[CEDET-devel] CEDET updates

[Eric M. L. <er...@si...>]

I've checked in some changes over the weekend that some of you should
enjoy.

First, `semantic-adebug.el' is a new application level debugger you
can use.  It is basically a data structure navigator that will give
you a high level overview of the data structure in question, AND let
you drill in to see what's going on.  Since it's semantic specific,
TAGS, FIND-RESULTS, and EIEIO objects are all handled nicely and with
proper labeling.

Some commands to try are:

semantic-adebug-bovinate, semantic-adebug-searchdb, and
semantic-adebug-analyze.  These do what you might expect, but then go
into the application debugger.


I also updated semantic-analyze-possible-completions to do operator
stripping as per my question earlier from this week.  As usual, let me
know if it causes grief, and give semantic-adebug-analyze a try if it
does. ;)


I also updated EDE Emacs Lisp handling code, which caused me to change
all the Makefiles.  There was some waste in the loadpaths which is now
gone.  Yay!


Next, I updated lots of items in the semantic doc from sources, and
added a few new sections, including a section in the application
developers guide for the application debugger.

I've gotten comments in the past that it's hard to get through the
massive doc.  I'm still not really sure what to do about it.


Lastly, and this is a pretty old item, there is a new CEDET module
which is not yet in the CEDET default compilation stream.  If you
update from CVS regularly, you've probably seen it.  It is called
SRecode, or the "Semantic Recoder".  The purpose is to do the opposite
of Semantic, and write code from tags.  This is the tool I always
wanted to write, but needed things like EIEIO, Semantic, and EDE to
provide the backbone for.

The new file cedet/USING_CEDET_FROM_CVS has a brief synopsis of all
unreleased software in CEDET, including a blurb about starting with
srecode, though the srecode manual might be a better start.

I started srecode several years ago, but it suffered several false
starts.  The problem was much harder than I had expected, and the
various pre-written templating systems for Emacs weren't up for the
task.  Eventually i found Google Templates which I used as a model,
and rolled my own template manager, which has let me get to a point
where I think the baseline is established enough to move forward.

If you feel brave, give it a try and let me know what you think.
There's a lot of work left to do on it.

Oh, and getting srecode working is NOT a CEDET 1.0 goal.  srecode is a
bit too immature for that.  I'm just trying to the other CEDET
module's bugs and issues fixed for the 1.0 release.

Enjoy
Eric

-- 
          Eric Ludlam:                 za...@gn..., er...@si...
   Home: http://www.ludlam.net            Siege: www.siege-engine.com
Emacs: http://cedet.sourceforge.net               GNU: www.gnu.org