This is an experiment to compress stored fields in Lucene using a new CompressionCodec that supports block based compression. We can specify chunkSize and blockSize trading performance over memory by extending CompressingStoredFieldsFormat.
Inspired from https://issues.apache.org/jira/browse/LUCENE-4226
##Build Instructions
mvn clean package
## To create Index arg1-indexStoragePath arg2-Zipped file for index
mvn exec:java -Dexec.mainClass="com.mycompany.indexcompression.JsonIndexWriter" -Dexec.args="../index/ ../RC_2007-10.bz2"
## To search within Index
mvn exec:java -Dexec.mainClass="com.mycompany.indexcompression.Searcher" -Dexec.args="../index/"