MapReduce Tutorial : Preparing the environment

To use the Hadoop Java API, you must be able to compile the Java sources with the Hadoop library. An easy way is to use a prepared Makefile:

1. Create a directory for the Java sources.
2. Create a Makefile containing

   Makefile

   include /net/projects/hadoop/java/Makefile

When typing make, every Java file is compiled (with the Hadoop library and some extra classes) to a jar file. It is of course possible to compile just selected sources using:

make ClassName1.jar ClassName2.jar

or

make -f /net/projects/hadoop/java/Makefile ClassName1.jar

Command make clean remove for every Java file its jar file and subdirectory with compiled classes.

Create a directory and a Makefile as described and download CompileTest.java. Try compiling it using make.

If you like using VIM, there is a special configuration available for editing Hadoop java files. Just execute

/net/projects/hadoop/bin/vim

instead of vim.

This configuration of VIM has following features:

• Code completion is enabled for Java files. After pressing ., list of members is displayed (if there are any). The code completion is not perfect, but works very well.
• VIM can run make¹⁾ in the current directory by pressing F7 (both in normal and insert mode). The file being edited is automatically saved. Compilation errors and warnings are parsed. If there are no errors, the window with compilation errors is closed. If there are errors, the window with compilation errors is opened and the cursor is moved to the first error.
• The windows with compilation errors is opened/closed using F9.
• Pressing F11 / F12 goes to previous / next compilation error, if there are any.
• Key F8 opens a taglist – small window with classes, fields and functions in the current file.
• Pressing \\ comments/uncomments current (or selected) line.