This is the mail archive of the
mailing list for the GCC project.
Re: Using of parse tree externally
- To: Joe Buck <jbuck at racerx dot synopsys dot com>, dmj+ at andrew dot cmu dot edu (Daniel Jacobowitz)
- Subject: Re: Using of parse tree externally
- From: Denis Perchine <dyp at perchine dot com>
- Date: Tue, 10 Oct 2000 12:07:35 +0700
- Cc: gcc at gcc dot gnu dot org
- References: <200010100141.SAA09375@racerx.synopsys.com>
> > > Denis> I work on static analisys tool. And would like to use parse
> > > Denis> tree produced by C++ gcc frontend. The question is did
> > > Denis> someone did some movements to store parse tree in a file
> > > Denis> and read it back in a program which is outside gcc tree?
> > Isn't this the same discussion which has gotten repeatedly squashed for
> > making it too easy to leverage GCC with a non-GPL'd project via a
> > defined interface?
> Yes, that is a concern: proprietary compilers that use the GCC front ends.
> This is a especially serious risk for languages that are hard to parse and
> analyze, like C++. There are commercial license-able front ends
> available, like the EDG front end, so any proprietary developers sniffing
> around would do better to go in that direction, though it will cost.
> That doesn't mean that it would necessarily be ruled out, but we'd need
> good arguments for how free software can benefit from the capability.
> So Denis, what do you want it for? What would you do with it if you had
Ough, year... I will use it in commercial compiler... Sure... Guys... Did you
read my mail?
There is an existing static analisys tool called Wasp. It is written by my
science advisor Vladimir Shelekhov. (http://www.iis.nsk.su/wasp/)
It targets Java, Module-2 and Oberon-2.
I personally prefer C++. That is why I like to use gcc frontend information.
I plans to write static analisys tool for C++. Will it be commercial, or
not... I do not know yet. For me it is an interesting science problem for my