-
Notifications
You must be signed in to change notification settings - Fork 1
CG Compilation Framework
Once the code is generated, it must be compiled. Compilation is a bit of an involved process. Basic flow:
- The operator invokes
FragmentContext.getImplementationClass()
(which really gets the implementation object.) - The
FragmentContext
invokes the globalCodeCompiler.getImplementationClass()
method. -
CodeCompiler
looks up the code in the code cache. - If not found,
CodeCompiler
creates aQueryClassLoader
to hold the generated class. - Next
CodeCompiler
callsClassTransformer.getImplementationClass( )
to get the generated class (the actual class this time.) -
ClassTransformer
callsQueryClassLoader.getClassByteCode( )
to compile the code. -
ClassTransformer
then merges the generated code with the template to create the actual class byte codes. - This code is placed into the cache.
- The
CodeCompiler
uses theQueryClassLoader
to create an instance of the class.
The actual compilation is controlled by the ClassCompilerSelector
which selects either the Janio or JDK compilers.
The "JDK compiler" uses the built-in Java compiler (which is, in modern Java, available in the JRE itself.) Java compilation in Java itself is based on the JavaCompiler
class. The javax.tools
provides a dynamic compilation framework. The IBM article provides a sample app that compiles Java classes from in-memory sources.
Drill's implementation is similar to the approach outlined in the IBM article. The DrillJavaFileManager
manages the (virtual) source and class files, DrillJavaFileObject
manages each file.
You can view the generated code in one of three ways:
- Enable debug logging for
AbstractClassCompiler
. The generated code is written to the log file. Set the config optiondrill.exec.compile.debug
totrue
if you want line numbers added to each line. - Uncomment the obvious block of code in
AbstractClassCompiler
to write each generated class to/tmp
. This version is handy as it uses the class name to name each file. - Pass the following to the command line when starting Drill:
-Dorg.codehaus.janino.source_debugging.enable=true -Dorg.codehaus.janino.source_debugging.dir=/tmp
. (See this post for more information.) This version writes all files using generic temporary names, making it hard to find the particular file of interest.
You can gain access to the generated class files by uncommenting the obvious lines near line 256 in MergeAdapter
.