Hadoop plugin 1 0 3 for eclipse

Question

I'm new to Hadoop. Can anyone tell me how to create Hadoop Plugin (version 1.0.3) for eclipse? In fact, they removed the plugin from /hadoop-x.x.x/contrib/ (in my case, x.x.x = 1.0.3) There's a eclipse-plugin in /hadoop-x.x.x/src/contrib/.

By the way, What's the "typical way" to develop a MapReduce app using eclipse (words count for example) in term of:

• Configuration (Standalone or Pseudo distributed...)
• Coding convention (Folder structure, code, debug...)

Answer 1

When you have Hadoop Eclipse plugin installed and configured. In Eclipse create a mapreduce project, it provides required dependencies of hadoop and other jars. Then you need to create Main class, Mapper class and Reducer class, In Main class you need to configure Job. wordcount example

once done you can run main program as run on hadoop, no need to start hadoop before running the program

for 1.0.3 plugin: Apache has remove the plugin from Hadoop installation folder. Instead you can find Eclipse Plugin source code with build.xml file at "${HADOOP_HOME}\hadoop-1.0.3\src\contrib\eclipse-plugin". or you can simply download it from here