To install mercury under Ubuntu ,

XXX-------- BELOW is not needed anymore since build now includes working version --------XXX

checkout or copy the dataone-cn-mercury to a local dir

edit the control file for your local configuration ( machine type and pre-requisites)
- or wait for error and then edit them.

Change back to where the dataone-cn-mercury directory was copied to  and

	sudo dpkg-deb -b dataone-cn-mercury

	sudo dpkg -i dataone-cn-mercury.deb


After apache / tomcat / mysql / dataone-cn-mercury are installed with no errors, 
verify tomcat & apache are accessible, then 

XXX----- AND MAY NOT NEED ANY OF THE REST OF THIS EITHER ---XXX

login as root user to mysql.

// change root pw below to suit environment

	mysql -u root -ppassword
		pw=dataone


	create database mercury3_harvests_datanet;

	use mercury3_harvests_datanet;

	create user mercuryuser identified by 'mercury3user';

	grant all privileges on *.* to mercuryuser@localhost identified by 'mercury3user' ;

To harvest and Index metadata files, change to the local mercury dev directory

	cd /var/mercury/mercury_dev/datanet

XXX----- YOU DO NOT NEED TO RUN UPDATE.SH, BUT CURRENTLY IT HAS SOME COMMENTED OUT CODE THAT WILL NEED TO BE RUN SEPARATELY FROM THE COMMAND LINE FOR MERCURY POPULATION ---XXX

	run update.sh

With some more luck, http://localhost/mercury3 should now give us usable data.


for the Harvest_dataone.jar executable
the parameters are : instance,  source, debug, htmlurl.

	
	instance		name of the directory for locating the data files

	source			specifies data source directory for walking. Can be 
	  			used to automatically walk multiple directories
	 			if the application is configured entirely from 
	 			a properties file. Otherwise, defaults to a single
	 			value specified on the command line.

	debug flag		when set to "debug", writes debugging information. 
	 						else, not.


	htmlurl			used as switch for selecting XPATH for file location
	 			saved in the harvestInfo and transformedFiles tables