Treats all files in a directory structure as HTML log files and parses them into event sequences and a GUI model. Also sub directories are parsed.

The parsing process can be parameterized. This allows to ignore ids or indexes of GUI elements in the log files. If they are ignored, the GUI model is more harmonized and GUI elements are considered equal although they are not. This may be helpful, e.g., if you have a table where each row is semantically the same. Without ignoring indexes or ids of the rows, each row is treated separately. But with ignored indexes or ids, all rows are considered the same.

To ignore the indexes, add -clearIndex=<path to GUI element> as parameter to the command call. To ignore ids, add -clearId=<path to GUI element> to the command call. The path to the GUI element is written using the HTML tag names and either their index or their id as identification. E.g., to denote all rows in a table where the table has the id "table_1" you can specify "table(htmlId=table_1)/tbody/tr". To denote e.g. all divs being the child of a div with an index 1, you specify "div[1]/div".  

$USAGE$
<directory> path to the directory 
[<sequenceNames>] array of sequences into which the parsed events shall be stored
[<clearId>] used to define GUI elements of which the ids shall be ignored
[<clearIndex>] used to define GUI elements of which the indexes shall be ignored

Example(s):
parseDirHTML /path/to/directory
parseDirHTML /path/to/directory sequences -clearId=table(htmlId=overview)/tbody[0]/tr