This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Using of parse tree externally


<<There's a trade-off there. It depends on what you expect to do, for
example if you're goal is to do an expander with a different strategy
than the GNAT one, having the GNAT expansion done might be a problem
(but there are flags to disable it IIRC...).
>>

Laurent, you should look at ASIS more closely, it has many times been
discussed that the information there is not really sufficient for a
compiler, and the point is that this was not designed for this purpose.

> There are absolutely no licensing differences between using ASIS or using
> the low level tree that underlies it.

But you need ACT to fix the license on the ultra important gnatvsn
package, at least for 3.13p ;-) ;-).

One mistake people make is assuming that the notice in a file is necessarily
the correct statement of the licensing conditions. That's not the case. 
You don't even need a copyright statement in the text for something to
be copyrighted.

If you see a GPL notice in a file, it just means that someone pressed a
button on a file to put it there, nothing more, nothing less. You don't
know if that statement was put there by someone authorized to do so. Of
course generally such notices can be mistakes. In this case there is no
mistake, gnatvsn is quite deliberately GPL'ed, so that you need to supply
your own under some circumstances.

(that makes people indicate clearly that the version they are using is
differentiated if they try to make a proprietary tool from the public
version).

So no fixing is about to be done here!


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]