Starting standard analysis:

java -jar graphana.jar analysis

This will start the analysis with the default configuration. That means, that the graphs of the 'graphs/' directory are 
loaded and some graph parameters are calculated. Finally, the results are written into 'output.txt' as CSV format.

----------Example call with additional configuration--------

The following sample calls the analysis with some additional arguments:

java -jar graphana.jar analysis "graphDir='graphs/'" "extensions=('dim','txt')" directed=false "separator=','" -post QUIT

Some arguments are put in double quotes to avoid shell syntax conflicts.
The input graph directory is explicitly set to 'graphs/'. By setting the extensions parameter, only .dim and .txt files 
are loaded. With directed=false the loaded graphs will be undirected. The separator determines the character between two values
of the output file. The configuration parameters are explained later.
With the -post QUIT arguments Graphana will close after finishing the analysis. Otherwise it awaits a command line input 
and must be closed manually.

Note that you have to regard the shell syntax too, when executing Graphana in this way. For example you must not insert 
white spaces arbitrarily because these are separators for shell arguments. To ensure correct argument passes, put every 
argument in quotes as it is done with the extensions argument in the example shown above.


--------Graph parameter configuration--------

By default, the file 'analysis_graphparameters.txt' is used to determine which graph parameters to compute.
One line in the file is one graph parameter. The following syntax is used:

<name> :: <call>

The name will appear in the output file. The call must be a valid Graphana call of the algorithm. See 'graphana_ops.pdf'
for a list of algorithms. The file can be manipulated for example by inserting the following line:

Treewidth upper bound :: treewidthUpperBound()

When starting the analysis now, an upper bound heuristic for treewidth is started and 'Treewidth upper bound' will appear
in the heading of the CSV.
If a line starts with '//' it will not be executed.


--------List of optional parameters--------

Optional arguments separated by white spaces can be given when calling the jar.
An argument is given with the following syntax:

<key>=<value>

String values are given in single quotes.
What follows is a list of possible argument keys:


graphDir	[default: 'graphs/']
	Filename or directory (relative or absolute) of input graph file(s) (non-recursive)

extensions	[default: ('dim','dot')]
	File extensions as a vector of strings. Only files with the given extensions will be loaded.
	If an empty vector () is given then all files will be loaded.

graphParams	[default: 'analysis_graphparameters.txt']
	Determines the relative or absolute filename of the text file containing valid Graphana graphparameter calls.
	
output		[default: "output.txt"]
 	Relative or absolute filename of the output file. The file will contain the results of the analysis in CSV format.
 	
directed	[default: false]
	If a graph is loaded from a file and the file format does not indicate whether the graph is directed (only) then 
	this value will be used to decide whether the graph is directed or undirected. So for example if DIMACS files
	are used then this parameter will effect the computation results and must be chosen consciously. Note that some
	algorithms do not support directed graphs.
	
weighted	[default: false]
	The same principle as for the 'directed' parameter
	
separator	[default: ';']
	The output of the analysis is a CSV file. With this parameter, the separator for the keys and values can be set.
	







See graphana_manual.pdf for more detailed usage information.