Search Results

Search found 28873 results on 1155 pages for 'simple injector'.

Page 10/1155 | < Previous Page | 6 7 8 9 10 11 12 13 14 15 16 17  | Next Page >

  • BigQuery: Simple example of a data collection and analysis pipeline + Your questions

    BigQuery: Simple example of a data collection and analysis pipeline + Your questions Join Michael Manoochehri and Ryan Boyd live to talk about Google BigQuery. We'll give an overview of how we're using our cars, phones, App Engine and BigQuery to collect and analyze data. We'll be discussing our trusted tester feature which allows analyzing data from the App Engine datastore. We'll also review some of the more interesting questions from Stack Overflow and take questions via Google Moderator. From: GoogleDevelopers Views: 250 16 ratings Time: 26:53 More in Science & Technology

    Read the article

  • Looking for feedback on design pattern for simple 2D environment

    - by Le Mot Juiced
    I'm working in iOS. I am trying to make a very simple 2D environment where there are some basic shapes you can drag around with your finger. These shapes should interact in various ways when dropped on each other, or when single-tapped versus double-tapped, etc. I don't know the name for the design pattern I'm thinking of. Basically, you have a bunch of arrays named after attributes, such as "double-tappable" or "draggable" or "stackable". You assign these attributes to the shapes by putting the shapes in the arrays. So, if there's a double-tap event, the code gets the location of it, then iterates through the "double-tappable" array to see if any of its members are in that location. And so on: every interactive event causes a scan through the appropriate array or arrays. It seems like that should work, but I'm wondering if there's a better pattern for the purpose.

    Read the article

  • Application of LGPL license on a simple algorithm

    - by georgesl
    The "scope" of the GNU license is troubling me : I know it has been answered many times ( here, here, ... ) but shouldn't we take into consideration the complexity and originality of a code before using GPL license ? I explain : I'm working on a pet project using the DTW algorithm that I have written in C using the pseudo-code given on the wikipedia page . At one point I decided to change it for a C++ implementation ( just for hone my c++ skill ) . After doing so, I've looked for an existing implementation on the web, to compare the "cleanliness" of it, and I found this one : Vectored DTW implementation, which is part of limproved, a C++ library licensed under GPL v3 . Personnally, I don't mind the GNU license because it is a personnal project, which will never led to any kind of commercial purpose, but I wonder if this implementation can abide a company using it to open their code ( and other FOSS permissions ). Theoretically, I think it can ( I may be wrong :p ), but the algorithm in question is so simple (and old) that it should not.

    Read the article

  • Java - learning with a simple project

    - by phodu_insaan
    I remember when I learnt PHP, it was suggested that I code a simple blog or a forum after reading the language fundamentals and syntax. I was told/read that this would cover everything that I would learn about PHP in a beginners book. This advice was out there in a number of places, and after following and working with PHP it seems quite good advice. I am now learning Java, and reading the book Thinking in Java by Bruce Eckel. I wonder if there is any such set of similar, small projects that I could take up, that would cover all the essentials and most of what is covered in the book. Thanks.

    Read the article

  • A simple Dynamic Proxy

    - by Abhijeet Patel
    Frameworks such as EF4 and MOQ do what most developers consider "dark magic". For instance in EF4, when you use a POCO for an entity you can opt-in to get behaviors such as "lazy-loading" and "change tracking" at runtime merely by ensuring that your type has the following characteristics: The class must be public and not sealed. The class must have a public or protected parameter-less constructor. The class must have public or protected properties Adhere to this and your type is magically endowed with these behaviors without any additional programming on your part. Behind the scenes the framework subclasses your type at runtime and creates a "dynamic proxy" which has these additional behaviors and when you navigate properties of your POCO, the framework replaces the POCO type with derived type instances. The MOQ framework does simlar magic. Let's say you have a simple interface:   public interface IFoo      {          int GetNum();      }   We can verify that the GetNum() was invoked on a mock like so:   var mock = new Mock<IFoo>(MockBehavior.Default);   mock.Setup(f => f.GetNum());   var num = mock.Object.GetNum();   mock.Verify(f => f.GetNum());   Beind the scenes the MOQ framework is generating a dynamic proxy by implementing IFoo at runtime. the call to moq.Object returns the dynamic proxy on which we then call "GetNum" and then verify that this method was invoked. No dark magic at all, just clever programming is what's going on here, just not visible and hence appears magical! Let's create a simple dynamic proxy generator which accepts an interface type and dynamically creates a proxy implementing the interface type specified at runtime.     public static class DynamicProxyGenerator   {       public static T GetInstanceFor<T>()       {           Type typeOfT = typeof(T);           var methodInfos = typeOfT.GetMethods();           AssemblyName assName = new AssemblyName("testAssembly");           var assBuilder = AppDomain.CurrentDomain.DefineDynamicAssembly(assName, AssemblyBuilderAccess.RunAndSave);           var moduleBuilder = assBuilder.DefineDynamicModule("testModule", "test.dll");           var typeBuilder = moduleBuilder.DefineType(typeOfT.Name + "Proxy", TypeAttributes.Public);              typeBuilder.AddInterfaceImplementation(typeOfT);           var ctorBuilder = typeBuilder.DefineConstructor(                     MethodAttributes.Public,                     CallingConventions.Standard,                     new Type[] { });           var ilGenerator = ctorBuilder.GetILGenerator();           ilGenerator.EmitWriteLine("Creating Proxy instance");           ilGenerator.Emit(OpCodes.Ret);           foreach (var methodInfo in methodInfos)           {               var methodBuilder = typeBuilder.DefineMethod(                   methodInfo.Name,                   MethodAttributes.Public | MethodAttributes.Virtual,                   methodInfo.ReturnType,                   methodInfo.GetParameters().Select(p => p.GetType()).ToArray()                   );               var methodILGen = methodBuilder.GetILGenerator();               methodILGen.EmitWriteLine("I'm a proxy");               if (methodInfo.ReturnType == typeof(void))               {                   methodILGen.Emit(OpCodes.Ret);               }               else               {                   if (methodInfo.ReturnType.IsValueType || methodInfo.ReturnType.IsEnum)                   {                       MethodInfo getMethod = typeof(Activator).GetMethod(/span>"CreateInstance",new Type[]{typeof((Type)});                                               LocalBuilder lb = methodILGen.DeclareLocal(methodInfo.ReturnType);                       methodILGen.Emit(OpCodes.Ldtoken, lb.LocalType);                       methodILGen.Emit(OpCodes.Call, typeofype).GetMethod("GetTypeFromHandle"));  ));                       methodILGen.Emit(OpCodes.Callvirt, getMethod);                       methodILGen.Emit(OpCodes.Unbox_Any, lb.LocalType);                                                              }                 else                   {                       methodILGen.Emit(OpCodes.Ldnull);                   }                   methodILGen.Emit(OpCodes.Ret);               }               typeBuilder.DefineMethodOverride(methodBuilder, methodInfo);           }                     Type constructedType = typeBuilder.CreateType();           var instance = Activator.CreateInstance(constructedType);           return (T)instance;       }   }   Dynamic proxies are created by calling into the following main types: AssemblyBuilder, TypeBuilder, Modulebuilder and ILGenerator. These types enable dynamically creating an assembly and emitting .NET modules and types in that assembly, all using IL instructions. Let's break down the code above a bit and examine it piece by piece                Type typeOfT = typeof(T);              var methodInfos = typeOfT.GetMethods();              AssemblyName assName = new AssemblyName("testAssembly");              var assBuilder = AppDomain.CurrentDomain.DefineDynamicAssembly(assName, AssemblyBuilderAccess.RunAndSave);              var moduleBuilder = assBuilder.DefineDynamicModule("testModule", "test.dll");              var typeBuilder = moduleBuilder.DefineType(typeOfT.Name + "Proxy", TypeAttributes.Public);   We are instructing the runtime to create an assembly caled "test.dll"and in this assembly we then emit a new module called "testModule". We then emit a new type definition of name "typeName"Proxy into this new module. This is the definition for the "dynamic proxy" for type T                 typeBuilder.AddInterfaceImplementation(typeOfT);               var ctorBuilder = typeBuilder.DefineConstructor(                         MethodAttributes.Public,                         CallingConventions.Standard,                         new Type[] { });               var ilGenerator = ctorBuilder.GetILGenerator();               ilGenerator.EmitWriteLine("Creating Proxy instance");               ilGenerator.Emit(OpCodes.Ret);   The newly created type implements type T and defines a default parameterless constructor in which we emit a call to Console.WriteLine. This call is not necessary but we do this so that we can see first hand that when the proxy is constructed, when our default constructor is invoked.   var methodBuilder = typeBuilder.DefineMethod(                      methodInfo.Name,                      MethodAttributes.Public | MethodAttributes.Virtual,                      methodInfo.ReturnType,                      methodInfo.GetParameters().Select(p => p.GetType()).ToArray()                      );   We then iterate over each method declared on type T and add a method definition of the same name into our "dynamic proxy" definition     if (methodInfo.ReturnType == typeof(void))   {       methodILGen.Emit(OpCodes.Ret);   }   If the return type specified in the method declaration of T is void we simply return.     if (methodInfo.ReturnType.IsValueType || methodInfo.ReturnType.IsEnum)   {                               MethodInfo getMethod = typeof(Activator).GetMethod("CreateInstance",                                                         new Type[]{typeof(Type)});                               LocalBuilder lb = methodILGen.DeclareLocal(methodInfo.ReturnType);                                                     methodILGen.Emit(OpCodes.Ldtoken, lb.LocalType);       methodILGen.Emit(OpCodes.Call, typeof(Type).GetMethod("GetTypeFromHandle"));       methodILGen.Emit(OpCodes.Callvirt, getMethod);       methodILGen.Emit(OpCodes.Unbox_Any, lb.LocalType);   }   If the return type in the method declaration of T is either a value type or an enum, then we need to create an instance of the value type and return that instance the caller. In order to accomplish that we need to do the following: 1) Get a handle to the Activator.CreateInstance method 2) Declare a local variable which represents the Type of the return type(i.e the type object of the return type) specified on the method declaration of T(obtained from the MethodInfo) and push this Type object onto the evaluation stack. In reality a RuntimeTypeHandle is what is pushed onto the stack. 3) Invoke the "GetTypeFromHandle" method(a static method in the Type class) passing in the RuntimeTypeHandle pushed onto the stack previously as an argument, the result of this invocation is a Type object (representing the method's return type) which is pushed onto the top of the evaluation stack. 4) Invoke Activator.CreateInstance passing in the Type object from step 3, the result of this invocation is an instance of the value type boxed as a reference type and pushed onto the top of the evaluation stack. 5) Unbox the result and place it into the local variable of the return type defined in step 2   methodILGen.Emit(OpCodes.Ldnull);   If the return type is a reference type then we just load a null onto the evaluation stack   methodILGen.Emit(OpCodes.Ret);   Emit a a return statement to return whatever is on top of the evaluation stack(null or an instance of a value type) back to the caller     Type constructedType = typeBuilder.CreateType();   var instance = Activator.CreateInstance(constructedType);   return (T)instance;   Now that we have a definition of the "dynamic proxy" implementing all the methods declared on T, we can now create an instance of the proxy type and return that out typed as T. The caller can now invoke the generator and request a dynamic proxy for any type T. In our example when the client invokes GetNum() we get back "0". Lets add a new method on the interface called DayOfWeek GetDay()   public interface IFoo      {          int GetNum();          DayOfWeek GetDay();      }   When GetDay() is invoked, the "dynamic proxy" returns "Sunday" since that is the default value for the DayOfWeek enum This is a very trivial example of dynammic proxies, frameworks like MOQ have a way more sophisticated implementation of this paradigm where in you can instruct the framework to create proxies which return specified values for a method implementation.

    Read the article

  • A Simple Collapsible Menu with jQuery

    - by Vincent Maverick Durano
    In this post I'll demonstrate how to make a simple collapsible menu using jQuery. To get started let's go ahead and fire up Visual Studio and create a new WebForm.  Now let's build our menu by adding some div, p and anchor tags. Since I'm using a masterpage then the ASPX mark-up should look something like this:   1: <asp:Content ID="Content2" ContentPlaceHolderID="MainContent" runat="server"> 2: <div id="Menu"> 3: <p>CARS</p> 4: <div class="section"> 5: <a href="#">Car 1</a> 6: <a href="#">Car 2</a> 7: <a href="#">Car 3</a> 8: <a href="#">Car 4</a> 9: </div> 10: <p>BIKES</p> 11: <div class="section"> 12: <a href="#">Bike 1</a> 13: <a href="#">Bike 2</a> 14: <a href="#">Bike 3</a> 15: <a href="#">Bike 4</a> 16: <a href="#">Bike 5</a> 17: <a href="#">Bike 6</a> 18: <a href="#">Bike 7</a> 19: <a href="#">Bike 8</a> 20: </div> 21: <p>COMPUTERS</p> 22: <div class="section"> 23: <a href="#">Computer 1</a> 24: <a href="#">Computer 2</a> 25: <a href="#">Computer 3</a> 26: <a href="#">Computer 4</a> 27: </div> 28: <p>OTHERS</p> 29: <div class="section"> 30: <a href="#">Other 1</a> 31: <a href="#">Other 2</a> 32: <a href="#">Other 3</a> 33: <a href="#">Other 4</a> 34: </div> 35: </div> 36: </asp:Content>   As you can see there's nothing fancy about the mark up above.. Now lets go ahead create a simple CSS to set the look and feel our our Menu. Just for for the simplicity of this demo, add the following CSS below under the <head> section of the page or if you are using master page then add it a the content head. Here's the CSS below:   1: <asp:Content ID="Content1" ContentPlaceHolderID="HeadContent" runat="server"> 2: <style type="text/css"> 3: #Menu{ 4: width:300px; 5: } 6: #Menu > p{ 7: background-color:#104D9E; 8: color:#F5F7FA; 9: margin:0; 10: padding:0; 11: border-bottom-style: solid; 12: border-bottom-width: medium; 13: border-bottom-color:#000000; 14: cursor:pointer; 15: } 16: #Menu .section{ 17: padding-left:5px; 18: background-color:#C0D9FA; 19: } 20: a{ 21: display:block; 22: color:#0A0A07; 23: } 24: </style> 25: </asp:Content>   Now let's add the collapsible effects on our menu using jQuery. To start using jQuery then register the following script at the very top of the <head> section of the page or if you are using master page then add it the very top of  the content head section.   <script type="text/javascript" src="https://ajax.googleapis.com/ajax/libs/jquery/1.4.4/jquery.min.js" ></script>   As you can see I'm using Google AJAX API CDN to host the jQuery file. You can also download the jQuery here and host it in your server if you'd like. Okay here's the the jQuery script below for adding the collapsible effects:   1: <script type="text/javascript"> 2: $(function () { 3: $("a").mouseover(function () { $(this).addClass("highlightRow"); }) 4: .mouseout(function () { $(this).removeClass("highlightRow"); }); 5:   6: $(".section").hide(); 7: $("#Menu > p").click(function () { 8: $(this).next().slideToggle("Slow"); 9: }); 10: }); 11: </script>   Okay to give you a little bit of explaination, at line 3.. what it does is it looks for all the "<a>" anchor elements on the page and attach the mouseover and mouseout event. On mouseover, the highlightRow css class is added to <a> element and on mouse out we remove the css class to revert the style to its default look. at line 6 we will hide all the elements that has a class name set as "section" and if you look at the mark up above it is refering to the <div> elements right after each <p> element. At line 7.. what it does is it looks for a <p> element that is a direct child of the element that has an ID of "Menu" and then attach the click event to toggle the visibilty of the section. Here's how it looks in the page: On Initial Load: After Clicking the Section Header:   That's it! I hope someone find this post usefu!   Technorati Tags: ASP.NET,JQuery,Master Page,JavaScript

    Read the article

  • SimpleInjector - Register a type for all it's interfaces

    - by Karl Cassar
    Is it possible to register a type for all it's implementing interfaces? E.g, I have a: public class Bow : IWeapon { #region IWeapon Members public string Attack() { return "Shooted with a bow"; } #endregion } public class HumanFighter { private readonly IWeapon weapon = null; public HumanFighter(IWeapon weapon) { this.weapon = weapon; } public string Fight() { return this.weapon.Attack(); } } [Test] public void Test2b() { Container container = new Container(); container.RegisterSingle<Bow>(); container.RegisterSingle<HumanFighter>(); // this would match the IWeapon to the Bow, as it // is implemented by Bow var humanFighter1 = container.GetInstance<HumanFighter>(); string s = humanFighter1.Fight(); }

    Read the article

  • Running a simple integration scenario using the Oracle Big Data Connectors on Hadoop/HDFS cluster

    - by hamsun
    Between the elephant ( the tradional image of the Hadoop framework) and the Oracle Iron Man (Big Data..) an english setter could be seen as the link to the right data Data, Data, Data, we are living in a world where data technology based on popular applications , search engines, Webservers, rich sms messages, email clients, weather forecasts and so on, have a predominant role in our life. More and more technologies are used to analyze/track our behavior, try to detect patterns, to propose us "the best/right user experience" from the Google Ad services, to Telco companies or large consumer sites (like Amazon:) ). The more we use all these technologies, the more we generate data, and thus there is a need of huge data marts and specific hardware/software servers (as the Exadata servers) in order to treat/analyze/understand the trends and offer new services to the users. Some of these "data feeds" are raw, unstructured data, and cannot be processed effectively by normal SQL queries. Large scale distributed processing was an emerging infrastructure need and the solution seemed to be the "collocation of compute nodes with the data", which in turn leaded to MapReduce parallel patterns and the development of the Hadoop framework, which is based on MapReduce and a distributed file system (HDFS) that runs on larger clusters of rather inexpensive servers. Several Oracle products are using the distributed / aggregation pattern for data calculation ( Coherence, NoSql, times ten ) so once that you are familiar with one of these technologies, lets says with coherence aggregators, you will find the whole Hadoop, MapReduce concept very similar. Oracle Big Data Appliance is based on the Cloudera Distribution (CDH), and the Oracle Big Data Connectors can be plugged on a Hadoop cluster running the CDH distribution or equivalent Hadoop clusters. In this paper, a "lab like" implementation of this concept is done on a single Linux X64 server, running an Oracle Database 11g Enterprise Edition Release 11.2.0.4.0, and a single node Apache hadoop-1.2.1 HDFS cluster, using the SQL connector for HDFS. The whole setup is fairly simple: Install on a Linux x64 server ( or virtual box appliance) an Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 server Get the Apache Hadoop distribution from: http://mir2.ovh.net/ftp.apache.org/dist/hadoop/common/hadoop-1.2.1. Get the Oracle Big Data Connectors from: http://www.oracle.com/technetwork/bdc/big-data-connectors/downloads/index.html?ssSourceSiteId=ocomen. Check the java version of your Linux server with the command: java -version java version "1.7.0_40" Java(TM) SE Runtime Environment (build 1.7.0_40-b43) Java HotSpot(TM) 64-Bit Server VM (build 24.0-b56, mixed mode) Decompress the hadoop hadoop-1.2.1.tar.gz file to /u01/hadoop-1.2.1 Modify your .bash_profile export HADOOP_HOME=/u01/hadoop-1.2.1 export PATH=$PATH:$HADOOP_HOME/bin export HIVE_HOME=/u01/hive-0.11.0 export PATH=$PATH:$HADOOP_HOME/bin:$HIVE_HOME/bin (also see my sample .bash_profile) Set up ssh trust for Hadoop process, this is a mandatory step, in our case we have to establish a "local trust" as will are using a single node configuration copy the new public keys to the list of authorized keys connect and test the ssh setup to your localhost: We will run a "pseudo-Hadoop cluster", in what is called "local standalone mode", all the Hadoop java components are running in one Java process, this is enough for our demo purposes. We need to "fine tune" some Hadoop configuration files, we have to go at our $HADOOP_HOME/conf, and modify the files: core-site.xml hdfs-site.xml mapred-site.xml check that the hadoop binaries are referenced correctly from the command line by executing: hadoop -version As Hadoop is managing our "clustered HDFS" file system we have to create "the mount point" and format it , the mount point will be declared to core-site.xml as: The layout under the /u01/hadoop-1.2.1/data will be created and used by other hadoop components (MapReduce = /mapred/...) HDFS is using the /dfs/... layout structure format the HDFS hadoop file system: Start the java components for the HDFS system As an additional check, you can use the GUI Hadoop browsers to check the content of your HDFS configurations: Once our HDFS Hadoop setup is done you can use the HDFS file system to store data ( big data : )), and plug them back and forth to Oracle Databases by the means of the Big Data Connectors ( which is the next configuration step). You can create / use a Hive db, but in our case we will make a simple integration of "raw data" , through the creation of an External Table to a local Oracle instance ( on the same Linux box, we run the Hadoop HDFS one node cluster and one Oracle DB). Download some public "big data", I use the site: http://france.meteofrance.com/france/observations, from where I can get *.csv files for my big data simulations :). Here is the data layout of my example file: Download the Big Data Connector from the OTN (oraosch-2.2.0.zip), unzip it to your local file system (see picture below) Modify your environment in order to access the connector libraries , and make the following test: [oracle@dg1 bin]$./hdfs_stream Usage: hdfs_stream locationFile [oracle@dg1 bin]$ Load the data to the Hadoop hdfs file system: hadoop fs -mkdir bgtest_data hadoop fs -put obsFrance.txt bgtest_data/obsFrance.txt hadoop fs -ls /user/oracle/bgtest_data/obsFrance.txt [oracle@dg1 bg-data-raw]$ hadoop fs -ls /user/oracle/bgtest_data/obsFrance.txt Found 1 items -rw-r--r-- 1 oracle supergroup 54103 2013-10-22 06:10 /user/oracle/bgtest_data/obsFrance.txt [oracle@dg1 bg-data-raw]$hadoop fs -ls hdfs:///user/oracle/bgtest_data/obsFrance.txt Found 1 items -rw-r--r-- 1 oracle supergroup 54103 2013-10-22 06:10 /user/oracle/bgtest_data/obsFrance.txt Check the content of the HDFS with the browser UI: Start the Oracle database, and run the following script in order to create the Oracle database user, the Oracle directories for the Oracle Big Data Connector (dg1 it’s my own db id replace accordingly yours): #!/bin/bash export ORAENV_ASK=NO export ORACLE_SID=dg1 . oraenv sqlplus /nolog <<EOF CONNECT / AS sysdba; CREATE OR REPLACE DIRECTORY osch_bin_path AS '/u01/orahdfs-2.2.0/bin'; CREATE USER BGUSER IDENTIFIED BY oracle; GRANT CREATE SESSION, CREATE TABLE TO BGUSER; GRANT EXECUTE ON sys.utl_file TO BGUSER; GRANT READ, EXECUTE ON DIRECTORY osch_bin_path TO BGUSER; CREATE OR REPLACE DIRECTORY BGT_LOG_DIR as '/u01/BG_TEST/logs'; GRANT READ, WRITE ON DIRECTORY BGT_LOG_DIR to BGUSER; CREATE OR REPLACE DIRECTORY BGT_DATA_DIR as '/u01/BG_TEST/data'; GRANT READ, WRITE ON DIRECTORY BGT_DATA_DIR to BGUSER; EOF Put the following in a file named t3.sh and make it executable, hadoop jar $OSCH_HOME/jlib/orahdfs.jar \ oracle.hadoop.exttab.ExternalTable \ -D oracle.hadoop.exttab.tableName=BGTEST_DP_XTAB \ -D oracle.hadoop.exttab.defaultDirectory=BGT_DATA_DIR \ -D oracle.hadoop.exttab.dataPaths="hdfs:///user/oracle/bgtest_data/obsFrance.txt" \ -D oracle.hadoop.exttab.columnCount=7 \ -D oracle.hadoop.connection.url=jdbc:oracle:thin:@//localhost:1521/dg1 \ -D oracle.hadoop.connection.user=BGUSER \ -D oracle.hadoop.exttab.printStackTrace=true \ -createTable --noexecute then test the creation fo the external table with it: [oracle@dg1 samples]$ ./t3.sh ./t3.sh: line 2: /u01/orahdfs-2.2.0: Is a directory Oracle SQL Connector for HDFS Release 2.2.0 - Production Copyright (c) 2011, 2013, Oracle and/or its affiliates. All rights reserved. Enter Database Password:] The create table command was not executed. The following table would be created. CREATE TABLE "BGUSER"."BGTEST_DP_XTAB" ( "C1" VARCHAR2(4000), "C2" VARCHAR2(4000), "C3" VARCHAR2(4000), "C4" VARCHAR2(4000), "C5" VARCHAR2(4000), "C6" VARCHAR2(4000), "C7" VARCHAR2(4000) ) ORGANIZATION EXTERNAL ( TYPE ORACLE_LOADER DEFAULT DIRECTORY "BGT_DATA_DIR" ACCESS PARAMETERS ( RECORDS DELIMITED BY 0X'0A' CHARACTERSET AL32UTF8 STRING SIZES ARE IN CHARACTERS PREPROCESSOR "OSCH_BIN_PATH":'hdfs_stream' FIELDS TERMINATED BY 0X'2C' MISSING FIELD VALUES ARE NULL ( "C1" CHAR(4000), "C2" CHAR(4000), "C3" CHAR(4000), "C4" CHAR(4000), "C5" CHAR(4000), "C6" CHAR(4000), "C7" CHAR(4000) ) ) LOCATION ( 'osch-20131022081035-74-1' ) ) PARALLEL REJECT LIMIT UNLIMITED; The following location files would be created. osch-20131022081035-74-1 contains 1 URI, 54103 bytes 54103 hdfs://localhost:19000/user/oracle/bgtest_data/obsFrance.txt Then remove the --noexecute flag and create the external Oracle table for the Hadoop data. Check the results: The create table command succeeded. CREATE TABLE "BGUSER"."BGTEST_DP_XTAB" ( "C1" VARCHAR2(4000), "C2" VARCHAR2(4000), "C3" VARCHAR2(4000), "C4" VARCHAR2(4000), "C5" VARCHAR2(4000), "C6" VARCHAR2(4000), "C7" VARCHAR2(4000) ) ORGANIZATION EXTERNAL ( TYPE ORACLE_LOADER DEFAULT DIRECTORY "BGT_DATA_DIR" ACCESS PARAMETERS ( RECORDS DELIMITED BY 0X'0A' CHARACTERSET AL32UTF8 STRING SIZES ARE IN CHARACTERS PREPROCESSOR "OSCH_BIN_PATH":'hdfs_stream' FIELDS TERMINATED BY 0X'2C' MISSING FIELD VALUES ARE NULL ( "C1" CHAR(4000), "C2" CHAR(4000), "C3" CHAR(4000), "C4" CHAR(4000), "C5" CHAR(4000), "C6" CHAR(4000), "C7" CHAR(4000) ) ) LOCATION ( 'osch-20131022081719-3239-1' ) ) PARALLEL REJECT LIMIT UNLIMITED; The following location files were created. osch-20131022081719-3239-1 contains 1 URI, 54103 bytes 54103 hdfs://localhost:19000/user/oracle/bgtest_data/obsFrance.txt This is the view from the SQL Developer: and finally the number of lines in the oracle table, imported from our Hadoop HDFS cluster SQL select count(*) from "BGUSER"."BGTEST_DP_XTAB"; COUNT(*) ---------- 1151 In a next post we will integrate data from a Hive database, and try some ODI integrations with the ODI Big Data connector. Our simplistic approach is just a step to show you how these unstructured data world can be integrated to Oracle infrastructure. Hadoop, BigData, NoSql are great technologies, they are widely used and Oracle is offering a large integration infrastructure based on these services. Oracle University presents a complete curriculum on all the Oracle related technologies: NoSQL: Introduction to Oracle NoSQL Database Using Oracle NoSQL Database Big Data: Introduction to Big Data Oracle Big Data Essentials Oracle Big Data Overview Oracle Data Integrator: Oracle Data Integrator 12c: New Features Oracle Data Integrator 11g: Integration and Administration Oracle Data Integrator: Administration and Development Oracle Data Integrator 11g: Advanced Integration and Development Oracle Coherence 12c: Oracle Coherence 12c: New Features Oracle Coherence 12c: Share and Manage Data in Clusters Oracle Coherence 12c: Oracle GoldenGate 11g: Fundamentals for Oracle Oracle GoldenGate 11g: Fundamentals for SQL Server Oracle GoldenGate 11g Fundamentals for Oracle Oracle GoldenGate 11g Fundamentals for DB2 Oracle GoldenGate 11g Fundamentals for Teradata Oracle GoldenGate 11g Fundamentals for HP NonStop Oracle GoldenGate 11g Management Pack: Overview Oracle GoldenGate 11g Troubleshooting and Tuning Oracle GoldenGate 11g: Advanced Configuration for Oracle Other Resources: Apache Hadoop : http://hadoop.apache.org/ is the homepage for these technologies. "Hadoop Definitive Guide 3rdEdition" by Tom White is a classical lecture for people who want to know more about Hadoop , and some active "googling " will also give you some more references. About the author: Eugene Simos is based in France and joined Oracle through the BEA-Weblogic Acquisition, where he worked for the Professional Service, Support, end Education for major accounts across the EMEA Region. He worked in the banking sector, ATT, Telco companies giving him extensive experience on production environments. Eugen currently specializes in Oracle Fusion Middleware teaching an array of courses on Weblogic/Webcenter, Content,BPM /SOA/Identity-Security/GoldenGate/Virtualisation/Unified Comm Suite) throughout the EMEA region.

    Read the article

  • SQL SERVER – Simple Installation of Master Data Services (MDS) and Sample Packages – Very Easy

    - by pinaldave
    I twitted recently about: ‘Installing #sql Server 2008 R2 – Master Data Services. Painless.‘ After doing so, I got quite a few emails from other users as to why I thought it was painless. The reason was very simple- I was able to install it rather quickly on my laptop without any issues. There were a few requests along with these emails sent to me, which regards to how to install MDS, as well sample databases. Please note that I am the admin of my machine and I installed this MDS as the admin as well. Talk to your network administrator to figure out the best suitable settings for better security of login users. Additionally, since MDS is only supported on a 64-bit machine, I had rebuilt my computer a week before with a 64-bit OS and 64-bit SQL Server to go with it. Here is a quick picture tour of the installation. First of all, go to your SQL Server 2008. Install self-extracted folder and find the .msi file for C:\1033_enu_lp\x64\setup\masterdataservices.msi. Once you clicked on the file, follow the image tour below. You can ask me any questions in case you are still confused with any of the steps of the installation. While searching the internet for a similar installation process, I have landed on the official blog of MDS team where they have many interesting posts about it. If any concept written on my posts are contradictory to the information on the official blog, I suggest that you should follow the advice of official blog. Reference : Pinal Dave (http://blog.SQLAuthority.com) Filed under: Pinal Dave, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, T SQL, Technology

    Read the article

  • Building Simple Workflows in Oozie

    - by dan.mcclary
    Introduction More often than not, data doesn't come packaged exactly as we'd like it for analysis. Transformation, match-merge operations, and a host of data munging tasks are usually needed before we can extract insights from our Big Data sources. Few people find data munging exciting, but it has to be done. Once we've suffered that boredom, we should take steps to automate the process. We want codify our work into repeatable units and create workflows which we can leverage over and over again without having to write new code. In this article, we'll look at how to use Oozie to create a workflow for the parallel machine learning task I described on Cloudera's site. Hive Actions: Prepping for Pig In my parallel machine learning article, I use data from the National Climatic Data Center to build weather models on a state-by-state basis. NCDC makes the data freely available as gzipped files of day-over-day observations stretching from the 1930s to today. In reading that post, one might get the impression that the data came in a handy, ready-to-model files with convenient delimiters. The truth of it is that I need to perform some parsing and projection on the dataset before it can be modeled. If I get more observations, I'll want to retrain and test those models, which will require more parsing and projection. This is a good opportunity to start building up a workflow with Oozie. I store the data from the NCDC in HDFS and create an external Hive table partitioned by year. This gives me flexibility of Hive's query language when I want it, but let's me put the dataset in a directory of my choosing in case I want to treat the same data with Pig or MapReduce code. CREATE EXTERNAL TABLE IF NOT EXISTS historic_weather(column 1, column2) PARTITIONED BY (yr string) STORED AS ... LOCATION '/user/oracle/weather/historic'; As new weather data comes in from NCDC, I'll need to add partitions to my table. That's an action I should put in the workflow. Similarly, the weather data requires parsing in order to be useful as a set of columns. Because of their long history, the weather data is broken up into fields of specific byte lengths: x bytes for the station ID, y bytes for the dew point, and so on. The delimiting is consistent from year to year, so writing SerDe or a parser for transformation is simple. Once that's done, I want to select columns on which to train, classify certain features, and place the training data in an HDFS directory for my Pig script to access. ALTER TABLE historic_weather ADD IF NOT EXISTS PARTITION (yr='2010') LOCATION '/user/oracle/weather/historic/yr=2011'; INSERT OVERWRITE DIRECTORY '/user/oracle/weather/cleaned_history' SELECT w.stn, w.wban, w.weather_year, w.weather_month, w.weather_day, w.temp, w.dewp, w.weather FROM ( FROM historic_weather SELECT TRANSFORM(...) USING '/path/to/hive/filters/ncdc_parser.py' as stn, wban, weather_year, weather_month, weather_day, temp, dewp, weather ) w; Since I'm going to prepare training directories with at least the same frequency that I add partitions, I should also add that to my workflow. Oozie is going to invoke these Hive actions using what's somewhat obviously referred to as a Hive action. Hive actions amount to Oozie running a script file containing our query language statements, so we can place them in a file called weather_train.hql. Starting Our Workflow Oozie offers two types of jobs: workflows and coordinator jobs. Workflows are straightforward: they define a set of actions to perform as a sequence or directed acyclic graph. Coordinator jobs can take all the same actions of Workflow jobs, but they can be automatically started either periodically or when new data arrives in a specified location. To keep things simple we'll make a workflow job; coordinator jobs simply require another XML file for scheduling. The bare minimum for workflow XML defines a name, a starting point, and an end point: <workflow-app name="WeatherMan" xmlns="uri:oozie:workflow:0.1"> <start to="ParseNCDCData"/> <end name="end"/> </workflow-app> To this we need to add an action, and within that we'll specify the hive parameters Also, keep in mind that actions require <ok> and <error> tags to direct the next action on success or failure. <action name="ParseNCDCData"> <hive xmlns="uri:oozie:hive-action:0.2"> <job-tracker>localhost:8021</job-tracker> <name-node>localhost:8020</name-node> <configuration> <property> <name>oozie.hive.defaults</name> <value>/user/oracle/weather_ooze/hive-default.xml</value> </property> </configuration> <script>ncdc_parse.hql</script> </hive> <ok to="WeatherMan"/> <error to="end"/> </action> There are a couple of things to note here: I have to give the FQDN (or IP) and port of my JobTracker and NameNode. I have to include a hive-default.xml file. I have to include a script file. The hive-default.xml and script file must be stored in HDFS That last point is particularly important. Oozie doesn't make assumptions about where a given workflow is being run. You might submit workflows against different clusters, or have different hive-defaults.xml on different clusters (e.g. MySQL or Postgres-backed metastores). A quick way to ensure that all the assets end up in the right place in HDFS is just to make a working directory locally, build your workflow.xml in it, and copy the assets you'll need to it as you add actions to workflow.xml. At this point, our local directory should contain: workflow.xml hive-defaults.xml (make sure this file contains your metastore connection data) ncdc_parse.hql Adding Pig to the Ooze Adding our Pig script as an action is slightly simpler from an XML standpoint. All we do is add an action to workflow.xml as follows: <action name="WeatherMan"> <pig> <job-tracker>localhost:8021</job-tracker> <name-node>localhost:8020</name-node> <script>weather_train.pig</script> </pig> <ok to="end"/> <error to="end"/> </action> Once we've done this, we'll copy weather_train.pig to our working directory. However, there's a bit of a "gotcha" here. My pig script registers the Weka Jar and a chunk of jython. If those aren't also in HDFS, our action will fail from the outset -- but where do we put them? The Jython script goes into the working directory at the same level as the pig script, because pig attempts to load Jython files in the directory from which the script executes. However, that's not where our Weka jar goes. While Oozie doesn't assume much, it does make an assumption about the Pig classpath. Anything under working_directory/lib gets automatically added to the Pig classpath and no longer requires a REGISTER statement in the script. Anything that uses a REGISTER statement cannot be in the working_directory/lib directory. Instead, it needs to be in a different HDFS directory and attached to the pig action with an <archive> tag. Yes, that's as confusing as you think it is. You can get the exact rules for adding Jars to the distributed cache from Oozie's Pig Cookbook. Making the Workflow Work We've got a workflow defined and have collected all the components we'll need to run. But we can't run anything yet, because we still have to define some properties about the job and submit it to Oozie. We need to start with the job properties, as this is essentially the "request" we'll submit to the Oozie server. In the same working directory, we'll make a file called job.properties as follows: nameNode=hdfs://localhost:8020 jobTracker=localhost:8021 queueName=default weatherRoot=weather_ooze mapreduce.jobtracker.kerberos.principal=foo dfs.namenode.kerberos.principal=foo oozie.libpath=${nameNode}/user/oozie/share/lib oozie.wf.application.path=${nameNode}/user/${user.name}/${weatherRoot} outputDir=weather-ooze While some of the pieces of the properties file are familiar (e.g., JobTracker address), others take a bit of explaining. The first is weatherRoot: this is essentially an environment variable for the script (as are jobTracker and queueName). We're simply using them to simplify the directives for the Oozie job. The oozie.libpath pieces is extremely important. This is a directory in HDFS which holds Oozie's shared libraries: a collection of Jars necessary for invoking Hive, Pig, and other actions. It's a good idea to make sure this has been installed and copied up to HDFS. The last two lines are straightforward: run the application defined by workflow.xml at the application path listed and write the output to the output directory. We're finally ready to submit our job! After all that work we only need to do a few more things: Validate our workflow.xml Copy our working directory to HDFS Submit our job to the Oozie server Run our workflow Let's do them in order. First validate the workflow: oozie validate workflow.xml Next, copy the working directory up to HDFS: hadoop fs -put working_dir /user/oracle/working_dir Now we submit the job to the Oozie server. We need to ensure that we've got the correct URL for the Oozie server, and we need to specify our job.properties file as an argument. oozie job -oozie http://url.to.oozie.server:port_number/ -config /path/to/working_dir/job.properties -submit We've submitted the job, but we don't see any activity on the JobTracker? All I got was this funny bit of output: 14-20120525161321-oozie-oracle This is because submitting a job to Oozie creates an entry for the job and places it in PREP status. What we got back, in essence, is a ticket for our workflow to ride the Oozie train. We're responsible for redeeming our ticket and running the job. oozie -oozie http://url.to.oozie.server:port_number/ -start 14-20120525161321-oozie-oracle Of course, if we really want to run the job from the outset, we can change the "-submit" argument above to "-run." This will prep and run the workflow immediately. Takeaway So, there you have it: the somewhat laborious process of building an Oozie workflow. It's a bit tedious the first time out, but it does present a pair of real benefits to those of us who spend a great deal of time data munging. First, when new data arrives that requires the same processing, we already have the workflow defined and ready to run. Second, as we build up a set of useful action definitions over time, creating new workflows becomes quicker and quicker.

    Read the article

  • C# Simple Twitter Update

    - by mroberts
    For what it's worth a simple twitter update. 1: using System; 2: using System.IO; 3: using System.Net; 4: using System.Text; 5:   6: namespace Server.Actions 7: { 8: public class TwitterUpdate 9: { 10: public string Body { get; set; } 11: public string Login { get; set; } 12: public string Password { get; set; } 13:   14: public override void Execute() 15: { 16: try 17: { 18: //encode user name and password 19: string creds = Convert.ToBase64String(Encoding.ASCII.GetBytes(string.Format("{0}:{1}", this.Login, this.Password))); 20:   21: //encode tweet 22: byte[] tweet = Encoding.ASCII.GetBytes("status=" + this.Body); 23:   24: //setup request 25: HttpWebRequest request = (HttpWebRequest)WebRequest.Create("http://twitter.com/statuses/update.xml"); 26: request.Method = "POST"; 27: request.ServicePoint.Expect100Continue = false; 28: request.Headers.Add("Authorization", string.Format("Basic {0}", creds)); 29: request.ContentType = "application/x-www-form-urlencoded"; 30: request.ContentLength = tweet.Length; 31:   32: //write to stream 33: Stream reqStream = request.GetRequestStream(); 34: reqStream.Write(tweet, 0, tweet.Length); 35: reqStream.Close(); 36:   37: //check response 38: HttpWebResponse response = (HttpWebResponse)request.GetResponse(); 39:   40: //... 41: } 42: catch (Exception e) 43: { 44: //... 45: } 46: } 47: } 48: }   BTW, this is my first blog post.  Nothing earth shattering, I admit, but I needed to figure out how to post formatted code.  In the past I’ve used Alex Gorbatchev’s Syntax Highlighter with great success, but here at GWB I couldn’t get it to work. Windows Live Writer though, being a stand alone writer, worked with no problems.  For now, that’s what I’ll use.

    Read the article

  • iOS 5: Enable Android Style Auto Correction Feature With A Simple Trick

    - by Gopinath
    Apple generally don’t let its users to play with their devices, but seems to be these days there are few things slipping through the nets. Smart users are able find some hacks and enable new features on iOS devices! Few days ago we heard about the hidden panorama feature built into iOS 5 and it could be enabled on a jail broken device. Here come another hidden feature unearthed by a smart geek in iOS 5 : enable Android style auto-correction on on-screen keyboard. Luckily to enable this feature you don’t need to jailbreak, all you need to do is to take backup of your device, edit a file and restore it back. Boom!  That’s it. To enable auto corrections feature on the on-screen keyboard of iOS 5 follow these steps Download iBackupBot and install it on your machine. It’s works on both Windows and Mac OS X. Backup your iPhone, iPod, or iPad with iTunes – plug in your iOS device and sync it. Open iBackupBot, locate your most recent backup and click on it Scroll down to Library/Preferences/com.apple.keyboard.plist and double-click on it.   Replace everything between the two <dict> with the following <key>KeyboardAutocorrectionLists</key> <string>YES</string> Save the plist file, then hit the "Restore From Backup" button in iBackupbot. Reboot your device to see the auto correction feature in action on your device’s on-screen keyboard. via lifehacker This article titled,iOS 5: Enable Android Style Auto Correction Feature With A Simple Trick, was originally published at Tech Dreams. Grab our rss feed or fan us on Facebook to get updates from us.

    Read the article

  • Simple ADF page using BAM Data Control

    - by [email protected]
    var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www."); document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E")); try { var pageTracker = _gat._getTracker("UA-15829414-1"); pageTracker._trackPageview(); } catch(err) {} Purpose : In this blog I will walk you through very simple steps to create an ADF page using BAM data control connection.Details : Create the projectOpen JDeveloper (make sure you have installed the SOA extension for JDev)Create new Application using "Generic Application" template.Click on "Next"Shuttle  "ADF Faces" to right pane for the project technology.Click "Finish"Create a BAM connectionIn the resource palette click on "Folder->New Connection -> BAM"Enter the connection name and click "Next"Enter Connection details Click on "Test connection" and "Finish"Create the BAM Data Control Open the IDE connection created in above step.Drag and drop "Employees" to "Data controls" palette.Select "Flat Query" and Click "Finish".Create the View Create a new JSF page.From Data control Panel drag and drop "Employees->Query->ADF Read Only table"Right click and Run the page.

    Read the article

  • Test a simple multi-player (upto four players) Android game in single developer machine

    - by Kush
    I'm working on a multi-player Android game (very simple it is that it doesn't have any game-engine used). The game is based on Java Socket. Four devices will connect the game server and a new thread will manage their session. The game server will server many such sessions (having 4 players each). What I'm worried about is the testing of this game. I know it is possible to run multiple android emulators, but my development laptop is very limited in capabilities (3 GB RAM, 2 Ghz Intel Core2Duo and on-board Graphics). And I'm already using Ubuntu to develop the game so that I have more user memory available than I'd have with Windows. Hence, the laptop will burn-to-death on running 4 emulator instances. I don't have access to any android device, neither I have another machine with higher configuration. And I still have to develop and test this game. P.S. : I'm a CS student, and currently don't work anywhere, and this game is college project, so if there are any paid solutions, I cannot afford it. What can I do to test the app seamlessly? ability to test even only 4 clients (i.e. only 1 session) would suffice, its alright if I can't simulate real environment with some 10-20 active game sessions (having 4 players each).

    Read the article

  • Great Example of a Simple Cost-Benefit Analysis

    - by BuckWoody
    I saw a post the other day that you should definitely go check out. It’s a cost/benefit decision, and although the author gives it a quick treatment and doesn’t take all points in the decision into account, you should focus on the process he follows. It’s a quick and simple example of the kind of thought process we should have as data professionals when we pick a server, a process, or application and even platform software. The key is to include more than just the price of a piece of software or hardware. You need to think about the “other” costs in the decision, and then make the right one. Sometimes the cheapest option is the cheapest, and other times, well, it isn’t. I’ve seen this played out not only in the decision to go with a certain selection, but in the options or editions it comes in. You have to put all of the decision points in the analysis to come up with the right answer, and you have to be able to explain your logic to your team and your company. This is the way you become a data professional, not just a DBA. You can check out the post here – it deals with Azure, but the point is the process, not Azure itself: http://blogs.msdn.com/eugeniop/archive/2010/03/19/windows-azure-guidance-a-simplistic-economic-analysis-of-a-expense-migration.aspx Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • To encryption=on or encryption=off a simple ZFS Crypto demo

    - by darrenm
    I've just been asked twice this week how I would demonstrate ZFS encryption really is encrypting the data on disk.  It needs to be really simple and the target isn't forensics or cryptanalysis just a quick demo to show the before and after. I usually do this small demo using a pool based on files so I can run strings(1) on the "disks" that make up the pool. The demo will work with real disks too but it will take a lot longer (how much longer depends on the size of your disks).  The file hamlet.txt is this one from gutenberg.org # mkfile 64m /tmp/pool1_file # zpool create clear_pool /tmp/pool1_file # cp hamlet.txt /clear_pool # grep -i hamlet /clear_pool/hamlet.txt | wc -l Note the number of times hamlet appears # zpool export clear_pool # strings /tmp/pool1_file | grep -i hamlet | wc -l Note the number of times hamlet appears on disk - it is 2 more because the file is called hamlet.txt and file names are in the clear as well and we keep at least two copies of metadata. Now lets encrypt the file systems in the pool. Note you MUST use a new pool file don't reuse the one from above. # mkfile 64m /tmp/pool2_file # zpool create -O encryption=on enc_pool /tmp/pool2_file Enter passphrase for 'enc_pool': Enter again: # cp hamlet.txt /enc_pool # grep -i hamlet /enc_pool/hamlet.txt | wc -l Note the number of times hamlet appears is the same as before # zpool export enc_pool # strings /tmp/pool2_file | grep -i hamlet | wc -l Note the word hamlet doesn't appear at all! As a said above this isn't indended as "proof" that ZFS does encryption properly just as a quick to do demo.

    Read the article

  • Better solution then simple factory method when concrete implementations have different attributes

    - by danip
    abstract class Animal { function eat() {..} function sleep() {..} function isSmart() } class Dog extends Animal { public $blnCanBark; function isSmart() { return $this->blnCanBark; } } class Cat extends Animal { public $blnCanJumpHigh; function isSmart() { return $this->blnCanJumpHigh; } } .. and so on up to 10-20 animals. Now I created a factory using simple factory method and try to create instances like this: class AnimalFactory { public static function create($strName) { switch($strName) { case 'Dog': return new Dog(); case 'Cat': return new Cat(); default: break; } } } The problem is I can't set the specific attributes like blnCanBark, blnCanJumpHigh in an efficient way. I can send all of them as extra params to create but this will not scale to more then a few classes. Also I can't break the inheritance because a lot of the basic functionality is the same. Is there a better pattern to solve this?

    Read the article

  • C# Simple Twitter Update

    - by mroberts
    For what it's worth a simple twitter update. using System; using System.IO; using System.Net; using System.Text; namespace Server.Actions { public class TwitterUpdate { public string Body { get; set; } public string Login { get; set; } public string Password { get; set; } public override void Execute() { try { //encode user name and password string creds = Convert.ToBase64String(Encoding.ASCII.GetBytes(string.Format("{0}:{1}", this.Login, this.Password))); //encode tweet byte[] tweet = Encoding.ASCII.GetBytes("status=" + this.Body); //setup request HttpWebRequest request = (HttpWebRequest)WebRequest.Create("http://twitter.com/statuses/update.xml"); request.Method = "POST"; request.ServicePoint.Expect100Continue = false; request.Headers.Add("Authorization", string.Format("Basic {0}", creds)); request.ContentType = "application/x-www-form-urlencoded"; request.ContentLength = tweet.Length; //write to stream Stream reqStream = request.GetRequestStream(); reqStream.Write(tweet, 0, tweet.Length); reqStream.Close(); //check response HttpWebResponse response = (HttpWebResponse)request.GetResponse(); //... } catch (Exception e) { //... } } } }

    Read the article

  • New Book - Oracle ADF Enterprise Application Development Made Simple

    - by Shay Shmeltzer
    It's nice to see another ADF book out there, this one from Sten Vesteli titled "Oracle ADF Enterprise Application Development Made Simple" comes from Packet Publishing Unlike other ADF books out there, this one doesn't aim to teach you Oracle ADF, but rather focuses on the right way to structure and manage a project that leverages ADF. This is a welcomed addition to the bookshelf for people who are looking into ADF based development. One thing I find is that some organization just start developing an ADF application without first doing much planning, something that is understandable given that it is very easy to start building a prototype with ADF and then just grow it into a full blown application. However, as the book points out, doing a bit of planning before you delve into the actual project development can save you a lot of time in the future. For example it is much better to have the right breakdown and structure of your project to allow you to do efficient team development right out of the gate, then to find out 1 year down the road that you are dealing with one monolithic size project which is hard to manage. The book touches on such topics as project organization (workspaces, projects, packages), planning your infrastructure (templates, framework classes), coding standards, team structure, etc. It also covers various aspects of application lifecycle management such as versioning, build, testing, deployment and managing requirements and tasks and how all of those are done when using JDeveloper and Oracle ADF. It's nice to see that the book covers working with Oracle Team Productivity Center - a solution that might not be getting the exposure it deserves. The book also has some chapters about security, internalization and customization of applications both with MDS and with ADF Faces skins (and it even covers the brand new skin editor). Overall I think this is definitely a book you should read if you are about to start your way on a new enterprise scale ADF application. Taking into account the topics that the book discusses before you start your work will save you time and effort down the road. By the way, don't forget that as an OTN member you can get discount on this and other books.

    Read the article

  • Simple collision detection for pong

    - by Dave Voyles
    I'm making a simple pong game, and things are great so far, but I have an odd bug which causes my ball (well, it's a box really) to get stuck on occasion when detecting collision against the ceiling or floor. It looks as though it is trying to update too frequently to get out of the collision check. Basically the box slides against the top or bottom of the screen from one paddle to the other, and quickly bounces on and off the wall while doing so, but only bounces a few pixels from the wall. What can I do to avoid this problem? It seems to occur at random. Below is my collision detection for the wall, as well as my update method for the ball. public void UpdatePosition() { size.X = (int)position.X; size.Y = (int)position.Y; position.X += speed * (float)Math.Cos(direction); position.Y += speed * (float)Math.Sin(direction); CheckWallHit(); } // Checks for collision with the ceiling or floor. // 2*Math.pi = 360 degrees // TODO: Change collision so that ball bounces from wall after getting caught private void CheckWallHit() { while (direction > 2 * Math.PI) { direction -= 2 * Math.PI; } while (direction < 0) { direction += 2 * Math.PI; } if (position.Y <= 0 || (position.Y > resetPos.Y * 2 - size.Height)) { direction = 2 * Math.PI - direction; } }

    Read the article

  • Simple monitoring of a Raspberry Pi powered screen - Part 2

    - by Chris Houston
    If you have read my previous blog post Raspberry Pi entrance signed backed by Umbraco - Part 1 which describes how we used a Raspberry Pi to drive an Entrance sign for QV Offices you will have seen I mentioned a follow up post about monitoring the sign.As the sign is mounted in the entrance of the building on the ground floor and the reception is on the 1st floor, this meant that if there was a fault of any kind showing on the screen, the first person to see this was inevitably one of QV Offices' clients as they walked into the building.Although the QV Offices' team were able to check the Umbraco website address that the sign uses, this did not always mean that everything was working as expected. We noticed a couple of times that the sign had Wifi issues (it is now hard wired) and this caused the Chromium browser to render a 404 error when it tried to refresh the screen.The simple monitoring solutionWe added the following line to our refresh script, so that after the sign had been refreshed a screen shot of the Raspberry Pi would be taken:import -display :0 -window root ~/screenshot.jpgFinally we wrote a small Crontab task that ran on a QV Offices Mac that grabs this screen shot and saved it on the desktop, admittedly we have used a package that it not mega secure, but in reality this is an internal system that only runs an office sign, so we are not to concerned about it being hacked.*/5 * * * * /usr/local/bin/sshpass -p 'password' /usr/bin/scp [email protected]:screenshot.jpg Desktop/QVScreenShot.jpgAs the file icon updates, if the image changes, this gives a quick visual indication of the status of the sign, if for some reason the icon does not look correct the QV Offices administrator can just click on the file to see the exact image currently displayed on the sign.Sometimes a quick and easy solution is better than a more complex and expensive one.

    Read the article

  • Simple algorithm for a sudoku solver java

    - by user142050
    just a quick note first, I originally asked this question on stack overflow but was refered here instead. I've been stuck on this thing for a while, I just can't wrap my head around it. For a homework, I have to produce an algorithm for a sudoku solver that can check what number goes in a blank square in a row, in a column and in a block. It's a regular 9x9 sudoku and I'm assuming that the grid is already printed so I have to produce the part where it solves it. I've read a ton of stuff on the subject I just get stuck expressing it. I want the solver to do the following: If the value is smaller than 9, increase it by 1 If the value is 9, set it to zero and go back 1 If the value is invalid, increase by 1 I've already read about backtracking and such but I'm in the early stage of the class so I'd like to keep it as simple as possible. I'm more capable of writing in pseudo code but not so much with the algorithm itself and it's the algorithm that is needed for this exercise. Thanks in advance for your help guys.

    Read the article

  • Single complex or multiple simple autoload functions [on hold]

    - by Tyson of the Northwest
    Using the spl_autoload_register(), should I use a single autoload function that contains all the logic to determine where the include files are or should I break each include grouping into it's own function with it's own logic to include the files for the called function? As the places where include files may reside expands so too will the logic of a single function. If I break it into multiple functions I can add functions as new groupings are added, but the functions will be copy/pastes of each other with minor alterations. Currently I have a tool with a single registered autoload function that picks apart the class name and tries to predict where it is and then includes it. Due to naming conventions for the project this has been pretty simple. if has namespace if in template namespace look in Root\Templates else look in Root\Modules\Namespace else look in Root\System if file exists include But we are starting to include Interfaces and Traits into our codebase and it hurts me to include the type of a thing in it's name. So we are looking at instead of a single autoload function that digs through the class name and looks for the file and has increasingly complex logic to it, we are looking at having multiple autoload functions registered. But each one follows the same pattern and any time I see that I get paranoid about code copying. function systemAutoloadFunc logic to create probable filename if filename exists in system include it and return true else return false function moduleAutoloadFunc logic to create probable filename if filename exists in modules include it and return true else return false Every autoload function will follow that pattern and the last of each function if filename exists, include return true else return false is going to be identical code. This makes me paranoid about having to update it later across the board if the file_exists include pattern we are using ever changes. Or is it just that, paranoia and the multiple functions with some identical code is the best option?

    Read the article

  • Simple Introduction to using the Enterprise Manager SOA/BPM Facade API by Jaideep Ganguli

    - by JuergenKress
    There may be times when you need to expose just a small section of what is displayed in the Enterprise Manager console for SOA/BPM (EM console). A simple example can be where stakeholders on the systems integration or customer teams want to monitor a dashboard of statistics on how many instances of a composite have been created and how many have faulted. You can see this in the EM, as shown below Some of these stakeholders may not have knowledge of  EM console and they just want a quick view into the statistics, without having to navigate EM. This post describes how to use the Oracle Fusion Middleware Infrastructure Management Java API  for Oracle SOA Suite (also called the Facade API)  to build a custom ADF page to display this information. If you want a quick introduction in using the Facade API, this post is for you. Read the complete article here. SOA & BPM Partner Community For regular information on Oracle SOA Suite become a member in the SOA & BPM Partner Community for registration please visit www.oracle.com/goto/emea/soa (OPN account required) If you need support with your account please contact the Oracle Partner Business Center. Blog Twitter LinkedIn Facebook Wiki Technorati Tags: Enterprise Manager,SOA Community,Oracle SOA,Oracle BPM,Community,OPN,Jürgen Kress

    Read the article

  • Object desing problem for simple school application

    - by Aragornx
    I want to create simple school application that provides grades,notes,presence,etc. for students,teachers and parents. I'm trying to design objects for this problem and I'm little bit confused - because I'm not very experienced in class designing. Some of my present objects are : class PersonalData() { private String name; private String surename; private Calendar dateOfBirth; [...] } class Person { private PersonalData personalData; } class User extends Person { private String login; private char[] password; } class Student extends Person { private ArrayList<Counselor> counselors = new ArrayList<>(); } class Counselor extends Person { private ArrayList<Student> children = new ArrayList<>(); } class Teacher extends Person { private ArrayList<ChoolClass> schoolClasses = new ArrayList<>(); private ArrayList<Subject> subjects = new ArrayList<>(); } This is of course a general idea. But I'm sure it's not the best way. For example I want that one person could be a Teacher and also a Parent(Counselor) and present approach makes me to have two Person objects. I want that user after successful logging in get all roles that it has (Student or Teacher or (Teacher & Parent) ). I think I should make and use some interfaces but I'm not sure how to do this right. Maybe like this: interface Role { } interface TeacherRole implements Role { void addGrade( Student student, Grade grade, [...] ); } class Teacher implements TeacherRole { private Person person; [...] } class User extends Person{ ArrayList<Role> roles = new ArrayList<>(); } Please if anyone could help me to make this right or maybe just point me to some literature/article that covers practical objects design.

    Read the article

< Previous Page | 6 7 8 9 10 11 12 13 14 15 16 17  | Next Page >