"Eric M. Ludlam" <eric@...> writes:
> I've been through the Common and EIEIO sections already. I also
> checked the first bullet of the Semantic section.
> Pretty much everything under the "Semantic" heading needs to be done.
> I usually work in C or C++ when I develop, so trying out those items
> with other more obscure modes is good.
Oki, I use a lot of obscure modes so no problem :)
> If while studying one of these problems you discover a way to automate
> the process, that would be the ideal solution. Unfortunately much of
> the tests relate to interactive modes, and many of those modes were
> written to test some inner piece of semantic, and sometimes aren't
> useful for anything other than that, such as highlight edits mode.
Im not sure what you mean. Do you mean that its difficult to do tests
that check the interactive behaviour of cedet? Do you have a concrete example?
> Another good thing is to try cedet against dependent tools like JDEE
> or ECB. If there are others, they should be added to the list.
I use ECB and JDEE all the time, seems fine, but Ill try to do some
more structured tests.
> Lastly, just let me know what you accomplish so it doesn't get
> Sadly, I have a wide range of minor changes I need to check in, and I
> haven't been able to check stuff in since the subversion upgrade on
> sourceforge. The biggest change I have relates to how multiple files
> of different modes interact through searches and such. I found that
> while going through some of th semantic checklist for C.
> It's still not clear to me why I can't get CVS to work.
David Ponce did checked in a fix for the dreaded "recursive require"
problem recently, so at least sourceforge should be ok.
Its not simply the server naming change they did?
> Eric Ludlam: zappo@..., eric@...
> Home: http://www.ludlam.net Siege: http://www.siege-engine.com
> Emacs: http://cedet.sourceforge.net GNU: http://www.gnu.org