If anyone stumbles here trying to find an incremental parser, here is the 1998 paper (Efficient and Flexible Incremental Parsing) that I should have found before embarking in this failed (expectations not met) project.

If you are still interested in Parsley, go read the old README