Search Results

Search found 9178 results on 368 pages for 'nth element'.

Page 105/368 | < Previous Page | 101 102 103 104 105 106 107 108 109 110 111 112  | Next Page >

  • Unexpected output on initializing array by using both `element-by-element` & `designated` technique

    - by haccks
    C99 provides a feature to initialize arrays by using both element-by-element & designated method together as: int a[] = {2,1,[3] = 5,[5] = 9,6,[8] = 4}; On running the code: #include <stdio.h> int main() { int a[] = {2,1,[3] = 5,[0] = 9,4,[6] = 25}; for(int i = 0; i < sizeof(a)/sizeof(a[0]); i++) printf("%d ",a[i]); return 0; } (Note that Element 0 is initialized to 2 and then again initialised by designator [0] to 9) I was expecting that element 0(which is 2) will be replaced by 9(as designator [0] = 9) and hence o/p will become 9 1 0 5 4 0 25 Unfortunately I was wrong as o/p came; 9 4 0 5 0 0 25 Any explanation for unexpected o/p?

    Read the article

  • Howto create own Jquery Like Function in Javascript?

    - by streetparade
    How can i create a function which looks like the jquery callback $ Say i want to call a element with id= "mydiv". i want to be able to call it like var div = $("mydiv").value; i think the function should look like function $(element) { return document.getElementById(element); } Is that the right way to do it, or do you prefer another way to solve it?

    Read the article

  • CSS 100% height with padding/margin

    - by Toji
    This has been driving me crazy for a couple of days now, but in reality it's a problem that I've hit off and on for the last few years: With HTML/CSS how can I make an element that has a width and/or height that is 100% of it's parent element and still has proper padding or margins? By "proper" I mean that if my parent element is 200px tall and I specify 100% height with 5px padding I would expect that I should get a 190px high element with 5px "border" on all sides, nicely centered in the parent element. Now, I know that that's not how the standard box model specifies it should work (although I'd like to know why, exactly...), so the obvious answer doesn't work: #myDiv { width: 100% height: 100%; padding: 5px; } But it would seem to me that there must be SOME way of reliably producing this effect for a parent of arbitrary size. Does anyone know of a way of accomplishing this (seemingly simple) task? Oh, and for the record I'm not terribly interested in IE compatibility so that should (hopefully) make things a bit easier. EDIT: Since an example was asked for, here's the simplest one I can think of: <html style="height: 100%"> <body style="height: 100%"> <div style="background-color: black; height: 100%; padding: 25px"></div> </body> </html> The challenge is then to get the black box to show up with a 25 pixel padding on all edges without the page growing big enough to require scrollbars.

    Read the article

  • An Xml Serializable PropertyBag Dictionary Class for .NET

    - by Rick Strahl
    I don't know about you but I frequently need property bags in my applications to store and possibly cache arbitrary data. Dictionary<T,V> works well for this although I always seem to be hunting for a more specific generic type that provides a string key based dictionary. There's string dictionary, but it only works with strings. There's Hashset<T> but it uses the actual values as keys. In most key value pair situations for me string is key value to work off. Dictionary<T,V> works well enough, but there are some issues with serialization of dictionaries in .NET. The .NET framework doesn't do well serializing IDictionary objects out of the box. The XmlSerializer doesn't support serialization of IDictionary via it's default serialization, and while the DataContractSerializer does support IDictionary serialization it produces some pretty atrocious XML. What doesn't work? First off Dictionary serialization with the Xml Serializer doesn't work so the following fails: [TestMethod] public void DictionaryXmlSerializerTest() { var bag = new Dictionary<string, object>(); bag.Add("key", "Value"); bag.Add("Key2", 100.10M); bag.Add("Key3", Guid.NewGuid()); bag.Add("Key4", DateTime.Now); bag.Add("Key5", true); bag.Add("Key7", new byte[3] { 42, 45, 66 }); TestContext.WriteLine(this.ToXml(bag)); } public string ToXml(object obj) { if (obj == null) return null; StringWriter sw = new StringWriter(); XmlSerializer ser = new XmlSerializer(obj.GetType()); ser.Serialize(sw, obj); return sw.ToString(); } The error you get with this is: System.NotSupportedException: The type System.Collections.Generic.Dictionary`2[[System.String, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089],[System.Object, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089]] is not supported because it implements IDictionary. Got it! BTW, the same is true with binary serialization. Running the same code above against the DataContractSerializer does work: [TestMethod] public void DictionaryDataContextSerializerTest() { var bag = new Dictionary<string, object>(); bag.Add("key", "Value"); bag.Add("Key2", 100.10M); bag.Add("Key3", Guid.NewGuid()); bag.Add("Key4", DateTime.Now); bag.Add("Key5", true); bag.Add("Key7", new byte[3] { 42, 45, 66 }); TestContext.WriteLine(this.ToXmlDcs(bag)); } public string ToXmlDcs(object value, bool throwExceptions = false) { var ser = new DataContractSerializer(value.GetType(), null, int.MaxValue, true, false, null); MemoryStream ms = new MemoryStream(); ser.WriteObject(ms, value); return Encoding.UTF8.GetString(ms.ToArray(), 0, (int)ms.Length); } This DOES work but produces some pretty heinous XML (formatted with line breaks and indentation here): <ArrayOfKeyValueOfstringanyType xmlns="http://schemas.microsoft.com/2003/10/Serialization/Arrays" xmlns:i="http://www.w3.org/2001/XMLSchema-instance"> <KeyValueOfstringanyType> <Key>key</Key> <Value i:type="a:string" xmlns:a="http://www.w3.org/2001/XMLSchema">Value</Value> </KeyValueOfstringanyType> <KeyValueOfstringanyType> <Key>Key2</Key> <Value i:type="a:decimal" xmlns:a="http://www.w3.org/2001/XMLSchema">100.10</Value> </KeyValueOfstringanyType> <KeyValueOfstringanyType> <Key>Key3</Key> <Value i:type="a:guid" xmlns:a="http://schemas.microsoft.com/2003/10/Serialization/">2cd46d2a-a636-4af4-979b-e834d39b6d37</Value> </KeyValueOfstringanyType> <KeyValueOfstringanyType> <Key>Key4</Key> <Value i:type="a:dateTime" xmlns:a="http://www.w3.org/2001/XMLSchema">2011-09-19T17:17:05.4406999-07:00</Value> </KeyValueOfstringanyType> <KeyValueOfstringanyType> <Key>Key5</Key> <Value i:type="a:boolean" xmlns:a="http://www.w3.org/2001/XMLSchema">true</Value> </KeyValueOfstringanyType> <KeyValueOfstringanyType> <Key>Key7</Key> <Value i:type="a:base64Binary" xmlns:a="http://www.w3.org/2001/XMLSchema">Ki1C</Value> </KeyValueOfstringanyType> </ArrayOfKeyValueOfstringanyType> Ouch! That seriously hurts the eye! :-) Worse though it's extremely verbose with all those repetitive namespace declarations. It's good to know that it works in a pinch, but for a human readable/editable solution or something lightweight to store in a database it's not quite ideal. Why should I care? As a little background, in one of my applications I have a need for a flexible property bag that is used on a free form database field on an otherwise static entity. Basically what I have is a standard database record to which arbitrary properties can be added in an XML based string field. I intend to expose those arbitrary properties as a collection from field data stored in XML. The concept is pretty simple: When loading write the data to the collection, when the data is saved serialize the data into an XML string and store it into the database. When reading the data pick up the XML and if the collection on the entity is accessed automatically deserialize the XML into the Dictionary. (I'll talk more about this in another post). While the DataContext Serializer would work, it's verbosity is problematic both for size of the generated XML strings and the fact that users can manually edit this XML based property data in an advanced mode. A clean(er) layout certainly would be preferable and more user friendly. Custom XMLSerialization with a PropertyBag Class So… after a bunch of experimentation with different serialization formats I decided to create a custom PropertyBag class that provides for a serializable Dictionary. It's basically a custom Dictionary<TType,TValue> implementation with the keys always set as string keys. The result are PropertyBag<TValue> and PropertyBag (which defaults to the object type for values). The PropertyBag<TType> and PropertyBag classes provide these features: Subclassed from Dictionary<T,V> Implements IXmlSerializable with a cleanish XML format ToXml() and FromXml() methods to export and import to and from XML strings Static CreateFromXml() method to create an instance It's simple enough as it's merely a Dictionary<string,object> subclass but that supports serialization to a - what I think at least - cleaner XML format. The class is super simple to use: [TestMethod] public void PropertyBagTwoWayObjectSerializationTest() { var bag = new PropertyBag(); bag.Add("key", "Value"); bag.Add("Key2", 100.10M); bag.Add("Key3", Guid.NewGuid()); bag.Add("Key4", DateTime.Now); bag.Add("Key5", true); bag.Add("Key7", new byte[3] { 42,45,66 } ); bag.Add("Key8", null); bag.Add("Key9", new ComplexObject() { Name = "Rick", Entered = DateTime.Now, Count = 10 }); string xml = bag.ToXml(); TestContext.WriteLine(bag.ToXml()); bag.Clear(); bag.FromXml(xml); Assert.IsTrue(bag["key"] as string == "Value"); Assert.IsInstanceOfType( bag["Key3"], typeof(Guid)); Assert.IsNull(bag["Key8"]); //Assert.IsNull(bag["Key10"]); Assert.IsInstanceOfType(bag["Key9"], typeof(ComplexObject)); } This uses the PropertyBag class which uses a PropertyBag<string,object> - which means it returns untyped values of type object. I suspect for me this will be the most common scenario as I'd want to store arbitrary values in the PropertyBag rather than one specific type. The same code with a strongly typed PropertyBag<decimal> looks like this: [TestMethod] public void PropertyBagTwoWayValueTypeSerializationTest() { var bag = new PropertyBag<decimal>(); bag.Add("key", 10M); bag.Add("Key1", 100.10M); bag.Add("Key2", 200.10M); bag.Add("Key3", 300.10M); string xml = bag.ToXml(); TestContext.WriteLine(bag.ToXml()); bag.Clear(); bag.FromXml(xml); Assert.IsTrue(bag.Get("Key1") == 100.10M); Assert.IsTrue(bag.Get("Key3") == 300.10M); } and produces typed results of type decimal. The types can be either value or reference types the combination of which actually proved to be a little more tricky than anticipated due to null and specific string value checks required - getting the generic typing right required use of default(T) and Convert.ChangeType() to trick the compiler into playing nice. Of course the whole raison d'etre for this class is the XML serialization. You can see in the code above that we're doing a .ToXml() and .FromXml() to serialize to and from string. The XML produced for the first example looks like this: <?xml version="1.0" encoding="utf-8"?> <properties> <item> <key>key</key> <value>Value</value> </item> <item> <key>Key2</key> <value type="decimal">100.10</value> </item> <item> <key>Key3</key> <value type="___System.Guid"> <guid>f7a92032-0c6d-4e9d-9950-b15ff7cd207d</guid> </value> </item> <item> <key>Key4</key> <value type="datetime">2011-09-26T17:45:58.5789578-10:00</value> </item> <item> <key>Key5</key> <value type="boolean">true</value> </item> <item> <key>Key7</key> <value type="base64Binary">Ki1C</value> </item> <item> <key>Key8</key> <value type="nil" /> </item> <item> <key>Key9</key> <value type="___Westwind.Tools.Tests.PropertyBagTest+ComplexObject"> <ComplexObject> <Name>Rick</Name> <Entered>2011-09-26T17:45:58.5789578-10:00</Entered> <Count>10</Count> </ComplexObject> </value> </item> </properties>   The format is a bit cleaner than the DataContractSerializer. Each item is serialized into <key> <value> pairs. If the value is a string no type information is written. Since string tends to be the most common type this saves space and serialization processing. All other types are attributed. Simple types are mapped to XML types so things like decimal, datetime, boolean and base64Binary are encoded using their Xml type values. All other types are embedded with a hokey format that describes the .NET type preceded by a three underscores and then are encoded using the XmlSerializer. You can see this best above in the ComplexObject encoding. For custom types this isn't pretty either, but it's more concise than the DCS and it works as long as you're serializing back and forth between .NET clients at least. The XML generated from the second example that uses PropertyBag<decimal> looks like this: <?xml version="1.0" encoding="utf-8"?> <properties> <item> <key>key</key> <value type="decimal">10</value> </item> <item> <key>Key1</key> <value type="decimal">100.10</value> </item> <item> <key>Key2</key> <value type="decimal">200.10</value> </item> <item> <key>Key3</key> <value type="decimal">300.10</value> </item> </properties>   How does it work As I mentioned there's nothing fancy about this solution - it's little more than a subclass of Dictionary<T,V> that implements custom Xml Serialization and a couple of helper methods that facilitate getting the XML in and out of the class more easily. But it's proven very handy for a number of projects for me where dynamic data storage is required. Here's the code: /// <summary> /// Creates a serializable string/object dictionary that is XML serializable /// Encodes keys as element names and values as simple values with a type /// attribute that contains an XML type name. Complex names encode the type /// name with type='___namespace.classname' format followed by a standard xml /// serialized format. The latter serialization can be slow so it's not recommended /// to pass complex types if performance is critical. /// </summary> [XmlRoot("properties")] public class PropertyBag : PropertyBag<object> { /// <summary> /// Creates an instance of a propertybag from an Xml string /// </summary> /// <param name="xml">Serialize</param> /// <returns></returns> public static PropertyBag CreateFromXml(string xml) { var bag = new PropertyBag(); bag.FromXml(xml); return bag; } } /// <summary> /// Creates a serializable string for generic types that is XML serializable. /// /// Encodes keys as element names and values as simple values with a type /// attribute that contains an XML type name. Complex names encode the type /// name with type='___namespace.classname' format followed by a standard xml /// serialized format. The latter serialization can be slow so it's not recommended /// to pass complex types if performance is critical. /// </summary> /// <typeparam name="TValue">Must be a reference type. For value types use type object</typeparam> [XmlRoot("properties")] public class PropertyBag<TValue> : Dictionary<string, TValue>, IXmlSerializable { /// <summary> /// Not implemented - this means no schema information is passed /// so this won't work with ASMX/WCF services. /// </summary> /// <returns></returns> public System.Xml.Schema.XmlSchema GetSchema() { return null; } /// <summary> /// Serializes the dictionary to XML. Keys are /// serialized to element names and values as /// element values. An xml type attribute is embedded /// for each serialized element - a .NET type /// element is embedded for each complex type and /// prefixed with three underscores. /// </summary> /// <param name="writer"></param> public void WriteXml(System.Xml.XmlWriter writer) { foreach (string key in this.Keys) { TValue value = this[key]; Type type = null; if (value != null) type = value.GetType(); writer.WriteStartElement("item"); writer.WriteStartElement("key"); writer.WriteString(key as string); writer.WriteEndElement(); writer.WriteStartElement("value"); string xmlType = XmlUtils.MapTypeToXmlType(type); bool isCustom = false; // Type information attribute if not string if (value == null) { writer.WriteAttributeString("type", "nil"); } else if (!string.IsNullOrEmpty(xmlType)) { if (xmlType != "string") { writer.WriteStartAttribute("type"); writer.WriteString(xmlType); writer.WriteEndAttribute(); } } else { isCustom = true; xmlType = "___" + value.GetType().FullName; writer.WriteStartAttribute("type"); writer.WriteString(xmlType); writer.WriteEndAttribute(); } // Actual deserialization if (!isCustom) { if (value != null) writer.WriteValue(value); } else { XmlSerializer ser = new XmlSerializer(value.GetType()); ser.Serialize(writer, value); } writer.WriteEndElement(); // value writer.WriteEndElement(); // item } } /// <summary> /// Reads the custom serialized format /// </summary> /// <param name="reader"></param> public void ReadXml(System.Xml.XmlReader reader) { this.Clear(); while (reader.Read()) { if (reader.NodeType == XmlNodeType.Element && reader.Name == "key") { string xmlType = null; string name = reader.ReadElementContentAsString(); // item element reader.ReadToNextSibling("value"); if (reader.MoveToNextAttribute()) xmlType = reader.Value; reader.MoveToContent(); TValue value; if (xmlType == "nil") value = default(TValue); // null else if (string.IsNullOrEmpty(xmlType)) { // value is a string or object and we can assign TValue to value string strval = reader.ReadElementContentAsString(); value = (TValue) Convert.ChangeType(strval, typeof(TValue)); } else if (xmlType.StartsWith("___")) { while (reader.Read() && reader.NodeType != XmlNodeType.Element) { } Type type = ReflectionUtils.GetTypeFromName(xmlType.Substring(3)); //value = reader.ReadElementContentAs(type,null); XmlSerializer ser = new XmlSerializer(type); value = (TValue)ser.Deserialize(reader); } else value = (TValue)reader.ReadElementContentAs(XmlUtils.MapXmlTypeToType(xmlType), null); this.Add(name, value); } } } /// <summary> /// Serializes this dictionary to an XML string /// </summary> /// <returns>XML String or Null if it fails</returns> public string ToXml() { string xml = null; SerializationUtils.SerializeObject(this, out xml); return xml; } /// <summary> /// Deserializes from an XML string /// </summary> /// <param name="xml"></param> /// <returns>true or false</returns> public bool FromXml(string xml) { this.Clear(); // if xml string is empty we return an empty dictionary if (string.IsNullOrEmpty(xml)) return true; var result = SerializationUtils.DeSerializeObject(xml, this.GetType()) as PropertyBag<TValue>; if (result != null) { foreach (var item in result) { this.Add(item.Key, item.Value); } } else // null is a failure return false; return true; } /// <summary> /// Creates an instance of a propertybag from an Xml string /// </summary> /// <param name="xml"></param> /// <returns></returns> public static PropertyBag<TValue> CreateFromXml(string xml) { var bag = new PropertyBag<TValue>(); bag.FromXml(xml); return bag; } } } The code uses a couple of small helper classes SerializationUtils and XmlUtils for mapping Xml types to and from .NET, both of which are from the WestWind,Utilities project (which is the same project where PropertyBag lives) from the West Wind Web Toolkit. The code implements ReadXml and WriteXml for the IXmlSerializable implementation using old school XmlReaders and XmlWriters (because it's pretty simple stuff - no need for XLinq here). Then there are two helper methods .ToXml() and .FromXml() that basically allow your code to easily convert between XML and a PropertyBag object. In my code that's what I use to actually to persist to and from the entity XML property during .Load() and .Save() operations. It's sweet to be able to have a string key dictionary and then be able to turn around with 1 line of code to persist the whole thing to XML and back. Hopefully some of you will find this class as useful as I've found it. It's a simple solution to a common requirement in my applications and I've used the hell out of it in the  short time since I created it. Resources You can find the complete code for the two classes plus the helpers in the Subversion repository for Westwind.Utilities. You can grab the source files from there or download the whole project. You can also grab the full Westwind.Utilities assembly from NuGet and add it to your project if that's easier for you. PropertyBag Source Code SerializationUtils and XmlUtils Westwind.Utilities Assembly on NuGet (add from Visual Studio) © Rick Strahl, West Wind Technologies, 2005-2011Posted in .NET  CSharp   Tweet (function() { var po = document.createElement('script'); po.type = 'text/javascript'; po.async = true; po.src = 'https://apis.google.com/js/plusone.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(po, s); })();

    Read the article

  • Automating Solaris 11 Zones Installation Using The Automated Install Server

    - by Orgad Kimchi
    Introduction How to use the Oracle Solaris 11 Automated install server in order to automate the Solaris 11 Zones installation. In this document I will demonstrate how to setup the Automated Install server in order to provide hands off installation process for the Global Zone and two Non Global Zones located on the same system. Architecture layout: Figure 1. Architecture layout Prerequisite Setup the Automated install server (AI) using the following instructions “How to Set Up Automated Installation Services for Oracle Solaris 11” The first step in this setup will be creating two Solaris 11 Zones configuration files. Step 1: Create the Solaris 11 Zones configuration files  The Solaris Zones configuration files should be in the format of the zonecfg export command. # zonecfg -z zone1 export > /var/tmp/zone1# cat /var/tmp/zone1 create -b set brand=solaris set zonepath=/rpool/zones/zone1 set autoboot=true set ip-type=exclusive add anet set linkname=net0 set lower-link=auto set configure-allowed-address=true set link-protection=mac-nospoof set mac-address=random end  Create a backup copy of this file under a different name, for example, zone2. # cp /var/tmp/zone1 /var/tmp/zone2 Modify the second configuration file with the zone2 configuration information You should change the zonepath for example: set zonepath=/rpool/zones/zone2 Step2: Copy and share the Zones configuration files  Create the NFS directory for the Zones configuration files # mkdir /export/zone_config Share the directory for the Zones configuration file # share –o ro /export/zone_config Copy the Zones configuration files into the NFS shared directory # cp /var/tmp/zone1 /var/tmp/zone2  /export/zone_config Verify that the NFS share has been created using the following command # share export_zone_config      /export/zone_config     nfs     sec=sys,ro Step 3: Add the Global Zone as client to the Install Service Use the installadm create-client command to associate client (Global Zone) with the install service To find the MAC address of a system, use the dladm command as described in the dladm(1M) man page. The following command adds the client (Global Zone) with MAC address 0:14:4f:2:a:19 to the s11x86service install service. # installadm create-client -e “0:14:4f:2:a:19" -n s11x86service You can verify the client creation using the following command # installadm list –c Service Name  Client Address     Arch   Image Path ------------  --------------     ----   ---------- s11x86service 00:14:4F:02:0A:19  i386   /export/auto_install/s11x86service We can see the client install service name (s11x86service), MAC address (00:14:4F:02:0A:19 and Architecture (i386). Step 4: Global Zone manifest setup  First, get a list of the installation services and the manifests associated with them: # installadm list -m Service Name   Manifest        Status ------------   --------        ------ default-i386   orig_default   Default s11x86service  orig_default   Default Then probe the s11x86service and the default manifest associated with it. The -m switch reflects the name of the manifest associated with a service. Since we want to capture that output into a file, we redirect the output of the command as follows: # installadm export -n s11x86service -m orig_default >  /var/tmp/orig_default.xml Create a backup copy of this file under a different name, for example, orig-default2.xml, and edit the copy. # cp /var/tmp/orig_default.xml /var/tmp/orig_default2.xml Use the configuration element in the AI manifest for the client system to specify non-global zones. Use the name attribute of the configuration element to specify the name of the zone. Use the source attribute to specify the location of the config file for the zone.The source location can be any http:// or file:// location that the client can access during installation. The following sample AI manifest specifies two Non-Global Zones: zone1 and zone2 You should replace the server_ip with the ip address of the NFS server. <!DOCTYPE auto_install SYSTEM "file:///usr/share/install/ai.dtd.1"> <auto_install>   <ai_instance>     <target>       <logical>         <zpool name="rpool" is_root="true">           <filesystem name="export" mountpoint="/export"/>           <filesystem name="export/home"/>           <be name="solaris"/>         </zpool>       </logical>     </target>     <software type="IPS">       <source>         <publisher name="solaris">           <origin name="http://pkg.oracle.com/solaris/release"/>         </publisher>       </source>       <software_data action="install">         <name>pkg:/entire@latest</name>         <name>pkg:/group/system/solaris-large-server</name>       </software_data>     </software>     <configuration type="zone" name="zone1" source="file:///net/server_ip/export/zone_config/zone1"/>     <configuration type="zone" name="zone2" source="file:///net/server_ip/export/zone_config/zone2"/>   </ai_instance> </auto_install> The following example adds the /var/tmp/orig_default2.xml AI manifest to the s11x86service install service # installadm create-manifest -n s11x86service -f /var/tmp/orig_default2.xml -m gzmanifest You can verify the manifest creation using the following command # installadm list -n s11x86service  -m Service/Manifest Name  Status   Criteria ---------------------  ------   -------- s11x86service    orig_default        Default  None    gzmanifest          Inactive None We can see from the command output that the new manifest named gzmanifest has been created and associated with the s11x86service install service. Step 5: Non Global Zone manifest setup The AI manifest for non-global zone installation is similar to the AI manifest for installing the global zone. If you do not provide a custom AI manifest for a non-global zone, the default AI manifest for Zones is used The default AI manifest for Zones is available at /usr/share/auto_install/manifest/zone_default.xml. In this example we should use the default AI manifest for zones The following sample default AI manifest for zones # cat /usr/share/auto_install/manifest/zone_default.xml <?xml version="1.0" encoding="UTF-8"?> <!--  Copyright (c) 2011, 2012, Oracle and/or its affiliates. All rights reserved. --> <!DOCTYPE auto_install SYSTEM "file:///usr/share/install/ai.dtd.1"> <auto_install>     <ai_instance name="zone_default">         <target>             <logical>                 <zpool name="rpool">                     <!--                       Subsequent <filesystem> entries instruct an installer                       to create following ZFS datasets:                           <root_pool>/export         (mounted on /export)                           <root_pool>/export/home    (mounted on /export/home)                       Those datasets are part of standard environment                       and should be always created.                       In rare cases, if there is a need to deploy a zone                       without these datasets, either comment out or remove                       <filesystem> entries. In such scenario, it has to be also                       assured that in case of non-interactive post-install                       configuration, creation of initial user account is                       disabled in related system configuration profile.                       Otherwise the installed zone would fail to boot.                     -->                     <filesystem name="export" mountpoint="/export"/>                     <filesystem name="export/home"/>                     <be name="solaris">                         <options>                             <option name="compression" value="on"/>                         </options>                     </be>                 </zpool>             </logical>         </target>         <software type="IPS">             <destination>                 <image>                     <!-- Specify locales to install -->                     <facet set="false">facet.locale.*</facet>                     <facet set="true">facet.locale.de</facet>                     <facet set="true">facet.locale.de_DE</facet>                     <facet set="true">facet.locale.en</facet>                     <facet set="true">facet.locale.en_US</facet>                     <facet set="true">facet.locale.es</facet>                     <facet set="true">facet.locale.es_ES</facet>                     <facet set="true">facet.locale.fr</facet>                     <facet set="true">facet.locale.fr_FR</facet>                     <facet set="true">facet.locale.it</facet>                     <facet set="true">facet.locale.it_IT</facet>                     <facet set="true">facet.locale.ja</facet>                     <facet set="true">facet.locale.ja_*</facet>                     <facet set="true">facet.locale.ko</facet>                     <facet set="true">facet.locale.ko_*</facet>                     <facet set="true">facet.locale.pt</facet>                     <facet set="true">facet.locale.pt_BR</facet>                     <facet set="true">facet.locale.zh</facet>                     <facet set="true">facet.locale.zh_CN</facet>                     <facet set="true">facet.locale.zh_TW</facet>                 </image>             </destination>             <software_data action="install">                 <name>pkg:/group/system/solaris-small-server</name>             </software_data>         </software>     </ai_instance> </auto_install> (optional) We can customize the default AI manifest for Zones Create a backup copy of this file under a different name, for example, zone_default2.xml and edit the copy # cp /usr/share/auto_install/manifest/zone_default.xml /var/tmp/zone_default2.xml Edit the copy (/var/tmp/zone_default2.xml) The following example adds the /var/tmp/zone_default2.xml AI manifest to the s11x86service install service and specifies that zone1 and zone2 should use this manifest. # installadm create-manifest -n s11x86service -f /var/tmp/zone_default2.xml -m zones_manifest -c zonename="zone1 zone2" Note: Do not use the following elements or attributes in a non-global zone AI manifest:     The auto_reboot attribute of the ai_instance element     The http_proxy attribute of the ai_instance element     The disk child element of the target element     The noswap attribute of the logical element     The nodump attribute of the logical element     The configuration element Step 6: Global Zone profile setup We are going to create a global zone configuration profile which includes the host information for example: host name, ip address name services etc… # sysconfig create-profile –o /var/tmp/gz_profile.xml You need to provide the host information for example:     Default router     Root password     DNS information The output should eventually disappear and be replaced by the initial screen of the System Configuration Tool (see Figure 2), where you can do the final configuration. Figure 2. Profile creation menu You can validate the profile using the following command # installadm validate -n s11x86service –P /var/tmp/gz_profile.xml Validating static profile gz_profile.xml...  Passed Next, instantiate a profile with the install service. In our case, use the following syntax for doing this # installadm create-profile -n s11x86service  -f /var/tmp/gz_profile.xml -p  gz_profile You can verify profile creation using the following command # installadm list –n s11x86service  -p Service/Profile Name  Criteria --------------------  -------- s11x86service    gz_profile         None We can see that the gz_profie has been created and associated with the s11x86service Install service. Step 7: Setup the Solaris Zones configuration profiles The step should be similar to the Global zone profile creation on step 6 # sysconfig create-profile –o /var/tmp/zone1_profile.xml # sysconfig create-profile –o /var/tmp/zone2_profile.xml You can validate the profiles using the following command # installadm validate -n s11x86service -P /var/tmp/zone1_profile.xml Validating static profile zone1_profile.xml...  Passed # installadm validate -n s11x86service -P /var/tmp/zone2_profile.xml Validating static profile zone2_profile.xml...  Passed Next, associate the profiles with the install service The following example adds the zone1_profile.xml configuration profile to the s11x86service  install service and specifies that zone1 should use this profile. # installadm create-profile -n s11x86service  -f  /var/tmp/zone1_profile.xml -p zone1_profile -c zonename=zone1 The following example adds the zone2_profile.xml configuration profile to the s11x86service  install service and specifies that zone2 should use this profile. # installadm create-profile -n s11x86service  -f  /var/tmp/zone2_profile.xml -p zone2_profile -c zonename=zone2 You can verify the profiles creation using the following command # installadm list -n s11x86service -p Service/Profile Name  Criteria --------------------  -------- s11x86service    zone1_profile      zonename = zone1    zone2_profile      zonename = zone2    gz_profile         None We can see that we have three profiles in the s11x86service  install service     Global Zone  gz_profile     zone1            zone1_profile     zone2            zone2_profile. Step 8: Global Zone setup Associate the global zone client with the manifest and the profile that we create in the previous steps The following example adds the manifest and profile to the client (global zone), where: gzmanifest  is the name of the manifest. gz_profile  is the name of the configuration profile. mac="0:14:4f:2:a:19" is the client (global zone) mac address s11x86service is the install service name. # installadm set-criteria -m  gzmanifest  –p  gz_profile  -c mac="0:14:4f:2:a:19" -n s11x86service You can verify the manifest and profile association using the following command # installadm list -n s11x86service -p  -m Service/Manifest Name  Status   Criteria ---------------------  ------   -------- s11x86service    gzmanifest                   mac  = 00:14:4F:02:0A:19    orig_default        Default  None Service/Profile Name  Criteria --------------------  -------- s11x86service    gz_profile         mac      = 00:14:4F:02:0A:19    zone2_profile      zonename = zone2    zone1_profile      zonename = zone1 Step 9: Provision the host with the Non-Global Zones The next step is to boot the client system off the network and provision it using the Automated Install service that we just set up. First, boot the client system. Figure 3 shows the network boot attempt (when done on an x86 system): Figure 3. Network Boot Then you will be prompted by a GRUB menu, with a timer, as shown in Figure 4. The default selection (the "Text Installer and command line" option) is highlighted.  Press the down arrow to highlight the second option labeled Automated Install, and then press Enter. The reason we need to do this is because we want to prevent a system from being automatically re-installed if it were to be booted from the network accidentally. Figure 4. GRUB Menu What follows is the continuation of a networked boot from the Automated Install server,. The client downloads a mini-root (a small set of files in which to successfully run the installer), identifies the location of the Automated Install manifest on the network, retrieves that manifest, and then processes it to identify the address of the IPS repository from which to obtain the desired software payload. Non-Global Zones are installed and configured on the first reboot after the Global Zone is installed. You can list all the Solaris Zones status using the following command # zoneadm list -civ Once the Zones are in running state you can login into the Zone using the following command # zlogin –z zone1 Troubleshooting Automated Installations If an installation to a client system failed, you can find the client log at /system/volatile/install_log. NOTE: Zones are not installed if any of the following errors occurs:     A zone config file is not syntactically correct.     A collision exists among zone names, zone paths, or delegated ZFS datasets in the set of zones to be installed     Required datasets are not configured in the global zone. For more troubleshooting information see “Installing Oracle Solaris 11 Systems” Conclusion This paper demonstrated the benefits of using the Automated Install server to simplify the Non Global Zones setup, including the creation and configuration of the global zone manifest and the Solaris Zones profiles.

    Read the article

  • Parallelism in .NET – Part 5, Partitioning of Work

    - by Reed
    When parallelizing any routine, we start by decomposing the problem.  Once the problem is understood, we need to break our work into separate tasks, so each task can be run on a different processing element.  This process is called partitioning. Partitioning our tasks is a challenging feat.  There are opposing forces at work here: too many partitions adds overhead, too few partitions leaves processors idle.  Trying to work the perfect balance between the two extremes is the goal for which we should aim.  Luckily, the Task Parallel Library automatically handles much of this process.  However, there are situations where the default partitioning may not be appropriate, and knowledge of our routines may allow us to guide the framework to making better decisions. First off, I’d like to say that this is a more advanced topic.  It is perfectly acceptable to use the parallel constructs in the framework without considering the partitioning taking place.  The default behavior in the Task Parallel Library is very well-behaved, even for unusual work loads, and should rarely be adjusted.  I have found few situations where the default partitioning behavior in the TPL is not as good or better than my own hand-written partitioning routines, and recommend using the defaults unless there is a strong, measured, and profiled reason to avoid using them.  However, understanding partitioning, and how the TPL partitions your data, helps in understanding the proper usage of the TPL. I indirectly mentioned partitioning while discussing aggregation.  Typically, our systems will have a limited number of Processing Elements (PE), which is the terminology used for hardware capable of processing a stream of instructions.  For example, in a standard Intel i7 system, there are four processor cores, each of which has two potential hardware threads due to Hyperthreading.  This gives us a total of 8 PEs – theoretically, we can have up to eight operations occurring concurrently within our system. In order to fully exploit this power, we need to partition our work into Tasks.  A task is a simple set of instructions that can be run on a PE.  Ideally, we want to have at least one task per PE in the system, since fewer tasks means that some of our processing power will be sitting idle.  A naive implementation would be to just take our data, and partition it with one element in our collection being treated as one task.  When we loop through our collection in parallel, using this approach, we’d just process one item at a time, then reuse that thread to process the next, etc.  There’s a flaw in this approach, however.  It will tend to be slower than necessary, often slower than processing the data serially. The problem is that there is overhead associated with each task.  When we take a simple foreach loop body and implement it using the TPL, we add overhead.  First, we change the body from a simple statement to a delegate, which must be invoked.  In order to invoke the delegate on a separate thread, the delegate gets added to the ThreadPool’s current work queue, and the ThreadPool must pull this off the queue, assign it to a free thread, then execute it.  If our collection had one million elements, the overhead of trying to spawn one million tasks would destroy our performance. The answer, here, is to partition our collection into groups, and have each group of elements treated as a single task.  By adding a partitioning step, we can break our total work into small enough tasks to keep our processors busy, but large enough tasks to avoid overburdening the ThreadPool.  There are two clear, opposing goals here: Always try to keep each processor working, but also try to keep the individual partitions as large as possible. When using Parallel.For, the partitioning is always handled automatically.  At first, partitioning here seems simple.  A naive implementation would merely split the total element count up by the number of PEs in the system, and assign a chunk of data to each processor.  Many hand-written partitioning schemes work in this exactly manner.  This perfectly balanced, static partitioning scheme works very well if the amount of work is constant for each element.  However, this is rarely the case.  Often, the length of time required to process an element grows as we progress through the collection, especially if we’re doing numerical computations.  In this case, the first PEs will finish early, and sit idle waiting on the last chunks to finish.  Sometimes, work can decrease as we progress, since previous computations may be used to speed up later computations.  In this situation, the first chunks will be working far longer than the last chunks.  In order to balance the workload, many implementations create many small chunks, and reuse threads.  This adds overhead, but does provide better load balancing, which in turn improves performance. The Task Parallel Library handles this more elaborately.  Chunks are determined at runtime, and start small.  They grow slowly over time, getting larger and larger.  This tends to lead to a near optimum load balancing, even in odd cases such as increasing or decreasing workloads.  Parallel.ForEach is a bit more complicated, however. When working with a generic IEnumerable<T>, the number of items required for processing is not known in advance, and must be discovered at runtime.  In addition, since we don’t have direct access to each element, the scheduler must enumerate the collection to process it.  Since IEnumerable<T> is not thread safe, it must lock on elements as it enumerates, create temporary collections for each chunk to process, and schedule this out.  By default, it uses a partitioning method similar to the one described above.  We can see this directly by looking at the Visual Partitioning sample shipped by the Task Parallel Library team, and available as part of the Samples for Parallel Programming.  When we run the sample, with four cores and the default, Load Balancing partitioning scheme, we see this: The colored bands represent each processing core.  You can see that, when we started (at the top), we begin with very small bands of color.  As the routine progresses through the Parallel.ForEach, the chunks get larger and larger (seen by larger and larger stripes). Most of the time, this is fantastic behavior, and most likely will out perform any custom written partitioning.  However, if your routine is not scaling well, it may be due to a failure in the default partitioning to handle your specific case.  With prior knowledge about your work, it may be possible to partition data more meaningfully than the default Partitioner. There is the option to use an overload of Parallel.ForEach which takes a Partitioner<T> instance.  The Partitioner<T> class is an abstract class which allows for both static and dynamic partitioning.  By overriding Partitioner<T>.SupportsDynamicPartitions, you can specify whether a dynamic approach is available.  If not, your custom Partitioner<T> subclass would override GetPartitions(int), which returns a list of IEnumerator<T> instances.  These are then used by the Parallel class to split work up amongst processors.  When dynamic partitioning is available, GetDynamicPartitions() is used, which returns an IEnumerable<T> for each partition.  If you do decide to implement your own Partitioner<T>, keep in mind the goals and tradeoffs of different partitioning strategies, and design appropriately. The Samples for Parallel Programming project includes a ChunkPartitioner class in the ParallelExtensionsExtras project.  This provides example code for implementing your own, custom allocation strategies, including a static allocator of a given chunk size.  Although implementing your own Partitioner<T> is possible, as I mentioned above, this is rarely required or useful in practice.  The default behavior of the TPL is very good, often better than any hand written partitioning strategy.

    Read the article

  • Using Durandal to Create Single Page Apps

    - by Stephen.Walther
    A few days ago, I gave a talk on building Single Page Apps on the Microsoft Stack. In that talk, I recommended that people use Knockout, Sammy, and RequireJS to build their presentation layer and use the ASP.NET Web API to expose data from their server. After I gave the talk, several people contacted me and suggested that I investigate a new open-source JavaScript library named Durandal. Durandal stitches together Knockout, Sammy, and RequireJS to make it easier to use these technologies together. In this blog entry, I want to provide a brief walkthrough of using Durandal to create a simple Single Page App. I am going to demonstrate how you can create a simple Movies App which contains (virtual) pages for viewing a list of movies, adding new movies, and viewing movie details. The goal of this blog entry is to give you a sense of what it is like to build apps with Durandal. Installing Durandal First things first. How do you get Durandal? The GitHub project for Durandal is located here: https://github.com/BlueSpire/Durandal The Wiki — located at the GitHub project — contains all of the current documentation for Durandal. Currently, the documentation is a little sparse, but it is enough to get you started. Instead of downloading the Durandal source from GitHub, a better option for getting started with Durandal is to install one of the Durandal NuGet packages. I built the Movies App described in this blog entry by first creating a new ASP.NET MVC 4 Web Application with the Basic Template. Next, I executed the following command from the Package Manager Console: Install-Package Durandal.StarterKit As you can see from the screenshot of the Package Manager Console above, the Durandal Starter Kit package has several dependencies including: · jQuery · Knockout · Sammy · Twitter Bootstrap The Durandal Starter Kit package includes a sample Durandal application. You can get to the Starter Kit app by navigating to the Durandal controller. Unfortunately, when I first tried to run the Starter Kit app, I got an error because the Starter Kit is hard-coded to use a particular version of jQuery which is already out of date. You can fix this issue by modifying the App_Start\DurandalBundleConfig.cs file so it is jQuery version agnostic like this: bundles.Add( new ScriptBundle("~/scripts/vendor") .Include("~/Scripts/jquery-{version}.js") .Include("~/Scripts/knockout-{version}.js") .Include("~/Scripts/sammy-{version}.js") // .Include("~/Scripts/jquery-1.9.0.min.js") // .Include("~/Scripts/knockout-2.2.1.js") // .Include("~/Scripts/sammy-0.7.4.min.js") .Include("~/Scripts/bootstrap.min.js") ); The recommendation is that you create a Durandal app in a folder off your project root named App. The App folder in the Starter Kit contains the following subfolders and files: · durandal – This folder contains the actual durandal JavaScript library. · viewmodels – This folder contains all of your application’s view models. · views – This folder contains all of your application’s views. · main.js — This file contains all of the JavaScript startup code for your app including the client-side routing configuration. · main-built.js – This file contains an optimized version of your application. You need to build this file by using the RequireJS optimizer (unfortunately, before you can run the optimizer, you must first install NodeJS). For the purpose of this blog entry, I wanted to start from scratch when building the Movies app, so I deleted all of these files and folders except for the durandal folder which contains the durandal library. Creating the ASP.NET MVC Controller and View A Durandal app is built using a single server-side ASP.NET MVC controller and ASP.NET MVC view. A Durandal app is a Single Page App. When you navigate between pages, you are not navigating to new pages on the server. Instead, you are loading new virtual pages into the one-and-only-one server-side view. For the Movies app, I created the following ASP.NET MVC Home controller: public class HomeController : Controller { public ActionResult Index() { return View(); } } There is nothing special about the Home controller – it is as basic as it gets. Next, I created the following server-side ASP.NET view. This is the one-and-only server-side view used by the Movies app: @{ Layout = null; } <!DOCTYPE html> <html> <head> <title>Index</title> </head> <body> <div id="applicationHost"> Loading app.... </div> @Scripts.Render("~/scripts/vendor") <script type="text/javascript" src="~/App/durandal/amd/require.js" data-main="/App/main"></script> </body> </html> Notice that I set the Layout property for the view to the value null. If you neglect to do this, then the default ASP.NET MVC layout will be applied to the view and you will get the <!DOCTYPE> and opening and closing <html> tags twice. Next, notice that the view contains a DIV element with the Id applicationHost. This marks the area where virtual pages are loaded. When you navigate from page to page in a Durandal app, HTML page fragments are retrieved from the server and stuck in the applicationHost DIV element. Inside the applicationHost element, you can place any content which you want to display when a Durandal app is starting up. For example, you can create a fancy splash screen. I opted for simply displaying the text “Loading app…”: Next, notice the view above includes a call to the Scripts.Render() helper. This helper renders out all of the JavaScript files required by the Durandal library such as jQuery and Knockout. Remember to fix the App_Start\DurandalBundleConfig.cs as described above or Durandal will attempt to load an old version of jQuery and throw a JavaScript exception and stop working. Your application JavaScript code is not included in the scripts rendered by the Scripts.Render helper. Your application code is loaded dynamically by RequireJS with the help of the following SCRIPT element located at the bottom of the view: <script type="text/javascript" src="~/App/durandal/amd/require.js" data-main="/App/main"></script> The data-main attribute on the SCRIPT element causes RequireJS to load your /app/main.js JavaScript file to kick-off your Durandal app. Creating the Durandal Main.js File The Durandal Main.js JavaScript file, located in your App folder, contains all of the code required to configure the behavior of Durandal. Here’s what the Main.js file looks like in the case of the Movies app: require.config({ paths: { 'text': 'durandal/amd/text' } }); define(function (require) { var app = require('durandal/app'), viewLocator = require('durandal/viewLocator'), system = require('durandal/system'), router = require('durandal/plugins/router'); //>>excludeStart("build", true); system.debug(true); //>>excludeEnd("build"); app.start().then(function () { //Replace 'viewmodels' in the moduleId with 'views' to locate the view. //Look for partial views in a 'views' folder in the root. viewLocator.useConvention(); //configure routing router.useConvention(); router.mapNav("movies/show"); router.mapNav("movies/add"); router.mapNav("movies/details/:id"); app.adaptToDevice(); //Show the app by setting the root view model for our application with a transition. app.setRoot('viewmodels/shell', 'entrance'); }); }); There are three important things to notice about the main.js file above. First, notice that it contains a section which enables debugging which looks like this: //>>excludeStart(“build”, true); system.debug(true); //>>excludeEnd(“build”); This code enables debugging for your Durandal app which is very useful when things go wrong. When you call system.debug(true), Durandal writes out debugging information to your browser JavaScript console. For example, you can use the debugging information to diagnose issues with your client-side routes: (The funny looking //> symbols around the system.debug() call are RequireJS optimizer pragmas). The main.js file is also the place where you configure your client-side routes. In the case of the Movies app, the main.js file is used to configure routes for three page: the movies show, add, and details pages. //configure routing router.useConvention(); router.mapNav("movies/show"); router.mapNav("movies/add"); router.mapNav("movies/details/:id");   The route for movie details includes a route parameter named id. Later, we will use the id parameter to lookup and display the details for the right movie. Finally, the main.js file above contains the following line of code: //Show the app by setting the root view model for our application with a transition. app.setRoot('viewmodels/shell', 'entrance'); This line of code causes Durandal to load up a JavaScript file named shell.js and an HTML fragment named shell.html. I’ll discuss the shell in the next section. Creating the Durandal Shell You can think of the Durandal shell as the layout or master page for a Durandal app. The shell is where you put all of the content which you want to remain constant as a user navigates from virtual page to virtual page. For example, the shell is a great place to put your website logo and navigation links. The Durandal shell is composed from two parts: a JavaScript file and an HTML file. Here’s what the HTML file looks like for the Movies app: <h1>Movies App</h1> <div class="container-fluid page-host"> <!--ko compose: { model: router.activeItem, //wiring the router afterCompose: router.afterCompose, //wiring the router transition:'entrance', //use the 'entrance' transition when switching views cacheViews:true //telling composition to keep views in the dom, and reuse them (only a good idea with singleton view models) }--><!--/ko--> </div> And here is what the JavaScript file looks like: define(function (require) { var router = require('durandal/plugins/router'); return { router: router, activate: function () { return router.activate('movies/show'); } }; }); The JavaScript file contains the view model for the shell. This view model returns the Durandal router so you can access the list of configured routes from your shell. Notice that the JavaScript file includes a function named activate(). This function loads the movies/show page as the first page in the Movies app. If you want to create a different default Durandal page, then pass the name of a different age to the router.activate() method. Creating the Movies Show Page Durandal pages are created out of a view model and a view. The view model contains all of the data and view logic required for the view. The view contains all of the HTML markup for rendering the view model. Let’s start with the movies show page. The movies show page displays a list of movies. The view model for the show page looks like this: define(function (require) { var moviesRepository = require("repositories/moviesRepository"); return { movies: ko.observable(), activate: function() { this.movies(moviesRepository.listMovies()); } }; }); You create a view model by defining a new RequireJS module (see http://requirejs.org). You create a RequireJS module by placing all of your JavaScript code into an anonymous function passed to the RequireJS define() method. A RequireJS module has two parts. You retrieve all of the modules which your module requires at the top of your module. The code above depends on another RequireJS module named repositories/moviesRepository. Next, you return the implementation of your module. The code above returns a JavaScript object which contains a property named movies and a method named activate. The activate() method is a magic method which Durandal calls whenever it activates your view model. Your view model is activated whenever you navigate to a page which uses it. In the code above, the activate() method is used to get the list of movies from the movies repository and assign the list to the view model movies property. The HTML for the movies show page looks like this: <table> <thead> <tr> <th>Title</th><th>Director</th> </tr> </thead> <tbody data-bind="foreach:movies"> <tr> <td data-bind="text:title"></td> <td data-bind="text:director"></td> <td><a data-bind="attr:{href:'#/movies/details/'+id}">Details</a></td> </tr> </tbody> </table> <a href="#/movies/add">Add Movie</a> Notice that this is an HTML fragment. This fragment will be stuffed into the page-host DIV element in the shell.html file which is stuffed, in turn, into the applicationHost DIV element in the server-side MVC view. The HTML markup above contains data-bind attributes used by Knockout to display the list of movies (To learn more about Knockout, visit http://knockoutjs.com). The list of movies from the view model is displayed in an HTML table. Notice that the page includes a link to a page for adding a new movie. The link uses the following URL which starts with a hash: #/movies/add. Because the link starts with a hash, clicking the link does not cause a request back to the server. Instead, you navigate to the movies/add page virtually. Creating the Movies Add Page The movies add page also consists of a view model and view. The add page enables you to add a new movie to the movie database. Here’s the view model for the add page: define(function (require) { var app = require('durandal/app'); var router = require('durandal/plugins/router'); var moviesRepository = require("repositories/moviesRepository"); return { movieToAdd: { title: ko.observable(), director: ko.observable() }, activate: function () { this.movieToAdd.title(""); this.movieToAdd.director(""); this._movieAdded = false; }, canDeactivate: function () { if (this._movieAdded == false) { return app.showMessage('Are you sure you want to leave this page?', 'Navigate', ['Yes', 'No']); } else { return true; } }, addMovie: function () { // Add movie to db moviesRepository.addMovie(ko.toJS(this.movieToAdd)); // flag new movie this._movieAdded = true; // return to list of movies router.navigateTo("#/movies/show"); } }; }); The view model contains one property named movieToAdd which is bound to the add movie form. The view model also has the following three methods: 1. activate() – This method is called by Durandal when you navigate to the add movie page. The activate() method resets the add movie form by clearing out the movie title and director properties. 2. canDeactivate() – This method is called by Durandal when you attempt to navigate away from the add movie page. If you return false then navigation is cancelled. 3. addMovie() – This method executes when the add movie form is submitted. This code adds the new movie to the movie repository. I really like the Durandal canDeactivate() method. In the code above, I use the canDeactivate() method to show a warning to a user if they navigate away from the add movie page – either by clicking the Cancel button or by hitting the browser back button – before submitting the add movie form: The view for the add movie page looks like this: <form data-bind="submit:addMovie"> <fieldset> <legend>Add Movie</legend> <div> <label> Title: <input data-bind="value:movieToAdd.title" required /> </label> </div> <div> <label> Director: <input data-bind="value:movieToAdd.director" required /> </label> </div> <div> <input type="submit" value="Add" /> <a href="#/movies/show">Cancel</a> </div> </fieldset> </form> I am using Knockout to bind the movieToAdd property from the view model to the INPUT elements of the HTML form. Notice that the FORM element includes a data-bind attribute which invokes the addMovie() method from the view model when the HTML form is submitted. Creating the Movies Details Page You navigate to the movies details Page by clicking the Details link which appears next to each movie in the movies show page: The Details links pass the movie ids to the details page: #/movies/details/0 #/movies/details/1 #/movies/details/2 Here’s what the view model for the movies details page looks like: define(function (require) { var router = require('durandal/plugins/router'); var moviesRepository = require("repositories/moviesRepository"); return { movieToShow: { title: ko.observable(), director: ko.observable() }, activate: function (context) { // Grab movie from repository var movie = moviesRepository.getMovie(context.id); // Add to view model this.movieToShow.title(movie.title); this.movieToShow.director(movie.director); } }; }); Notice that the view model activate() method accepts a parameter named context. You can take advantage of the context parameter to retrieve route parameters such as the movie Id. In the code above, the context.id property is used to retrieve the correct movie from the movie repository and the movie is assigned to a property named movieToShow exposed by the view model. The movie details view displays the movieToShow property by taking advantage of Knockout bindings: <div> <h2 data-bind="text:movieToShow.title"></h2> directed by <span data-bind="text:movieToShow.director"></span> </div> Summary The goal of this blog entry was to walkthrough building a simple Single Page App using Durandal and to get a feel for what it is like to use this library. I really like how Durandal stitches together Knockout, Sammy, and RequireJS and establishes patterns for using these libraries to build Single Page Apps. Having a standard pattern which developers on a team can use to build new pages is super valuable. Once you get the hang of it, using Durandal to create new virtual pages is dead simple. Just define a new route, view model, and view and you are done. I also appreciate the fact that Durandal did not attempt to re-invent the wheel and that Durandal leverages existing JavaScript libraries such as Knockout, RequireJS, and Sammy. These existing libraries are powerful libraries and I have already invested a considerable amount of time in learning how to use them. Durandal makes it easier to use these libraries together without losing any of their power. Durandal has some additional interesting features which I have not had a chance to play with yet. For example, you can use the RequireJS optimizer to combine and minify all of a Durandal app’s code. Also, Durandal supports a way to create custom widgets (client-side controls) by composing widgets from a controller and view. You can download the code for the Movies app by clicking the following link (this is a Visual Studio 2012 project): Durandal Movie App

    Read the article

  • Replacing jQuery.live() with jQuery.on()

    - by Rick Strahl
    jQuery 1.9 and 1.10 have introduced a host of changes, but for the most part these changes are mostly transparent to existing application usage of jQuery. After spending some time last week with a few of my projects and going through them with a specific eye for jQuery failures I found that for the most part there wasn't a big issue. The vast majority of code continues to run just fine with either 1.9 or 1.10 (which are supposed to be in sync but with 1.10 removing support for legacy Internet Explorer pre-9.0 versions). However, one particular change in the new versions has caused me quite a bit of update trouble, is the removal of the jQuery.live() function. This is my own fault I suppose - .live() has been deprecated for a while, but with 1.9 and later it was finally removed altogether from jQuery. In the past I had quite a bit of jQuery code that used .live() and it's one of the things that's holding back my upgrade process, although I'm slowly cleaning up my code and switching to the .on() function as the replacement. jQuery.live() jQuery.live() was introduced a long time ago to simplify handling events on matched elements that exist currently on the document and those that are are added in the future and also match the selector. jQuery uses event bubbling, special event binding, plus some magic using meta data attached to a parent level element to check and see if the original target event element matches the selected selected elements (for more info see Elijah Manor's comment below). An Example Assume a list of items like the following in HTML for example and further assume that the items in this list can be appended to at a later point. In this app there's a smallish initial list that loads to start, and as the user scrolls towards the end of the initial small list more items are loaded dynamically and added to the list.<div id="PostItemContainer" class="scrollbox"> <div class="postitem" data-id="4z6qhomm"> <div class="post-icon"></div> <div class="postitemheader"><a href="show/4z6qhomm" target="Content">1999 Buick Century For Sale!</a></div> <div class="postitemprice rightalign">$ 3,500 O.B.O.</div> <div class="smalltext leftalign">Jun. 07 @ 1:06am</div> <div class="post-byline">- Vehicles - Automobiles</div> </div> <div class="postitem" data-id="2jtvuu17"> <div class="postitemheader"><a href="show/2jtvuu17" target="Content">Toyota VAN 1987</a></div> <div class="postitemprice rightalign">$950</div> <div class="smalltext leftalign">Jun. 07 @ 12:29am</div> <div class="post-byline">- Vehicles - Automobiles</div> </div> … </div> With the jQuery.live() function you could easily select elements and hook up a click handler like this:$(".postitem").live("click", function() {...}); Simple and perfectly readable. The behavior of the .live handler generally was the same as the corresponding simple event handlers like .click(), except that you have to explicitly name the event instead of using one of the methods. Re-writing with jQuery.on() With .live() removed in 1.9 and later we have to re-write .live() code above with an alternative. The jQuery documentation points you at the .on() or .delegate() functions to update your code. jQuery.on() is a more generic event handler function, and it's what jQuery uses internally to map the high level event functions like .click(),.change() etc. that jQuery exposes. Using jQuery.on() however is not a one to one replacement of the .live() function. While .on() can handle events directly and use the same syntax as .live() did, you'll find if you simply switch out .live() with .on() that events on not-yet existing elements will not fire. IOW, the key feature of .live() is not working. You can use .on() to get the desired effect however, but you have to change the syntax to explicitly handle the event you're interested in on the container and then provide a filter selector to specify which elements you are actually interested in for handling the event for. Sounds more complicated than it is and it's easier to see with an example. For the list above hooking .postitem clicks, using jQuery.on() looks like this:$("#PostItemContainer").on("click", ".postitem", function() {...}); You specify a container that can handle the .click event and then provide a filter selector to find the child elements that trigger the  the actual event. So here #PostItemContainer contains many .postitems, whose click events I want to handle. Any container will do including document, but I tend to use the container closest to the elements I actually want to handle the events on to minimize the event bubbling that occurs to capture the event. With this code I get the same behavior as with .live() and now as new .postitem elements are added the click events are always available. Sweet. Here's the full event signature for the .on() function: .on( events [, selector ] [, data ], handler(eventObject) ) Note that the selector is optional - if you omit it you essentially create a simple event handler that handles the event directly on the selected object. The filter/child selector required if you want life-like - uh, .live() like behavior to happen. While it's a bit more verbose than what .live() did, .on() provides the same functionality by being more explicit on what your parent container for trapping events is. .on() is good Practice even for ordinary static Element Lists As a side note, it's a good practice to use jQuery.on() or jQuery.delegate() for events in most cases anyway, using this 'container event trapping' syntax. That's because rather than requiring lots of event handlers on each of the child elements (.postitem in the sample above), there's just one event handler on the container, and only when clicked does jQuery drill down to find the matching filter element and tries to match it to the originating element. In the early days of jQuery I used manually build handlers that did this and manually drilled from the event object into the originalTarget to determine if it's a matching element. With later versions of jQuery the various event functions in jQuery essentially provide this functionality out of the box with functions like .on() and .delegate(). All of this is nothing new, but I thought I'd write this up because I have on a few occasions forgotten what exactly was needed to replace the many .live() function calls that litter my code - especially older code. This will be a nice reminder next time I have a memory blank on this topic. And maybe along the way I've helped one or two of you as well to clean up your .live() code…© Rick Strahl, West Wind Technologies, 2005-2013Posted in jQuery   Tweet !function(d,s,id){var js,fjs=d.getElementsByTagName(s)[0];if(!d.getElementById(id)){js=d.createElement(s);js.id=id;js.src="//platform.twitter.com/widgets.js";fjs.parentNode.insertBefore(js,fjs);}}(document,"script","twitter-wjs"); (function() { var po = document.createElement('script'); po.type = 'text/javascript'; po.async = true; po.src = 'https://apis.google.com/js/plusone.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(po, s); })();

    Read the article

  • Extending NerdDinner: Adding Geolocated Flair

    - by Jon Galloway
    NerdDinner is a website with the audacious goal of “Organizing the world’s nerds and helping them eat in packs.” Because nerds aren’t likely to socialize with others unless a website tells them to do it. Scott Hanselman showed off a lot of the cool features we’ve added to NerdDinner lately during his popular talk at MIX10, Beyond File | New Company: From Cheesy Sample to Social Platform. Did you miss it? Go ahead and watch it, I’ll wait. One of the features we wanted to add was flair. You know about flair, right? It’s a way to let folks who like your site show it off in their own site. For example, here’s my StackOverflow flair: Great! So how could we add some of this flair stuff to NerdDinner? What do we want to show? If we’re going to encourage our users to give up a bit of their beautiful website to show off a bit of ours, we need to think about what they’ll want to show. For instance, my StackOverflow flair is all about me, not StackOverflow. So how will this apply to NerdDinner? Since NerdDinner is all about organizing local dinners, in order for the flair to be useful it needs to make sense for the person viewing the web page. If someone visits from Egypt visits my blog, they should see information about NerdDinners in Egypt. That’s geolocation – localizing site content based on where the browser’s sitting, and it makes sense for flair as well as entire websites. So we’ll set up a simple little callout that prompts them to host a dinner in their area: Hopefully our flair works and there is a dinner near your viewers, so they’ll see another view which lists upcoming dinners near them: The Geolocation Part Generally website geolocation is done by mapping the requestor’s IP address to a geographic area. It’s not an exact science, but I’ve always found it to be pretty accurate. There are (at least) three ways to handle it: You pay somebody like MaxMind for a database (with regular updates) that sits on your server, and you use their API to do lookups. I used this on a pretty big project a few years ago and it worked well. You use HTML 5 Geolocation API or Google Gears or some other browser based solution. I think those are cool (I use Google Gears a lot), but they’re both in flux right now and I don’t think either has a wide enough of an install base yet to rely on them. You might want to, but I’ve heard you do all kinds of crazy stuff, and sometimes it gets you in trouble. I don’t mean talk out of line, but we all laugh behind your back a bit. But, hey, it’s up to you. It’s your flair or whatever. There are some free webservices out there that will take an IP address and give you location information. Easy, and works for everyone. That’s what we’re doing. I looked at a few different services and settled on IPInfoDB. It’s free, has a great API, and even returns JSON, which is handy for Javascript use. The IP query is pretty simple. We hit a URL like this: http://ipinfodb.com/ip_query.php?ip=74.125.45.100&timezone=false … and we get an XML response back like this… <?xml version="1.0" encoding="UTF-8"?> <Response> <Ip>74.125.45.100</Ip> <Status>OK</Status> <CountryCode>US</CountryCode> <CountryName>United States</CountryName> <RegionCode>06</RegionCode> <RegionName>California</RegionName> <City>Mountain View</City> <ZipPostalCode>94043</ZipPostalCode> <Latitude>37.4192</Latitude> <Longitude>-122.057</Longitude> </Response> So we’ll build some data transfer classes to hold the location information, like this: public class LocationInfo { public string Country { get; set; } public string RegionName { get; set; } public string City { get; set; } public string ZipPostalCode { get; set; } public LatLong Position { get; set; } } public class LatLong { public float Lat { get; set; } public float Long { get; set; } } And now hitting the service is pretty simple: public static LocationInfo HostIpToPlaceName(string ip) { string url = "http://ipinfodb.com/ip_query.php?ip={0}&timezone=false"; url = String.Format(url, ip); var result = XDocument.Load(url); var location = (from x in result.Descendants("Response") select new LocationInfo { City = (string)x.Element("City"), RegionName = (string)x.Element("RegionName"), Country = (string)x.Element("CountryName"), ZipPostalCode = (string)x.Element("CountryName"), Position = new LatLong { Lat = (float)x.Element("Latitude"), Long = (float)x.Element("Longitude") } }).First(); return location; } Getting The User’s IP Okay, but first we need the end user’s IP, and you’d think it would be as simple as reading the value from HttpContext: HttpContext.Current.Request.UserHostAddress But you’d be wrong. Sorry. UserHostAddress just wraps HttpContext.Current.Request.ServerVariables["REMOTE_ADDR"], but that doesn’t get you the IP for users behind a proxy. That’s in another header, “HTTP_X_FORWARDED_FOR". So you can either hit a wrapper and then check a header, or just check two headers. I went for uniformity: string SourceIP = string.IsNullOrEmpty(Request.ServerVariables["HTTP_X_FORWARDED_FOR"]) ? Request.ServerVariables["REMOTE_ADDR"] : Request.ServerVariables["HTTP_X_FORWARDED_FOR"]; We’re almost set to wrap this up, but first let’s talk about our views. Yes, views, because we’ll have two. Selecting the View We wanted to make it easy for people to include the flair in their sites, so we looked around at how other people were doing this. The StackOverflow folks have a pretty good flair system, which allows you to include the flair in your site as either an IFRAME reference or a Javascript include. We’ll do both. We have a ServicesController to handle use of the site information outside of NerdDinner.com, so this fits in pretty well there. We’ll be displaying the same information for both HTML and Javascript flair, so we can use one Flair controller action which will return a different view depending on the requested format. Here’s our general flow for our controller action: Get the user’s IP Translate it to a location Grab the top three upcoming dinners that are near that location Select the view based on the format (defaulted to “html”) Return a FlairViewModel which contains the list of dinners and the location information public ActionResult Flair(string format = "html") { string SourceIP = string.IsNullOrEmpty( Request.ServerVariables["HTTP_X_FORWARDED_FOR"]) ? Request.ServerVariables["REMOTE_ADDR"] : Request.ServerVariables["HTTP_X_FORWARDED_FOR"]; var location = GeolocationService.HostIpToPlaceName(SourceIP); var dinners = dinnerRepository. FindByLocation(location.Position.Lat, location.Position.Long). OrderByDescending(p => p.EventDate).Take(3); // Select the view we'll return. // Using a switch because we'll add in JSON and other formats later. string view; switch (format.ToLower()) { case "javascript": view = "JavascriptFlair"; break; default: view = "Flair"; break; } return View( view, new FlairViewModel { Dinners = dinners.ToList(), LocationName = string.IsNullOrEmpty(location.City) ? "you" : String.Format("{0}, {1}", location.City, location.RegionName) } ); } Note: I’m not in love with the logic here, but it seems like overkill to extract the switch statement away when we’ll probably just have two or three views. What do you think? The HTML View The HTML version of the view is pretty simple – the only thing of any real interest here is the use of an extension method to truncate strings that are would cause the titles to wrap. public static string Truncate(this string s, int maxLength) { if (string.IsNullOrEmpty(s) || maxLength <= 0) return string.Empty; else if (s.Length > maxLength) return s.Substring(0, maxLength) + "..."; else return s; }   So here’s how the HTML view ends up looking: <%@ Page Title="" Language="C#" Inherits="System.Web.Mvc.ViewPage<FlairViewModel>" %> <%@ Import Namespace="NerdDinner.Helpers" %> <%@ Import Namespace="NerdDinner.Models" %> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head> <title>Nerd Dinner</title> <link href="/Content/Flair.css" rel="stylesheet" type="text/css" /> </head> <body> <div id="nd-wrapper"> <h2 id="nd-header">NerdDinner.com</h2> <div id="nd-outer"> <% if (Model.Dinners.Count == 0) { %> <div id="nd-bummer"> Looks like there's no Nerd Dinners near <%:Model.LocationName %> in the near future. Why not <a target="_blank" href="http://www.nerddinner.com/Dinners/Create">host one</a>?</div> <% } else { %> <h3> Dinners Near You</h3> <ul> <% foreach (var item in Model.Dinners) { %> <li> <%: Html.ActionLink(String.Format("{0} with {1} on {2}", item.Title.Truncate(20), item.HostedBy, item.EventDate.ToShortDateString()), "Details", "Dinners", new { id = item.DinnerID }, new { target = "_blank" })%></li> <% } %> </ul> <% } %> <div id="nd-footer"> More dinners and fun at <a target="_blank" href="http://nrddnr.com">http://nrddnr.com</a></div> </div> </div> </body> </html> You’d include this in a page using an IFRAME, like this: <IFRAME height=230 marginHeight=0 src="http://nerddinner.com/services/flair" frameBorder=0 width=160 marginWidth=0 scrolling=no></IFRAME> The Javascript view The Javascript flair is written so you can include it in a webpage with a simple script include, like this: <script type="text/javascript" src="http://nerddinner.com/services/flair?format=javascript"></script> The goal of this view is very similar to the HTML embed view, with a few exceptions: We’re creating a script element and adding it to the head of the document, which will then document.write out the content. Note that you have to consider if your users will actually have a <head> element in their documents, but for website flair use cases I think that’s a safe bet. Since the content is being added to the existing page rather than shown in an IFRAME, all links need to be absolute. That means we can’t use Html.ActionLink, since it generates relative routes. We need to escape everything since it’s being written out as strings. We need to set the content type to application/x-javascript. The easiest way to do that is to use the <%@ Page ContentType%> directive. <%@ Page Language="C#" Inherits="System.Web.Mvc.ViewPage<NerdDinner.Models.FlairViewModel>" ContentType="application/x-javascript" %> <%@ Import Namespace="NerdDinner.Helpers" %> <%@ Import Namespace="NerdDinner.Models" %> document.write('<script>var link = document.createElement(\"link\");link.href = \"http://nerddinner.com/content/Flair.css\";link.rel = \"stylesheet\";link.type = \"text/css\";var head = document.getElementsByTagName(\"head\")[0];head.appendChild(link);</script>'); document.write('<div id=\"nd-wrapper\"><h2 id=\"nd-header\">NerdDinner.com</h2><div id=\"nd-outer\">'); <% if (Model.Dinners.Count == 0) { %> document.write('<div id=\"nd-bummer\">Looks like there\'s no Nerd Dinners near <%:Model.LocationName %> in the near future. Why not <a target=\"_blank\" href=\"http://www.nerddinner.com/Dinners/Create\">host one</a>?</div>'); <% } else { %> document.write('<h3> Dinners Near You</h3><ul>'); <% foreach (var item in Model.Dinners) { %> document.write('<li><a target=\"_blank\" href=\"http://nrddnr.com/<%: item.DinnerID %>\"><%: item.Title.Truncate(20) %> with <%: item.HostedBy %> on <%: item.EventDate.ToShortDateString() %></a></li>'); <% } %> document.write('</ul>'); <% } %> document.write('<div id=\"nd-footer\"> More dinners and fun at <a target=\"_blank\" href=\"http://nrddnr.com\">http://nrddnr.com</a></div></div></div>'); Getting IP’s for Testing There are a variety of online services that will translate a location to an IP, which were handy for testing these out. I found http://www.itouchmap.com/latlong.html to be most useful, but I’m open to suggestions if you know of something better. Next steps I think the next step here is to minimize load – you know, in case people start actually using this flair. There are two places to think about – the NerdDinner.com servers, and the services we’re using for Geolocation. I usually think about caching as a first attack on server load, but that’s less helpful here since every user will have a different IP. Instead, I’d look at taking advantage of Asynchronous Controller Actions, a cool new feature in ASP.NET MVC 2. Async Actions let you call a potentially long-running webservice without tying up a thread on the server while waiting for the response. There’s some good info on that in the MSDN documentation, and Dino Esposito wrote a great article on Asynchronous ASP.NET Pages in the April 2010 issue of MSDN Magazine. But let’s think of the children, shall we? What about ipinfodb.com? Well, they don’t have specific daily limits, but they do throttle you if you put a lot of traffic on them. From their FAQ: We do not have a specific daily limit but queries that are at a rate faster than 2 per second will be put in "queue". If you stay below 2 queries/second everything will be normal. If you go over the limit, you will still get an answer for all queries but they will be slowed down to about 1 per second. This should not affect most users but for high volume websites, you can either use our IP database on your server or we can whitelist your IP for 5$/month (simply use the donate form and leave a comment with your server IP). Good programming practices such as not querying our API for all page views (you can store the data in a cookie or a database) will also help not reaching the limit. So the first step there is to save the geolocalization information in a time-limited cookie, which will allow us to look up the local dinners immediately without having to hit the geolocation service.

    Read the article

  • http-equiv=content-language alternative - the way of specifying document language

    - by tugberk
    Lots of web sites uses following meta tag to specify the default language of the document: <meta http-equiv="content-language" content="es-ES"> When I go to w3c site: http://www.w3.org/TR/2011/WD-html-markup-20110113/meta.http-equiv.content-language.html#meta.http-equiv.content-language I get this: Using the meta element to specify the document-wide default language is obsolete. Consider specifying the language on the root element instead. What is the way of specifying document language now?

    Read the article

  • REST to Objects in C#

    RESTful interfaces for web services are all the rage for many Web 2.0 sites.  If you want to consume these in a very simple fashion, LINQ to XML can do the job pretty easily in C#.  If you go searching for help on this, youll find a lot of incomplete solutions and fairly large toolkits and frameworks (guess how I know this) this quick article is meant to be a no fluff just stuff approach to making this work. POCO Objects Lets assume you have a Model that you want to suck data into from a RESTful web service.  Ideally this is a Plain Old CLR Object, meaning it isnt infected with any persistence or serialization goop.  It might look something like this: public class Entry { public int Id; public int UserId; public DateTime Date; public float Hours; public string Notes; public bool Billable;   public override string ToString() { return String.Format("[{0}] User: {1} Date: {2} Hours: {3} Notes: {4} Billable {5}", Id, UserId, Date, Hours, Notes, Billable); } } Not that this isnt a completely trivial object.  Lets look at the API for the service.  RESTful HTTP Service In this case, its TickSpots API, with the following sample output: <?xml version="1.0" encoding="UTF-8"?> <entries type="array"> <entry> <id type="integer">24</id> <task_id type="integer">14</task_id> <user_id type="integer">3</user_id> <date type="date">2008-03-08</date> <hours type="float">1.00</hours> <notes>Had trouble with tribbles.</notes> <billable>true</billable> # Billable is an attribute inherited from the task <billed>true</billed> # Billed is an attribute to track whether the entry has been invoiced <created_at type="datetime">Tue, 07 Oct 2008 14:46:16 -0400</created_at> <updated_at type="datetime">Tue, 07 Oct 2008 14:46:16 -0400</updated_at> # The following attributes are derived and provided for informational purposes: <user_email>[email protected]</user_email> <task_name>Remove converter assembly</task_name> <sum_hours type="float">2.00</sum_hours> <budget type="float">10.00</budget> <project_name>Realign dilithium crystals</project_name> <client_name>Starfleet Command</client_name> </entry> </entries> Im assuming in this case that I dont necessarily care about all of the data fields the service is returning I just need some of them for my applications purposes.  Thus, you can see there are more elements in the <entry> XML than I have in my Entry class. Get The XML with C# The next step is to get the XML.  The following snippet does the heavy lifting once you pass it the appropriate URL: protected XElement GetResponse(string uri) { var request = WebRequest.Create(uri) as HttpWebRequest; request.UserAgent = ".NET Sample"; request.KeepAlive = false;   request.Timeout = 15 * 1000;   var response = request.GetResponse() as HttpWebResponse;   if (request.HaveResponse == true && response != null) { var reader = new StreamReader(response.GetResponseStream()); return XElement.Parse(reader.ReadToEnd()); } throw new Exception("Error fetching data."); } This is adapted from the Yahoo Developer article on Web Service REST calls.  Once you have the XML, the last step is to get the data back as your POCO. Use LINQ-To-XML to Deserialize POCOs from XML This is done via the following code: public IEnumerable<Entry> List(DateTime startDate, DateTime endDate) { string additionalParameters = String.Format("start_date={0}&end_date={1}", startDate.ToShortDateString(), endDate.ToShortDateString()); string uri = BuildUrl("entries", additionalParameters);   XElement elements = GetResponse(uri);   var entries = from e in elements.Elements() where e.Name.LocalName == "entry" select new Entry { Id = int.Parse(e.Element("id").Value), UserId = int.Parse(e.Element("user_id").Value), Date = DateTime.Parse(e.Element("date").Value), Hours = float.Parse(e.Element("hours").Value), Notes = e.Element("notes").Value, Billable = bool.Parse(e.Element("billable").Value) }; return entries; }   For completeness, heres the BuildUrl method for my TickSpot API wrapper: // Change these to your settings protected const string projectDomain = "DOMAIN.tickspot.com"; private const string authParams = "[email protected]&password=MyTickSpotPassword";   protected string BuildUrl(string apiMethod, string additionalParams) { if (projectDomain.Contains("DOMAIN")) { throw new ApplicationException("You must update your domain in ProjectRepository.cs."); } if (authParams.Contains("MyTickSpotPassword")) { throw new ApplicationException("You must update your email and password in ProjectRepository.cs."); } return string.Format("https://{0}/api/{1}?{2}&{3}", projectDomain, apiMethod, authParams, additionalParams); } Thats it!  Now go forth and consume XML and map it to classes you actually want to work with.  Have fun! Did you know that DotNetSlackers also publishes .net articles written by top known .net Authors? We already have over 80 articles in several categories including Silverlight. Take a look: here.

    Read the article

  • Ranking - Part II

    - by PointsToShare
    © 2011 By: Dov Trietsch. All rights reserved   Ranking Part II In my introduction to ranking I also introduced the Ranking Game. This is actually a much more sophisticated program than the one we need to simply rate an item, but it introduced you to the sophisticated results that you may achieve by a bit of code and accompanying CSS. In this installment, I am going to handle simple rating with 5 stars. The extra sophistication will come in the form of creating new elements in run time. Why do I need this? I like to be able to extend the SharePoint New and Update forms and put the starts in them simply by using the code shown here. We do not even need to go into SPD. We may achieve this simply by adding a content editor web part; more about this in the next installment. I have created a new page – Rank the Author – in which you may praise me in 5 different ways, but not immediately. The ranking mechanism – the 5 stars – has to be created first. To achieve that, click the “Add Element” button on the screen and then proceed in giving me the appropriate number of stars. Now view the source and see how this extra 5 start element was added. Also see how the ranking is achieved. This, obviously, is not any different in principle than what we did in the Ranking game. We create some sophisticated HTML, Add some style and create the element by: var divString = "<div id="rateMe" title="Rate Me...">    <a onclick="rateIt(this)" id="_1" title="ehh..." onmouseover="rating(this)" onmouseout="off(this)"></a>    <a onclick="rateIt(this)" id="_2" title="So So" onmouseover="rating(this)" onmouseout="off(this)"></a>    <a onclick="rateIt(this)" id="_3" title="Passable" onmouseover="rating(this)" onmouseout="off(this)"></a>    <a onclick="rateIt(this)" id="_4" title="Not too Bad" onmouseover="rating(this)" onmouseout="off(this)"></a>    <a onclick="rateIt(this)" id="_5" title="Not Bad" onmouseover="rating(this)" onmouseout="off(this)"></a></div>";m = document.createElement("p");m.innerHTML = divString;m.className = "blah";function AddElement(){    y = document.getElementById("Rest");    y.parentNode.insertBefore(m, y);} When you look into the full code, you’ll notice that I have added an empty <div id=”Rest”> into the form. A div element, like p, creates a line break, but the main purpose here was to mark the place above which I wanted to add the stars. Now you may hover over the stars, see how they behave and click on one of them to see that the program can react to your selection. That’s all folks!

    Read the article

  • REST to Objects in C#

    RESTful interfaces for web services are all the rage for many Web 2.0 sites.  If you want to consume these in a very simple fashion, LINQ to XML can do the job pretty easily in C#.  If you go searching for help on this, youll find a lot of incomplete solutions and fairly large toolkits and frameworks (guess how I know this) this quick article is meant to be a no fluff just stuff approach to making this work. POCO Objects Lets assume you have a Model that you want to suck data into from a RESTful web service.  Ideally this is a Plain Old CLR Object, meaning it isnt infected with any persistence or serialization goop.  It might look something like this: public class Entry { public int Id; public int UserId; public DateTime Date; public float Hours; public string Notes; public bool Billable;   public override string ToString() { return String.Format("[{0}] User: {1} Date: {2} Hours: {3} Notes: {4} Billable {5}", Id, UserId, Date, Hours, Notes, Billable); } } Not that this isnt a completely trivial object.  Lets look at the API for the service.  RESTful HTTP Service In this case, its TickSpots API, with the following sample output: <?xml version="1.0" encoding="UTF-8"?> <entries type="array"> <entry> <id type="integer">24</id> <task_id type="integer">14</task_id> <user_id type="integer">3</user_id> <date type="date">2008-03-08</date> <hours type="float">1.00</hours> <notes>Had trouble with tribbles.</notes> <billable>true</billable> # Billable is an attribute inherited from the task <billed>true</billed> # Billed is an attribute to track whether the entry has been invoiced <created_at type="datetime">Tue, 07 Oct 2008 14:46:16 -0400</created_at> <updated_at type="datetime">Tue, 07 Oct 2008 14:46:16 -0400</updated_at> # The following attributes are derived and provided for informational purposes: <user_email>[email protected]</user_email> <task_name>Remove converter assembly</task_name> <sum_hours type="float">2.00</sum_hours> <budget type="float">10.00</budget> <project_name>Realign dilithium crystals</project_name> <client_name>Starfleet Command</client_name> </entry> </entries> Im assuming in this case that I dont necessarily care about all of the data fields the service is returning I just need some of them for my applications purposes.  Thus, you can see there are more elements in the <entry> XML than I have in my Entry class. Get The XML with C# The next step is to get the XML.  The following snippet does the heavy lifting once you pass it the appropriate URL: protected XElement GetResponse(string uri) { var request = WebRequest.Create(uri) as HttpWebRequest; request.UserAgent = ".NET Sample"; request.KeepAlive = false;   request.Timeout = 15 * 1000;   var response = request.GetResponse() as HttpWebResponse;   if (request.HaveResponse == true && response != null) { var reader = new StreamReader(response.GetResponseStream()); return XElement.Parse(reader.ReadToEnd()); } throw new Exception("Error fetching data."); } This is adapted from the Yahoo Developer article on Web Service REST calls.  Once you have the XML, the last step is to get the data back as your POCO. Use LINQ-To-XML to Deserialize POCOs from XML This is done via the following code: public IEnumerable<Entry> List(DateTime startDate, DateTime endDate) { string additionalParameters = String.Format("start_date={0}&end_date={1}", startDate.ToShortDateString(), endDate.ToShortDateString()); string uri = BuildUrl("entries", additionalParameters);   XElement elements = GetResponse(uri);   var entries = from e in elements.Elements() where e.Name.LocalName == "entry" select new Entry { Id = int.Parse(e.Element("id").Value), UserId = int.Parse(e.Element("user_id").Value), Date = DateTime.Parse(e.Element("date").Value), Hours = float.Parse(e.Element("hours").Value), Notes = e.Element("notes").Value, Billable = bool.Parse(e.Element("billable").Value) }; return entries; }   For completeness, heres the BuildUrl method for my TickSpot API wrapper: // Change these to your settings protected const string projectDomain = "DOMAIN.tickspot.com"; private const string authParams = "[email protected]&password=MyTickSpotPassword";   protected string BuildUrl(string apiMethod, string additionalParams) { if (projectDomain.Contains("DOMAIN")) { throw new ApplicationException("You must update your domain in ProjectRepository.cs."); } if (authParams.Contains("MyTickSpotPassword")) { throw new ApplicationException("You must update your email and password in ProjectRepository.cs."); } return string.Format("https://{0}/api/{1}?{2}&{3}", projectDomain, apiMethod, authParams, additionalParams); } Thats it!  Now go forth and consume XML and map it to classes you actually want to work with.  Have fun! Did you know that DotNetSlackers also publishes .net articles written by top known .net Authors? We already have over 80 articles in several categories including Silverlight. Take a look: here.

    Read the article

  • Multi-Part Map Troubleshooting

    - by Michael Stephenson
    Scenario I came across a nice little one with multi-part maps the other day. I had an orchestration where I needed to combine 4 input messages into one output message like in the below table:   Input Messages Output Messages Company Details Member Details Event Message Member Search Member Import   I thought my orchestration was working fine but for some reason when I was trying to send my message it had no content under the root node like below <ns0:ImportMemberChange xmlns:ns0="http://---------------/"></ns0:ImportMemberChange>   My map is displayed in the below picture. I knew that the member search message may not have any elements under it but its root element would always exist. The rest of the messages were expected to be fully populated. I tried a number of different things and testing my map outside of the orchestration it always worked fine. The Eureka Moment The eureka moment came when I was looking at the xslt produced by the map. Even though I'd tried swapping the order of the messages in the input of the map you can see in the below picture that the first part of the processing of the message (with the red circle around it) is doing a for-each over the GetCompanyDetailsResult element within the GetCompanyDetailsResponse message. This is because the processing is driven by the output message format and the first element to output is the OrganisationID which comes from the GetCompanyDetailsResponse message. At this point I could focus my attention on this message as the xslt shows that if this xpath statement doesn’t return the an element from the GetCompanyDetailsResponse message then the whole body of the output message will not be produced and the output from the map would look like the message I was getting. <ns0:ImportMemberChange xmlns:ns0="http://---------------/"></ns0:ImportMemberChange> I was quickly able to prove this in my map test which proved this was a likely candidate for the problem. I revisited the orchestration focusing on the creation of the GetCompanyDetailsResponse message and there was actually a bug in the orchestration which resulted in the message being incorrectly created, once this was fixed everything worked as expected. Conclusion Originally I thought it was a problem with the map itself, and looking online there wasn’t really much in the way of content around troubleshooting for multi-part map problems so I thought I'd write this up. I guess technically it isn't a multi-part map problem, but I spend a good couple of hours the other day thinking it was.

    Read the article

  • Kinds of Keywords

    One of the most important element in your blog or website is keyword. With proper use, this element will surely put you in higher rankings and it will surely generate more traffic. But before anything else, let's take a look at the two kinds of keywords.

    Read the article

  • How to return the output of a recursive function in Clojure

    - by Silanglaya Valerio
    Hi everyone! I'm new to functional languages and clojure, so please bear with me... I'm trying to construct a list of functions, with either random parameters or constants. The function that constructs the list of functions is already working, though it doesn't return the function itself. I verified this using println. Here is the snippet: (def operations (list #(- %1 %2) #(+ %1 %2) #(* %1 %2) #(/ %1 %2))) (def parameters (list \u \v \w \x \y \z)) (def parameterlistcount 6) (def paramcount 2) (def opcount 4) (defn generateFunction "Generates a random function list" ([] (generateFunction 2 4 0.5 0.6 '())) ([pc maxdepth fp pp function] (if (and (> maxdepth 0) (< (rand) fp)) (dotimes [i 2] (println(conj (generateFunction pc (dec maxdepth) fp pp function) {:op (nth operations (rand-int opcount))}))) (if (and (< (rand) pp) (> pc 0)) (do (dec pc) (conj function {:param (nth parameters (rand-int parameterlistcount))})) (conj function {:const (rand-int 100)}))))) Any help will be appreciated, thanks!

    Read the article

  • Making DiveIntoPython3 work in IE8 (fixing a Javascript performance issue)

    - by srid
    I am trying to fix the performance problem with Dive Into Python 3 on IE8. Visit this page in IE8 and, after a few moments, you will see the following popup: I traced down the culprit down to this line in j/dip3.js ... find("tr:nth-child(" + (i+1) + ") td:nth-child(2)"); If I disable it (and return from the function immediately), the "Stop executing this script?" dialog does not appear as the page now loads fairly fast. I am no Javascript/jquery expert, so I ask you fellow developers as to why this query is making IE slow. Is there a fix for it? Edit: you can download the entire webpage (980K) for local viewing/editing.

    Read the article

  • Help finding table cell using locator after tableCellAddress finds the cell

    - by Greg Weinman
    Hi All, I have a JSP page populated by javascript resulting in a typical table element cascade . After population I see the text I want using getTable("tableLister.listTable.1.1") The table also has a class=clsDisplayTableBody. I want to click the check box in column 0 for the row of interest. However, all of these tests fail isElementPresent("//*[text()="+cellContents+"]/../td[0]")) isElementPresent("//table[@class='clsDisplayTableBody']//tr[1]/td[0]")) isElementPresent("css='clsDisplayTableBody' tr:nth-child(1) td:nth-child(0)")) isElementPresent("//xpath=id('tableLister.listTable')/descendant::tr[1]/descendant::td[0]")) isElementPresent("//table[@id='tableLister.listTable']//tbody/tr[1]/td[0]")) isElementPresent("//table[@id='tableLister.listTable']//tr[1]/td[0]")) Is there anything else I could try? Regards, Greg Weinman

    Read the article

  • F# static member

    - by HeMan
    I was working through the PDC 2008 F# videos and seem to run into a problem type StockAnalyzer (lprices, days) = let prices = lprices | Seq.map snd | Seq.take days static member GetAnalyzers(tickers, days) = tickers | Seq.map loadPrices | Seq.map (fun prices - new StockAnalyzer(prices, days)) member s.Return = let lastPrice = prices |> Seq.nth 0 let startPrice = prices |> Seq.nth (days - 1) lastPrice / startPrice - 1. I am getting an error at the static. GetStockPrices.fs(31,6): error FS0010: Unexpected keyword 'static' in binding. Expected incomplete structured construct at or before this point or other token. Does anyone know if they have changed the syntax or can spot what I am doing wrong

    Read the article

  • IE JavaScript Mystery

    - by William Calleja
    I have the following JavaScript function function headerBanner(){ var current = $('.bannerImages img:nth-child('+bannerIndex+')').css('display', 'none'); if(bannerIndex== $('.bannerImages img').size()){ bannerIndex= 1; }else{ bannerIndex= (bannerIndex*1)+1; } var next = $('.bannerImages img:nth-child('+bannerIndex+')').css('display', 'block'); } In every browser on the planet, with the exception of IE (8, 7 or less), the above code is working correctly. In Internet Explorer it's going through it and having no effect. I've put alerts at every line of the function and they all fire, even in IE, but the banner simply doesn't change. Is there any reason as to why this is so?

    Read the article

  • How to add exception in this jquery code?

    - by metal-gear-solid
    How to add exception in this jquery code? $(function() { $("table tr:nth-child(even)").addClass("striped"); }); this code is applying on all tables. but for specfic pages i don't want strip effect. I've differnt body id on each page. I want to know how to add exception for a id. $(function() { $("table tr:nth-child(even)").addClass("striped"); //I want to add exception to not to add striped class to only to page with <body id="nostrip"> });

    Read the article

  • How do I make Pseudo classes work with Internet Explorer 7/8?

    - by Mel
    I've written the following code to create a three-column layout where the first and last columns have no left and right margins respectively: #content { background-color:#edeff4; margin:0 auto 30px auto; padding:0 30px 30px 30px; width:900px; } .column { float:left; margin:0 20px; } #content .column:nth-child(1) { margin-left:0; } #content .column:nth-child(3) { margin-right:0; } The problem is that this code does not work in Internet Explorer 7 and 8? The only pseudo class I can use with IE (in this case) would be "first-child," but this does not eliminate the right margin on the third and last column. Does anyone know of a way I can make this code work on IE 7/8?

    Read the article

  • tablerows hover with table even/odd styled via CSS

    - by root
    even/odd styles set on tablerows but hovering of table rows don't work . CSS style : #table_even tr:hover { background-color:#fffbae!important; } /* hovering */ #table_even tr:nth-child(odd) td { background-color:#fbfbfb } /*odd*/ #table_even tr:nth-child(even) td { background-color:#e8ecee } /* even*/ HTML Codes: <table id="table_even" style="width: 100%"> <tr> <td>##</td> <td>##</td> </tr> <tr> <td>##</td> <td>##</td> </tr> <tr> <td>##</td> <td>##</td> </tr> </table> how can be solve ?

    Read the article

  • Button to add value to array and then compare result with another array on IPhone.

    - by slickplaid
    I have ten buttons that each correspond to a different number. I'm looking to record the order that these buttons are pressed and enter them into an array and then compare it to another array that is static in the app. The check should be done on the Nth button press, where N equals the number of items in the other, static array. How do I A) have a button press add a value to an array B) stop after Nth button press equal to amount of numbers in the array I'm checking against and C) compare this array to the other array?

    Read the article

  • .focus() doesn't work on an input while orher attributes works

    - by Sirber
    I have a classic table / thead / tbody structure, which I add a line at the end of the tbody. The line contains only an input element. The code works in Firefox 3.6 but not in Chrome v5 or IE8. I'm using jQuery 1.4.1. Does not work: $("#" + AJAX_ID).parent().find('tr:last > td:nth-child(2) > input').focus(); Does work: $("#" + AJAX_ID).parent().find('tr:last > td:nth-child(2) > input').css('background-color', 'red'); even setting an ID on the input, and using document.getElementBuId('id').focus() doesn't work. Thank you!

    Read the article

< Previous Page | 101 102 103 104 105 106 107 108 109 110 111 112  | Next Page >