Julian Scheid wrote:
So I've finally got a chance to look at the ecj parser yesterday, and
indeed it seems to be perfectly suited for the task. It even comes
with a Javadoc comment parser which appears to be adaptable for
Gjdoc's purposes. Assuming that you're right about the licensing
issues, I agree that this lends itself very well. Thanks for the
heads-up.
Now one big problem remains, namely integrating the new parser into
Gjdoc - as I said, the parsing step is not cleanly abstracted away, so
unfortunately this amounts to quiet a lot of work - so much work
actually that I am considering doing a complete rewrite of the Gjdoc
core instead, something which is long overdue anyway.
Given that a lot of last year's work on Gjdoc went into the doclet,
which can be retained unmodified, and now that I'm quiet acquainted
with the requirements and peculiarities of the system, this isn't as
big an undertaking as it may sound at first.
I'll look further into this during the next week(s) and keep you up to
date.
Julian
Great! I'll volunteer to test out the refactored code as soon as you
have something ready...
Regards,
Dave
