There are two phases, data acquisition and usage.
The input format dtd is
DTD
market: meta? rows
meta: marketid period1 period2 period3 period4 period5 period6-9 priority: prioritized Price,Index,c y for use with indicators etc reset: like c y(y), name(reset frequency)
rows: row*
row: id marketid name date indexvalue price currency period1 period2 period3 period4 period5 period6
This have so far been used for gathering info from web on a daily basis. It may also be used for entering historic data. Typically, web pages (choose a name based on this) for markets contain rows with stocks, commodities etc, and the price, currency, index and increased value for given periods. The given periods are named in the meta to indicate the time period, be it one day, one week, one month. The included xslts show how to transform from the web to this xml format. Then run
java -jar target/stockstat-input-0.5-SNAPSHOT-jar-with-dependencies.jar my.xml
with a param file of that format, here my.xml. This may be run from cron, or once if importing historical data. If importing historical data, the period values must be computed in the xml file to be input.