Hi!
On Thursday 06 July 2006 12:01, Thomas N. wrote:
> > > Well, I have the old universaltowel sources
from March 2004, but I'm
> > > pretty sure they won't work with the actual
ArgoUML.
> >
> > Do you think, it makes sense to port them to the
current Argo version?
> >
> > Ciao,
> > Andreas
>
> Sure, why not? It includes reverse engineering of SQL
scripts, where is the
> RE of PHP? Do I miss something? And what about the
included proguard.jar?
The proguard wouldn't be needed for Argo, since it's main
purpose is
obfuscation of the code. There's also an optimization
sideeffect of proguard,
but I don't think, it makes that much sense for an
Opensource app.
The PHP parser is still incomplete AFAIK, although there
were several folks,
who wanted to work on it for different project. Never heard
of them again.
But I have to say, that the PHP parser stuff is ugly, but
that's mainly
because PHP gives you x ways to write the same statement. If
you don't like
braces, you just write the statement without them etc. Not
as easy as parsing
Java...
> Any volunteer for porting this RE code into an actual
ArgoUML module?
The JDBC stuff should be easier to work on. There are
several Antlr SQL
grammars available now. Maybe one could start from there?
But there's still no PHP grammar. Maybe find someone on the
Antlr list?
Ciao,
Andreas
------------------------------------------------------------
---------
To unsubscribe, e-mail: modules-dev-unsubscribe argouml.tigris.org
For additional commands, e-mail: modules-dev-helpargouml.tigris.org
|