In this article I showed you how to create separate routes for IIS Integrated and Classic mode. Hope you will enjoy this article too.   SyntaxHighlighter.all()

    Read the article

  • value types in the vm

    - by john.rose
    value types in the vm p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times} p.p2 {margin: 0.0px 0.0px 14.0px 0.0px; font: 14.0px Times} p.p3 {margin: 0.0px 0.0px 12.0px 0.0px; font: 14.0px Times} p.p4 {margin: 0.0px 0.0px 15.0px 0.0px; font: 14.0px Times} p.p5 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Courier} p.p6 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Courier; min-height: 17.0px} p.p7 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times; min-height: 18.0px} p.p8 {margin: 0.0px 0.0px 0.0px 36.0px; text-indent: -36.0px; font: 14.0px Times; min-height: 18.0px} p.p9 {margin: 0.0px 0.0px 12.0px 0.0px; font: 14.0px Times; min-height: 18.0px} p.p10 {margin: 0.0px 0.0px 12.0px 0.0px; font: 14.0px Times; color: #000000} li.li1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times} li.li7 {margin: 0.0px 0.0px 0.0px 0.0px; font: 14.0px Times; min-height: 18.0px} span.s1 {font: 14.0px Courier} span.s2 {color: #000000} span.s3 {font: 14.0px Courier; color: #000000} ol.ol1 {list-style-type: decimal} Or, enduring values for a changing world. Introduction A value type is a data type which, generally speaking, is designed for being passed by value in and out of methods, and stored by value in data structures. The only value types which the Java language directly supports are the eight primitive types. Java indirectly and approximately supports value types, if they are implemented in terms of classes. For example, both Integer and String may be viewed as value types, especially if their usage is restricted to avoid operations appropriate to Object. In this note, we propose a definition of value types in terms of a design pattern for Java classes, accompanied by a set of usage restrictions. We also sketch the relation of such value types to tuple types (which are a JVM-level notion), and point out JVM optimizations that can apply to value types. This note is a thought experiment to extend the JVM’s performance model in support of value types. The demonstration has two phases.  Initially the extension can simply use design patterns, within the current bytecode architecture, and in today’s Java language. But if the performance model is to be realized in practice, it will probably require new JVM bytecode features, changes to the Java language, or both.  We will look at a few possibilities for these new features. An Axiom of Value In the context of the JVM, a value type is a data type equipped with construction, assignment, and equality operations, and a set of typed components, such that, whenever two variables of the value type produce equal corresponding values for their components, the values of the two variables cannot be distinguished by any JVM operation. Here are some corollaries: A value type is immutable, since otherwise a copy could be constructed and the original could be modified in one of its components, allowing the copies to be distinguished. Changing the component of a value type requires construction of a new value. The equals and hashCode operations are strictly component-wise. If a value type is represented by a JVM reference, that reference cannot be successfully synchronized on, and cannot be usefully compared for reference equality. A value type can be viewed in terms of what it doesn’t do. We can say that a value type omits all value-unsafe operations, which could violate the constraints on value types.  These operations, which are ordinarily allowed for Java object types, are pointer equality comparison (the acmp instruction), synchronization (the monitor instructions), all the wait and notify methods of class Object, and non-trivial finalize methods. The clone method is also value-unsafe, although for value types it could be treated as the identity function. Finally, and most importantly, any side effect on an object (however visible) also counts as an value-unsafe operation. A value type may have methods, but such methods must not change the components of the value. It is reasonable and useful to define methods like toString, equals, and hashCode on value types, and also methods which are specifically valuable to users of the value type. Representations of Value Value types have two natural representations in the JVM, unboxed and boxed. An unboxed value consists of the components, as simple variables. For example, the complex number x=(1+2i), in rectangular coordinate form, may be represented in unboxed form by the following pair of variables: /*Complex x = Complex.valueOf(1.0, 2.0):*/ double x_re = 1.0, x_im = 2.0; These variables might be locals, parameters, or fields. Their association as components of a single value is not defined to the JVM. Here is a sample computation which computes the norm of the difference between two complex numbers: double distance(/*Complex x:*/ double x_re, double x_im,         /*Complex y:*/ double y_re, double y_im) {     /*Complex z = x.minus(y):*/     double z_re = x_re - y_re, z_im = x_im - y_im;     /*return z.abs():*/     return Math.sqrt(z_re*z_re + z_im*z_im); } A boxed representation groups component values under a single object reference. The reference is to a ‘wrapper class’ that carries the component values in its fields. (A primitive type can naturally be equated with a trivial value type with just one component of that type. In that view, the wrapper class Integer can serve as a boxed representation of value type int.) The unboxed representation of complex numbers is practical for many uses, but it fails to cover several major use cases: return values, array elements, and generic APIs. The two components of a complex number cannot be directly returned from a Java function, since Java does not support multiple return values. The same story applies to array elements: Java has no ’array of structs’ feature. (Double-length arrays are a possible workaround for complex numbers, but not for value types with heterogeneous components.) By generic APIs I mean both those which use generic types, like Arrays.asList and those which have special case support for primitive types, like String.valueOf and PrintStream.println. Those APIs do not support unboxed values, and offer some problems to boxed values. Any ’real’ JVM type should have a story for returns, arrays, and API interoperability. The basic problem here is that value types fall between primitive types and object types. Value types are clearly more complex than primitive types, and object types are slightly too complicated. Objects are a little bit dangerous to use as value carriers, since object references can be compared for pointer equality, and can be synchronized on. Also, as many Java programmers have observed, there is often a performance cost to using wrapper objects, even on modern JVMs. Even so, wrapper classes are a good starting point for talking about value types. If there were a set of structural rules and restrictions which would prevent value-unsafe operations on value types, wrapper classes would provide a good notation for defining value types. This note attempts to define such rules and restrictions. Let’s Start Coding Now it is time to look at some real code. Here is a definition, written in Java, of a complex number value type. @ValueSafe public final class Complex implements java.io.Serializable {     // immutable component structure:     public final double re, im;     private Complex(double re, double im) {         this.re = re; this.im = im;     }     // interoperability methods:     public String toString() { return "Complex("+re+","+im+")"; }     public List<Double> asList() { return Arrays.asList(re, im); }     public boolean equals(Complex c) {         return re == c.re && im == c.im;     }     public boolean equals(@ValueSafe Object x) {         return x instanceof Complex && equals((Complex) x);     }     public int hashCode() {         return 31*Double.valueOf(re).hashCode()                 + Double.valueOf(im).hashCode();     }     // factory methods:     public static Complex valueOf(double re, double im) {         return new Complex(re, im);     }     public Complex changeRe(double re2) { return valueOf(re2, im); }     public Complex changeIm(double im2) { return valueOf(re, im2); }     public static Complex cast(@ValueSafe Object x) {         return x == null ? ZERO : (Complex) x;     }     // utility methods and constants:     public Complex plus(Complex c)  { return new Complex(re+c.re, im+c.im); }     public Complex minus(Complex c) { return new Complex(re-c.re, im-c.im); }     public double abs() { return Math.sqrt(re*re + im*im); }     public static final Complex PI = valueOf(Math.PI, 0.0);     public static final Complex ZERO = valueOf(0.0, 0.0); } This is not a minimal definition, because it includes some utility methods and other optional parts.  The essential elements are as follows: The class is marked as a value type with an annotation. The class is final, because it does not make sense to create subclasses of value types. The fields of the class are all non-private and final.  (I.e., the type is immutable and structurally transparent.) From the supertype Object, all public non-final methods are overridden. The constructor is private. Beyond these bare essentials, we can observe the following features in this example, which are likely to be typical of all value types: One or more factory methods are responsible for value creation, including a component-wise valueOf method. There are utility methods for complex arithmetic and instance creation, such as plus and changeIm. There are static utility constants, such as PI. The type is serializable, using the default mechanisms. There are methods for converting to and from dynamically typed references, such as asList and cast. The Rules In order to use value types properly, the programmer must avoid value-unsafe operations.  A helpful Java compiler should issue errors (or at least warnings) for code which provably applies value-unsafe operations, and should issue warnings for code which might be correct but does not provably avoid value-unsafe operations.  No such compilers exist today, but to simplify our account here, we will pretend that they do exist. A value-safe type is any class, interface, or type parameter marked with the @ValueSafe annotation, or any subtype of a value-safe type.  If a value-safe class is marked final, it is in fact a value type.  All other value-safe classes must be abstract.  The non-static fields of a value class must be non-public and final, and all its constructors must be private. Under the above rules, a standard interface could be helpful to define value types like Complex.  Here is an example: @ValueSafe public interface ValueType extends java.io.Serializable {     // All methods listed here must get redefined.     // Definitions must be value-safe, which means     // they may depend on component values only.     List<? extends Object> asList();     int hashCode();     boolean equals(@ValueSafe Object c);     String toString(); } //@ValueSafe inherited from supertype: public final class Complex implements ValueType { … The main advantage of such a conventional interface is that (unlike an annotation) it is reified in the runtime type system.  It could appear as an element type or parameter bound, for facilities which are designed to work on value types only.  More broadly, it might assist the JVM to perform dynamic enforcement of the rules for value types. Besides types, the annotation @ValueSafe can mark fields, parameters, local variables, and methods.  (This is redundant when the type is also value-safe, but may be useful when the type is Object or another supertype of a value type.)  Working forward from these annotations, an expression E is defined as value-safe if it satisfies one or more of the following: The type of E is a value-safe type. E names a field, parameter, or local variable whose declaration is marked @ValueSafe. E is a call to a method whose declaration is marked @ValueSafe. E is an assignment to a value-safe variable, field reference, or array reference. E is a cast to a value-safe type from a value-safe expression. E is a conditional expression E0 ? E1 : E2, and both E1 and E2 are value-safe. Assignments to value-safe expressions and initializations of value-safe names must take their values from value-safe expressions. A value-safe expression may not be the subject of a value-unsafe operation.  In particular, it cannot be synchronized on, nor can it be compared with the “==” operator, not even with a null or with another value-safe type. In a program where all of these rules are followed, no value-type value will be subject to a value-unsafe operation.  Thus, the prime axiom of value types will be satisfied, that no two value type will be distinguishable as long as their component values are equal. More Code To illustrate these rules, here are some usage examples for Complex: Complex pi = Complex.valueOf(Math.PI, 0); Complex zero = pi.changeRe(0);  //zero = pi; zero.re = 0; ValueType vtype = pi; @SuppressWarnings("value-unsafe")   Object obj = pi; @ValueSafe Object obj2 = pi; obj2 = new Object();  // ok List<Complex> clist = new ArrayList<Complex>(); clist.add(pi);  // (ok assuming List.add param is @ValueSafe) List<ValueType> vlist = new ArrayList<ValueType>(); vlist.add(pi);  // (ok) List<Object> olist = new ArrayList<Object>(); olist.add(pi);  // warning: "value-unsafe" boolean z = pi.equals(zero); boolean z1 = (pi == zero);  // error: reference comparison on value type boolean z2 = (pi == null);  // error: reference comparison on value type boolean z3 = (pi == obj2);  // error: reference comparison on value type synchronized (pi) { }  // error: synch of value, unpredictable result synchronized (obj2) { }  // unpredictable result Complex qq = pi; qq = null;  // possible NPE; warning: “null-unsafe" qq = (Complex) obj;  // warning: “null-unsafe" qq = Complex.cast(obj);  // OK @SuppressWarnings("null-unsafe")   Complex empty = null;  // possible NPE qq = empty;  // possible NPE (null pollution) The Payoffs It follows from this that either the JVM or the java compiler can replace boxed value-type values with unboxed ones, without affecting normal computations.  Fields and variables of value types can be split into their unboxed components.  Non-static methods on value types can be transformed into static methods which take the components as value parameters. Some common questions arise around this point in any discussion of value types. Why burden the programmer with all these extra rules?  Why not detect programs automagically and perform unboxing transparently?  The answer is that it is easy to break the rules accidently unless they are agreed to by the programmer and enforced.  Automatic unboxing optimizations are tantalizing but (so far) unreachable ideal.  In the current state of the art, it is possible exhibit benchmarks in which automatic unboxing provides the desired effects, but it is not possible to provide a JVM with a performance model that assures the programmer when unboxing will occur.  This is why I’m writing this note, to enlist help from, and provide assurances to, the programmer.  Basically, I’m shooting for a good set of user-supplied “pragmas” to frame the desired optimization. Again, the important thing is that the unboxing must be done reliably, or else programmers will have no reason to work with the extra complexity of the value-safety rules.  There must be a reasonably stable performance model, wherein using a value type has approximately the same performance characteristics as writing the unboxed components as separate Java variables. There are some rough corners to the present scheme.  Since Java fields and array elements are initialized to null, value-type computations which incorporate uninitialized variables can produce null pointer exceptions.  One workaround for this is to require such variables to be null-tested, and the result replaced with a suitable all-zero value of the value type.  That is what the “cast” method does above. Generically typed APIs like List<T> will continue to manipulate boxed values always, at least until we figure out how to do reification of generic type instances.  Use of such APIs will elicit warnings until their type parameters (and/or relevant members) are annotated or typed as value-safe.  Retrofitting List<T> is likely to expose flaws in the present scheme, which we will need to engineer around.  Here are a couple of first approaches: public interface java.util.List<@ValueSafe T> extends Collection<T> { … public interface java.util.List<T extends Object|ValueType> extends Collection<T> { … (The second approach would require disjunctive types, in which value-safety is “contagious” from the constituent types.) With more transformations, the return value types of methods can also be unboxed.  This may require significant bytecode-level transformations, and would work best in the presence of a bytecode representation for multiple value groups, which I have proposed elsewhere under the title “Tuples in the VM”. But for starters, the JVM can apply this transformation under the covers, to internally compiled methods.  This would give a way to express multiple return values and structured return values, which is a significant pain-point for Java programmers, especially those who work with low-level structure types favored by modern vector and graphics processors.  The lack of multiple return values has a strong distorting effect on many Java APIs. Even if the JVM fails to unbox a value, there is still potential benefit to the value type.  Clustered computing systems something have copy operations (serialization or something similar) which apply implicitly to command operands.  When copying JVM objects, it is extremely helpful to know when an object’s identity is important or not.  If an object reference is a copied operand, the system may have to create a proxy handle which points back to the original object, so that side effects are visible.  Proxies must be managed carefully, and this can be expensive.  On the other hand, value types are exactly those types which a JVM can “copy and forget” with no downside. Array types are crucial to bulk data interfaces.  (As data sizes and rates increase, bulk data becomes more important than scalar data, so arrays are definitely accompanying us into the future of computing.)  Value types are very helpful for adding structure to bulk data, so a successful value type mechanism will make it easier for us to express richer forms of bulk data. Unboxing arrays (i.e., arrays containing unboxed values) will provide better cache and memory density, and more direct data movement within clustered or heterogeneous computing systems.  They require the deepest transformations, relative to today’s JVM.  There is an impedance mismatch between value-type arrays and Java’s covariant array typing, so compromises will need to be struck with existing Java semantics.  It is probably worth the effort, since arrays of unboxed value types are inherently more memory-efficient than standard Java arrays, which rely on dependent pointer chains. It may be sufficient to extend the “value-safe” concept to array declarations, and allow low-level transformations to change value-safe array declarations from the standard boxed form into an unboxed tuple-based form.  Such value-safe arrays would not be convertible to Object[] arrays.  Certain connection points, such as Arrays.copyOf and System.arraycopy might need additional input/output combinations, to allow smooth conversion between arrays with boxed and unboxed elements. Alternatively, the correct solution may have to wait until we have enough reification of generic types, and enough operator overloading, to enable an overhaul of Java arrays. Implicit Method Definitions The example of class Complex above may be unattractively complex.  I believe most or all of the elements of the example class are required by the logic of value types. If this is true, a programmer who writes a value type will have to write lots of error-prone boilerplate code.  On the other hand, I think nearly all of the code (except for the domain-specific parts like plus and minus) can be implicitly generated. Java has a rule for implicitly defining a class’s constructor, if no it defines no constructors explicitly.  Likewise, there are rules for providing default access modifiers for interface members.  Because of the highly regular structure of value types, it might be reasonable to perform similar implicit transformations on value types.  Here’s an example of a “highly implicit” definition of a complex number type: public class Complex implements ValueType {  // implicitly final     public double re, im;  // implicitly public final     //implicit methods are defined elementwise from te fields:     //  toString, asList, equals(2), hashCode, valueOf, cast     //optionally, explicit methods (plus, abs, etc.) would go here } In other words, with the right defaults, a simple value type definition can be a one-liner.  The observant reader will have noticed the similarities (and suitable differences) between the explicit methods above and the corresponding methods for List<T>. Another way to abbreviate such a class would be to make an annotation the primary trigger of the functionality, and to add the interface(s) implicitly: public @ValueType class Complex { … // implicitly final, implements ValueType (But to me it seems better to communicate the “magic” via an interface, even if it is rooted in an annotation.) Implicitly Defined Value Types So far we have been working with nominal value types, which is to say that the sequence of typed components is associated with a name and additional methods that convey the intention of the programmer.  A simple ordered pair of floating point numbers can be variously interpreted as (to name a few possibilities) a rectangular or polar complex number or Cartesian point.  The name and the methods convey the intended meaning. But what if we need a truly simple ordered pair of floating point numbers, without any further conceptual baggage?  Perhaps we are writing a method (like “divideAndRemainder”) which naturally returns a pair of numbers instead of a single number.  Wrapping the pair of numbers in a nominal type (like “QuotientAndRemainder”) makes as little sense as wrapping a single return value in a nominal type (like “Quotient”).  What we need here are structural value types commonly known as tuples. For the present discussion, let us assign a conventional, JVM-friendly name to tuples, roughly as follows: public class java.lang.tuple.$DD extends java.lang.tuple.Tuple {      double $1, $2; } Here the component names are fixed and all the required methods are defined implicitly.  The supertype is an abstract class which has suitable shared declarations.  The name itself mentions a JVM-style method parameter descriptor, which may be “cracked” to determine the number and types of the component fields. The odd thing about such a tuple type (and structural types in general) is it must be instantiated lazily, in response to linkage requests from one or more classes that need it.  The JVM and/or its class loaders must be prepared to spin a tuple type on demand, given a simple name reference, $xyz, where the xyz is cracked into a series of component types.  (Specifics of naming and name mangling need some tasteful engineering.) Tuples also seem to demand, even more than nominal types, some support from the language.  (This is probably because notations for non-nominal types work best as combinations of punctuation and type names, rather than named constructors like Function3 or Tuple2.)  At a minimum, languages with tuples usually (I think) have some sort of simple bracket notation for creating tuples, and a corresponding pattern-matching syntax (or “destructuring bind”) for taking tuples apart, at least when they are parameter lists.  Designing such a syntax is no simple thing, because it ought to play well with nominal value types, and also with pre-existing Java features, such as method parameter lists, implicit conversions, generic types, and reflection.  That is a task for another day. Other Use Cases Besides complex numbers and simple tuples there are many use cases for value types.  Many tuple-like types have natural value-type representations. These include rational numbers, point locations and pixel colors, and various kinds of dates and addresses. Other types have a variable-length ‘tail’ of internal values. The most common example of this is String, which is (mathematically) a sequence of UTF-16 character values. Similarly, bit vectors, multiple-precision numbers, and polynomials are composed of sequences of values. Such types include, in their representation, a reference to a variable-sized data structure (often an array) which (somehow) represents the sequence of values. The value type may also include ’header’ information. Variable-sized values often have a length distribution which favors short lengths. In that case, the design of the value type can make the first few values in the sequence be direct ’header’ fields of the value type. In the common case where the header is enough to represent the whole value, the tail can be a shared null value, or even just a null reference. Note that the tail need not be an immutable object, as long as the header type encapsulates it well enough. This is the case with String, where the tail is a mutable (but never mutated) character array. Field types and their order must be a globally visible part of the API.  The structure of the value type must be transparent enough to have a globally consistent unboxed representation, so that all callers and callees agree about the type and order of components  that appear as parameters, return types, and array elements.  This is a trade-off between efficiency and encapsulation, which is forced on us when we remove an indirection enjoyed by boxed representations.  A JVM-only transformation would not care about such visibility, but a bytecode transformation would need to take care that (say) the components of complex numbers would not get swapped after a redefinition of Complex and a partial recompile.  Perhaps constant pool references to value types need to declare the field order as assumed by each API user. This brings up the delicate status of private fields in a value type.  It must always be possible to load, store, and copy value types as coordinated groups, and the JVM performs those movements by moving individual scalar values between locals and stack.  If a component field is not public, what is to prevent hostile code from plucking it out of the tuple using a rogue aload or astore instruction?  Nothing but the verifier, so we may need to give it more smarts, so that it treats value types as inseparable groups of stack slots or locals (something like long or double). My initial thought was to make the fields always public, which would make the security problem moot.  But public is not always the right answer; consider the case of String, where the underlying mutable character array must be encapsulated to prevent security holes.  I believe we can win back both sides of the tradeoff, by training the verifier never to split up the components in an unboxed value.  Just as the verifier encapsulates the two halves of a 64-bit primitive, it can encapsulate the the header and body of an unboxed String, so that no code other than that of class String itself can take apart the values. Similar to String, we could build an efficient multi-precision decimal type along these lines: public final class DecimalValue extends ValueType {     protected final long header;     protected private final BigInteger digits;     public DecimalValue valueOf(int value, int scale) {         assert(scale >= 0);         return new DecimalValue(((long)value << 32) + scale, null);     }     public DecimalValue valueOf(long value, int scale) {         if (value == (int) value)             return valueOf((int)value, scale);         return new DecimalValue(-scale, new BigInteger(value));     } } Values of this type would be passed between methods as two machine words. Small values (those with a significand which fits into 32 bits) would be represented without any heap data at all, unless the DecimalValue itself were boxed. (Note the tension between encapsulation and unboxing in this case.  It would be better if the header and digits fields were private, but depending on where the unboxing information must “leak”, it is probably safer to make a public revelation of the internal structure.) Note that, although an array of Complex can be faked with a double-length array of double, there is no easy way to fake an array of unboxed DecimalValues.  (Either an array of boxed values or a transposed pair of homogeneous arrays would be reasonable fallbacks, in a current JVM.)  Getting the full benefit of unboxing and arrays will require some new JVM magic. Although the JVM emphasizes portability, system dependent code will benefit from using machine-level types larger than 64 bits.  For example, the back end of a linear algebra package might benefit from value types like Float4 which map to stock vector types.  This is probably only worthwhile if the unboxing arrays can be packed with such values. More Daydreams A more finely-divided design for dynamic enforcement of value safety could feature separate marker interfaces for each invariant.  An empty marker interface Unsynchronizable could cause suitable exceptions for monitor instructions on objects in marked classes.  More radically, a Interchangeable marker interface could cause JVM primitives that are sensitive to object identity to raise exceptions; the strangest result would be that the acmp instruction would have to be specified as raising an exception. @ValueSafe public interface ValueType extends java.io.Serializable,         Unsynchronizable, Interchangeable { … public class Complex implements ValueType {     // inherits Serializable, Unsynchronizable, Interchangeable, @ValueSafe     … It seems possible that Integer and the other wrapper types could be retro-fitted as value-safe types.  This is a major change, since wrapper objects would be unsynchronizable and their references interchangeable.  It is likely that code which violates value-safety for wrapper types exists but is uncommon.  It is less plausible to retro-fit String, since the prominent operation String.intern is often used with value-unsafe code. We should also reconsider the distinction between boxed and unboxed values in code.  The design presented above obscures that distinction.  As another thought experiment, we could imagine making a first class distinction in the type system between boxed and unboxed representations.  Since only primitive types are named with a lower-case initial letter, we could define that the capitalized version of a value type name always refers to the boxed representation, while the initial lower-case variant always refers to boxed.  For example: complex pi = complex.valueOf(Math.PI, 0); Complex boxPi = pi;  // convert to boxed myList.add(boxPi); complex z = myList.get(0);  // unbox Such a convention could perhaps absorb the current difference between int and Integer, double and Double. It might also allow the programmer to express a helpful distinction among array types. As said above, array types are crucial to bulk data interfaces, but are limited in the JVM.  Extending arrays beyond the present limitations is worth thinking about; for example, the Maxine JVM implementation has a hybrid object/array type.  Something like this which can also accommodate value type components seems worthwhile.  On the other hand, does it make sense for value types to contain short arrays?  And why should random-access arrays be the end of our design process, when bulk data is often sequentially accessed, and it might make sense to have heterogeneous streams of data as the natural “jumbo” data structure.  These considerations must wait for another day and another note. More Work It seems to me that a good sequence for introducing such value types would be as follows: Add the value-safety restrictions to an experimental version of javac. Code some sample applications with value types, including Complex and DecimalValue. Create an experimental JVM which internally unboxes value types but does not require new bytecodes to do so.  Ensure the feasibility of the performance model for the sample applications. Add tuple-like bytecodes (with or without generic type reification) to a major revision of the JVM, and teach the Java compiler to switch in the new bytecodes without code changes. A staggered roll-out like this would decouple language changes from bytecode changes, which is always a convenient thing. A similar investigation should be applied (concurrently) to array types.  In this case, it seems to me that the starting point is in the JVM: Add an experimental unboxing array data structure to a production JVM, perhaps along the lines of Maxine hybrids.  No bytecode or language support is required at first; everything can be done with encapsulated unsafe operations and/or method handles. Create an experimental JVM which internally unboxes value types but does not require new bytecodes to do so.  Ensure the feasibility of the performance model for the sample applications. Add tuple-like bytecodes (with or without generic type reification) to a major revision of the JVM, and teach the Java compiler to switch in the new bytecodes without code changes. That’s enough musing me for now.  Back to work!

    Read the article

  • Silverlight Cream for May 13, 2010 -- #861

    - by Dave Campbell
    In this Issue: Sigurd Snørteland, Jeff Prosise, DaveDev, Joe Zhou, Chris Eargle, John Papa(-2-, -3-), and David Anson(-2-). Shoutouts: In with the links I've listed below, Sigurd Snørteland also sent a link to this app he's working on which is actually pretty cool to see: ZuneLight. The code is not yet available. He also has a no-code demo of a Silverlight Media Center Pieter Voloshyn, Luiz Thadeu, and Jhun Iti have a very nice Silverlight image editor up: Thumba From SilverlightCream.com: WP7 - Silverlight on mobile Sigurd Snørteland submitted some links for me that have been translated to English from his blog. I hope the pages come out good because he's got a lot of good stuff on there. This one has a link to a presentation he did, and 4 projects you can load up in the emulator that he's converted to the phone: weather, worldclock, coverflow, and solitaire ... pretty cool... thanks for the links Sigurd! Understanding Page Orientation in Silverlight for Windows Phone Jeff Prosise has a really nice post up on page orientation in WP7 ... what it means to your app, how to detect it, and example code for what to do then... also love a quote by Jeff: "Silverlight for Windows Phone is the hottest thing since color TV" Why you should check out Expression Blend Behaviors when using Silverlight DaveDev has a post up describing Behaviors and why we should use them, plus tons of external links to resources, blogs, videos... all good stuff... Fiddler inspector for WCF Silverlight Polling Duplex and WCF RIA Joe Zhou announces and provides a link to a new Fiddler inspector that understands the framing in Polling Duplex and also raw binary xml and binary SOAP. Windows Phone Controls v0.7 Chris Eargle reports the release of Version 0.7 of the Windows Phone Controls project on CodePlex ... this includes a Pivot Control and a Panorama Control... both very nicely done. Binding to Silverlight ComboBox and Using SelectedValue, SelectedValuePath and DisplayMemberPath John Papa responds to a user question and put up a nice post about binding to a ComboBox and then go from the selected item to some other property ... code included No More Boxes! Exploring the PathListBox (Silverlight TV #25) Silverlight TV 25 went up on Tuesday ... thought it was going to be Thursday?? anyway ... John Papa and Adam Kinney are discussing the PathListBox and looking at some cool demos thereof. Exposing SOAP, OData, and JSON Endpoints for RIA Services (Silverlight TV 26) Since today IS Thursday, we have a new Silverlight TV, number 26, and John Papa is chatting with Deepesh Mohnani of the WCF RIA Services team about exposing all sorts of endpoints... should be something in there for everybody :) Workaround for a Silverlight data binding bug affecting various scenarios - including DataGrid+ContextMenu David Anson details the rabbit-trail he and others on the team followed in response to a problem reported via Twitter where the binding on a DataGrid seemed off by a row(!) ... weird but true, validated, and SL3/4 are bug-for-bug compatible with this too! ... But David wouldn't leave us there.. he also has a workaround. Sharing the code for a simple Silverlight 4 REST-based cloud-oriented file management app for Azure and S3 David Anson had an opportunity to build an app he's wanted to build for a while and shares it with us: Blobstore -- a small, lightweight Silverlight 4 application that acts as a basic front-end for the Windows Azure Simple Data Storage and the Amazon Simple Storage Service (S3) -- and remember I said he shared the source :) Stay in the 'Light! Twitter SilverlightNews | Twitter WynApse | WynApse.com | Tagged Posts | SilverlightCream Join me @ SilverlightCream | Phoenix Silverlight User Group Technorati Tags: Silverlight    Silverlight 3    Silverlight 4    Windows Phone MIX10

    Read the article

  • Adventures in MVVM &ndash; ViewModel Location and Creation

    - by Brian Genisio's House Of Bilz
    More Adventures in MVVM In this post, I am going to explore how I prefer to attach ViewModels to my Views.  I have published the code to my ViewModelSupport project on CodePlex in case you'd like to see how it works along with some examples.  Some History My approach to View-First ViewModel creation has evolved over time.  I have constructed ViewModels in code-behind.  I have instantiated ViewModels in the resources sectoin of the view. I have used Prism to resolve ViewModels via Dependency Injection. I have created attached properties that use Dependency Injection containers underneath.  Of all these approaches, I continue to find issues either in composability, blendability or maintainability.  Laurent Bugnion came up with a pretty good approach in MVVM Light Toolkit with his ViewModelLocator, but as John Papa points out, it has maintenance issues.  John paired up with Glen Block to make the ViewModelLocator more generic by using MEF to compose ViewModels.  It is a great approach, but I don’t like baking in specific resolution technologies into the ViewModelSupport project. I bring these people up, not to name drop, but to give them credit for the place I finally landed in my journey to resolve ViewModels.  I have come up with my own version of the ViewModelLocator that is both generic and container agnostic.  The solution is blendable, configurable and simple to use.  Use any resolution mechanism you want: MEF, Unity, Ninject, Activator.Create, Lookup Tables, new, whatever. How to use the locator 1. Create a class to contain your resolution configuration: public class YourViewModelResolver: IViewModelResolver { private YourFavoriteContainer container = new YourFavoriteContainer(); public YourViewModelResolver() { // Configure your container } public object Resolve(string viewModelName) { return container.Resolve(viewModelName); } } Examples of doing this are on CodePlex for MEF, Unity and Activator.CreateInstance. 2. Create your ViewModelLocator with your custom resolver in App.xaml: <VMS:ViewModelLocator x:Key="ViewModelLocator"> <VMS:ViewModelLocator.Resolver> <local:YourViewModelResolver /> </VMS:ViewModelLocator.Resolver> </VMS:ViewModelLocator> 3. Hook up your data context whenever you want a ViewModel (WPF): <Border DataContext="{Binding YourViewModelName, Source={StaticResource ViewModelLocator}}"> This example uses dynamic properties on the ViewModelLocator and passes the name to your resolver to figure out how to compose it. 4. What about Silverlight? Good question.  You can't bind to dynamic properties in Silverlight 4 (crossing my fingers for Silverlight 5), but you CAN use string indexing: <Border DataContext="{Binding [YourViewModelName], Source={StaticResource ViewModelLocator}}"> But, as John Papa points out in his article, there is a silly bug in Silverlight 4 (as of this writing) that will call into the indexer 6 times when it binds.  While this is little more than a nuisance when getting most properties, it can be much more of an issue when you are resolving ViewModels six times.  If this gets in your way, the solution (as pointed out by John), is to use an IndexConverter (instantiated in App.xaml and also included in the project): <Border DataContext="{Binding Source={StaticResource ViewModelLocator}, Converter={StaticResource IndexConverter}, ConverterParameter=YourViewModelName}"> It is a bit uglier than the WPF version (this method will also work in WPF if you prefer), but it is still not all that bad.  Conclusion This approach works really well (I suppose I am a bit biased).  It allows for composability from any mechanisim you choose.  It is blendable (consider serving up different objects in Design Mode if you wish... or different constructors… whatever makes sense to you).  It works in Cider.  It is configurable.  It is flexible.  It is the best way I have found to manage View-First ViewModel hookups.  Thanks to the guys mentioned in this article for getting me to something I love using.  Enjoy.

    Read the article

  • Silverlight Cream for March 10, 2011 -- #1058

    - by Dave Campbell
    In this Issue: Ian T. Lackey, Peter Kuhn, WindowsPhoneGeek(-2-), Jesse Liberty(-2-), Martin Krüger, John Papa, Jeremy Likness, Karl Shifflett, and Colin Eberhardt. Above the Fold: Silverlight: "Silverlight TV 65: 3D Graphics" John Papa WP7: "Developing a Windows Phone 7 Jump List Control" Colin Eberhardt Shoutouts: Telerik announced a special sale on their RadControls for WP7... check it out: RadControls for Windows Phone 7 - on Sale from March 16th at a Special Promo Price! From SilverlightCream.com: Prism BootStrapper Load ModuleCatalog Ansyc Ian T. Lackey has a post up about reading the module catalog for Prism from an XML file asynchronously... fun stuff... this is how we kick-started our app... XNA for Silverlight developers: Part 6 - Input (accelerometer) Peter Kuhn has Part 6 of his XNA for Silverlight devs up at SilverlightShow. This post is on the use of the accelerometer... some great diagrams and explanations of it's use along with some code to play with... including a 'problems and pitfalls' section, and some good external links. Getting Started with Unit Testing in Silverlight for WP7 WindowsPhoneGeek has an introduction to Unit Testing in general, and then moves into Unit Testing in Silverlight for WP7, providing 3 options with links to the materials and code demonstrating the concepts. Using DockPanel in WP7 Responding to reader's questions, WindowsPhoneGeek's next post is on the DockPanel from the Silverlight Toolkit, and using it in WP7... defined declaratively and in code. Reactive Extensions–More About Chaining Jesse Liberty has post number 10 on Rx up and is a follow-on to the last one on Chaining. This time he exercises the chaining aspect of SelectMany. Yet Another Podcast #26–Walt Ritscher In his next post, Jesse Liberty has his 26th 'Yet Another Podcast' up and is chatting with my friend Walt Ritscher. If you don't know who Walt is, check out the links Jesse has on the post... I'm sure you've crossed paths. How to: Create A half square from a regular polygon (triangle) Martin Krüger demonstrates the exact placement of a half-square (isosceles right triangle), formed with a regular polygon in Blend... this is much more involved than I've made it sound... check out his post. Silverlight TV 65: 3D Graphics John Papa has Silverlight TV number 65 up and it's all about the 3D graphics stuff we saw at the Firestarter. John is talking with Danny Riddel, the CEO of Archetype, the company that built the awesome 3D demo we all gushed over. Jounce Part 12: Providing History-Based Back Navigation Jeremy Likness has part 12 of his Jounce exploration up... and discussing the stack of navigated pages that Jounce retains and providing a 'go back' functionality... and provides a good example of using it all. Prism 4 Region Navigation with Silverlight Frame Navigation and Unity Karl Shifflett has a post for all us Prism afficianados... Prism, Unity, and the Silverlight Frame Navigation framework. Some great external links for 'required reading' too. Developing a Windows Phone 7 Jump List Control Colin Eberhardt has an awesome tutorial up for creating a JumpList control for WP7... what a bunch of effort... this is a step-by-step description of designing the control he built and blogged about a while back... and it's still cool! Stay in the 'Light! Twitter SilverlightNews | Twitter WynApse | WynApse.com | Tagged Posts | SilverlightCream Join me @ SilverlightCream | Phoenix Silverlight User Group Technorati Tags: Silverlight    Silverlight 3    Silverlight 4    Windows Phone MIX10

    Read the article

  • Three Key Tenets of Optimal Social Collaboration

    - by kellsey.ruppel
    Today's blog post comes to us from John Bruswick! This post is an abridged version of John’s white paper in which he discusses three principals to optimize social collaboration within an enterprise.   By john[email protected], Oracle Principal Sales Consultant Effective social collaboration is actionable, deeply contextual and inherently derives its value from business entities outside of itself. How does an organization begin the journey from traditional, siloed collaboration to natural, business entity based social collaboration? Successful enablement of enterprise social collaboration requires that organizations embrace the following tenets and understand that traditional collaborative functionality has inherent limits - it is innovation and integration in accordance with the following tenets that will provide net-new efficiency benefits. Key Tenets of Optimal Social Collaboration Leverage a Ubiquitous Social Fabric - Collaborative activities should be supported through a ubiquitous social fabric, providing a personalized experience, broadcasting key business events and connecting people and business processes.  This supports education of participants working in and around a specific business entity that will benefit from an implicit capture of tacit knowledge and provide continuity between participants.  In the absence of this ubiquitous platform activities can still occur but are essentially siloed causing frequent duplication of effort across similar tasks, with critical tacit knowledge eluding capture. Supply Continuous Context to Support Decision Making and Problem Solving - People generally engage in collaborative behavior to obtain a decision or the resolution for a specific issue.  The time to achieve resolution is referred to as "Solve Time".  Users have traditionally been forced to switch or "alt-tab" between business systems and synthesize their own context across disparate systems and processes.  The constant loss of context forces end users to exert a large amount of effort that could be spent on higher value problem solving. Extend the Collaborative Lifecycle into Back Office - Beyond the solve time from decision making efforts, additional time is expended formalizing the resolution that was generated from collaboration in a system of record.  Extending collaboration to result in the capture of an explicit decision maximizes efficiencies, creating a closed circuit for a particular thread.  This type of structured action may exist today within your organization's customer support system around opening, solving and closing support issues, but generally does not extend to Sales focused collaborative activities. Excelling in the Unstructured Future We will always have to deal with unstructured collaborative processes within our organizations.  Regardless of the participants and nature of the collaborate process, two things are certain – the origination and end points are generally known and relate to a business entity, perhaps a customer, opportunity, order, shipping location, product or otherwise. Imagine the benefits if an organization's key business systems supported a social fabric, provided continuous context and extended the lifecycle around the collaborative decision making to include output into back office systems of record.   The technical hurdle to embracing optimal social collaboration would fall away, leaving the company with an opportunity to focus on and refine how processes were approached.  Time and resources previously required could then be reallocated to focusing on innovation to support competitive differentiation unique to your business. How can you achieve optimal social collaboration? Oracle Social Network enables business users to collaborate with each other using a broad range of collaboration styles and integrates data from a variety of sources and business applications -- allowing you to achieve optimal social collaboration. Looking to learn more? Read John's white paper, where he discusses in further detail the three principals to optimize social collaboration within an enterprise. 

    Read the article

  • Windows Azure Recipe: Consumer Portal

    - by Clint Edmonson
    Nearly every company on the internet has a web presence. Many are merely using theirs for informational purposes. More sophisticated portals allow customers to register their contact information and provide some level of interaction or customer support. But as our understanding of how consumers use the web increases, the more progressive companies are taking advantage of social web and rich media delivery to connect at a deeper level with the consumers of their goods and services. Drivers Cost reduction Scalability Global distribution Time to market Solution Here’s a sketch of how a Windows Azure Consumer Portal might be built out: Ingredients Web Role – this will host the core of the solution. Each web role is a virtual machine hosting an application written in ASP.NET (or optionally php, or node.js). The number of web roles can be scaled up or down as needed to handle peak and non-peak traffic loads. Database – every modern web application needs to store data. SQL Azure databases look and act exactly like their on-premise siblings but are fault tolerant and have data redundancy built in. Access Control (optional) – if identity needs to be tracked within the solution, the access control service combined with the Windows Identity Foundation framework provides out-of-the-box support for several social media platforms including Windows LiveID, Google, Yahoo!, Facebook. It also has a provider model to allow integration with other platforms as well. Caching (optional) – for sites with high traffic with lots of read-only data and lists, the distributed in-memory caching service can be used to cache and serve up static data at higher scale and speed than direct database requests. It can also be used to manage user session state. Blob Storage (optional) – for sites that serve up unstructured data such as documents, video, audio, device drivers, and more. The data is highly available and stored redundantly across data centers. Each entry in blob storage is provided with it’s own unique URL for direct access by the browser. Content Delivery Network (CDN) (optional) – for sites that service users around the globe, the CDN is an extension to blob storage that, when enabled, will automatically cache frequently accessed blobs and static site content at edge data centers around the world. The data can be delivered statically or streamed in the case of rich media content. Training Labs These links point to online Windows Azure training labs where you can learn more about the individual ingredients described above. (Note: The entire Windows Azure Training Kit can also be downloaded for offline use.) Windows Azure (16 labs) Windows Azure is an internet-scale cloud computing and services platform hosted in Microsoft data centers, which provides an operating system and a set of developer services which can be used individually or together. It gives developers the choice to build web applications; applications running on connected devices, PCs, or servers; or hybrid solutions offering the best of both worlds. New or enhanced applications can be built using existing skills with the Visual Studio development environment and the .NET Framework. With its standards-based and interoperable approach, the services platform supports multiple internet protocols, including HTTP, REST, SOAP, and plain XML SQL Azure (7 labs) Microsoft SQL Azure delivers on the Microsoft Data Platform vision of extending the SQL Server capabilities to the cloud as web-based services, enabling you to store structured, semi-structured, and unstructured data. Windows Azure Services (9 labs) As applications collaborate across organizational boundaries, ensuring secure transactions across disparate security domains is crucial but difficult to implement. Windows Azure Services provides hosted authentication and access control using powerful, secure, standards-based infrastructure. See my Windows Azure Resource Guide for more guidance on how to get started, including links web portals, training kits, samples, and blogs related to Windows Azure.

    Read the article

  • RHEL - NFS4: Mounted/Exported as rw, user write permission denied

    - by brendanmac
    Hello, I have nfs4 configured between a RHEL 5.3 server (charlie) and a RHEL 5.4 client (simcom1). The machines are configured to authenticate users via kerberos by a Windows Server 2008 active directory machine called "alpha." Alpha also serves as a dns and dhcp machine for the local network. I notice that when a user logs in to a RHEL machine for the first time they are issued a unique uid to that machine; The first user to log on gets 10001. So, what I see is that users between simcom1 and charlie have different UIDs. When a user does an 'ls -la' command from within an nfs4 mount I would have thought that the usernames in the owner column would indicate 'nobody' or at least the wrong user name - since UIDs are different between the machines for each user, and not all users have logged into each machine. However, the simcom1 is able to resolve usernames in an 'ls -la' executed on files residing on charlie via nfs4 correctly. Most troubling is that users are unable to write to files across the nfs mount. The server, charlie, has the root directory exported as rw. The client, simcom1, mounts the export as rw. My configurations are shown below. My question is, how do I configure the RHEL machines to allow users to write files across nfs4 that is already mounted as read/write? [root@charlie ~]# more /etc/exports / 10.100.0.0/16(rw,no_root_squash,fsid=0) [root@charlie ~]#cat /etc/sysconfig/nfs # # Define which protocol versions mountd # will advertise. The values are "no" or "yes" # with yes being the default #MOUNTD_NFS_V1="no" #MOUNTD_NFS_V2="no" #MOUNTD_NFS_V3="no" # # # Path to remote quota server. See rquotad(8) #RQUOTAD="/usr/sbin/rpc.rquotad" # Port rquotad should listen on. #RQUOTAD_PORT=875 # Optinal options passed to rquotad #RPCRQUOTADOPTS="" # # # TCP port rpc.lockd should listen on. #LOCKD_TCPPORT=32803 # UDP port rpc.lockd should listen on. #LOCKD_UDPPORT=32769 # # # Optional arguments passed to rpc.nfsd. See rpc.nfsd(8) # Turn off v2 and v3 protocol support #RPCNFSDARGS="-N 2 -N 3" # Turn off v4 protocol support #RPCNFSDARGS="-N 4" # Number of nfs server processes to be started. # The default is 8. RPCNFSDCOUNT=8 # Stop the nfsd module from being pre-loaded #NFSD_MODULE="noload" # # # Optional arguments passed to rpc.mountd. See rpc.mountd(8) #STATDARG="" #RPCMOUNTDOPTS="" # Port rpc.mountd should listen on. #MOUNTD_PORT=892 # # # Optional arguments passed to rpc.statd. See rpc.statd(8) #RPCIDMAPDARGS="" # # Set to turn on Secure NFS mounts. SECURE_NFS="no" # Optional arguments passed to rpc.gssd. See rpc.gssd(8) #RPCGSSDARGS="-vvv" # Optional arguments passed to rpc.svcgssd. See rpc.svcgssd(8) #RPCSVCGSSDARGS="-vvv" # Don't load security modules in to the kernel #SECURE_NFS_MODS="noload" # # Don't load sunrpc module. #RPCMTAB="noload" # [root@simcom1 ~]# cat /etc/fstab --start snip-- charlie:/home /usr/local/dev/charlie nfs4 rw,nosuid, 0 0 --end snip-- [brendanmac@simcom1 /usr/local/dev/charlie/brendanmac]# touch file touch: cannot touch 'file': Permission denied [brendanmac@simcom1 /usr/local/dev/charlie/brendanmac]# su Password: [root@simcom1 /usr/local/dev/charlie/brendanmac]# touch file [root@simcom1 /usr/local/dev/charlie/brendanmac]# ls -la file -rw------- 1 root root 0 May 26 10:43 file Thank you for your assistance, Brendan

    Read the article

  • beginner's ruby question: how to use erb to output file after binding

    - by john
    Hi, I got the following example: require 'erb' names = [] names.push( { 'first' => "Jack", 'last' => "Herrington" } ) names.push( { 'first' => "LoriLi", 'last' => "Herrington" } ) names.push( { 'first' => "Megan", 'last' => "Herrington" } ) myname = "John Smith" File.open( ARGV[0] ) { |fh| erb = ERB.new( fh.read ) print erb.result( binding ) accompanied by text.txt <% name = "Jack" %> Hello <%= name %> <% names.each { |name| %> Hello <%= name[ 'first' ] %> <%= name[ 'last' ] %> <% } %> hi, my name is <%= myname %> } it prints nicely to screen. what is the simplest way to output another file: "text2.txt"? thank you!!!

    Read the article

< Previous Page | 32 33 34 35 36 37 38 39 40 41 42 43  | Next Page >