Specifying the option -i causes the interactive shell to be started once the parser has finished. This gives you a shell interface to probe the results of the parse. You can examine the chart, print specific results, view derivation trees (see below) and lots of other things.

Note that you should only do this for quick parses. If you leave the parser running a long time, you should output the results to a file. You can still load up the interactive shell and read in these results, as described below.

Using the Shell

Run the parser using the -i (or --interactive) option.

> ./jazzparser -i "AM7 Bm7 E7 AM7"

The parser runs as usual and outputs its results to the terminal. Instead of exiting, it then gives you a shell prompt, like this:

<<<< Jazz Parser interactive shell >>>>
Type 'help' for command list
>>

Type help for a list of commands that you can use.

>> help

This will give you quite a long list. Exactly what commands are available will depend on the options given to the parser. For example, some commands are specific to a particular supertagger and will only appear when that supertagger is used.

TODO: document some of these commands

Loading the Shell Later

TODO: write this