***********TFormat***********

Introduction
============

TFormat is a tool that select lines of a log file to analyse, separate them into
different trace files, and transform them on an easier syntax to work with.


For Users
=========

Usage : java -jar Mapper.jar -i <input> -r <regex> -o <output> -e <extract algo>

-i/-input : input file to analyse
-r/-regex : regex file to use on the input file
-o/-output : name of the output directory
-e/-extract : extract method used to separate generated files
(date/id/dateid/iddate)

Options :
-t show the duration of each step of the program
-s write non-matched lines on an other file named "trash"


regex syntax
============

You must write your regex on an other file and give it to the program with the
-r option.
This is the syntaxe you must adopt in this file :
-r <regex>
-d <date format> (if needed)

The regex must contain many groupname to work :
<label> is required
<param1> <param2> etc... if there exist (be careful to the order)
<date> if using an extraction method that need it
<idHOST> if using an extraction method that need it


Extraction Algorithm
====================

There are 4 extraction methods :

date : 	sort lines by date. If there is an unnatural gap between 2 events,
the algorithm detect it and create a new file to keep writing events.

id : sort line by idHost. It can be an ip adress or a name.

dateid : sort line by date first. Then, sort line by id.

iddate : sort line by ip first. Then, sort line by date.


Results
=======

Files created are transformed following the model :

<label>(<param1>;<param2>;<param3>...)
<label>(<param1>;<param2>;<param3>...)
.
.
.
.