Great chart, Three suggestions...

• break out "accept newlines" into two seperate qulifications: "accept newlines with other record seperator" and "accept newlines with newline seperator" ... that will eliminate the need for the "*" and teh special footnote.
• It might be usefull for people who aren't intimately familiar with *SV to see a simple sample set fo each of the cases described in that table.
• Given a sample set for each case, a small sample program for each module showing how to parse that data would provide both a good example of how to use the module, as well as provide benchmark critera. You could then concievable replace those "+" signs with timing info.

(All of which is just to say: I thought about doing this myself and posting as a followup, but then I remembered I'm at work, and I'm planning on leaving early today, so I really can't justify spending that kind of time ... but I thought I'd through it out in case anyone else thought it sounded like a good idea and wanted to take a stab at it.)