This package pulls out the contents of a table from a string containing
HTML. Each time a table is encountered, data will be stored in an array
consisting of a hash of whatever was discovered about the table -- id,
name, border, cellspacing etc, and of course data contained within the
table.

The format of each hash will look something like

attributes keys from the attributes of the <table> tag
@{$table_headers} array of table headers, in order found
@{$table_rows} rows discovered, in order