home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.lang.c++:11699 gnu.gcc.help:1801 gnu.g++.help:1054
- Path: sparky!uunet!decwrl!bu.edu!ai-lab!ai.mit.edu!gnulists
- From: martin@oahu.cs.ucla.edu (david l. martin)
- Newsgroups: comp.lang.c++,gnu.gcc.help,gnu.g++.help
- Subject: porting to g++ (from other "dialects")
- Message-ID: <gnusenet1992Jul28.160935.11839@cs.ucla.edu>
- Date: 28 Jul 92 16:09:35 GMT
- Followup-To: poster
- Organization: UCLA Computer Science Department
- Lines: 21
- Approved: info-gnu@prep.ai.mit.edu
- To: gnu-gcc-announce@uunet.uu.net
- Originator: martin@oahu.cs.ucla.edu
- Nntp-Posting-Host: oahu.cs.ucla.edu
-
- I am involved in a project in which we need to be able to use g++
- (that is, the g++ component of the latest version of gcc) to compile*
- existing source code which was developed for a variety of other
- leading C++ compilers - including CFRONT, Borland, Zortech, and
- Sabre. I know there are still incompatibilities between g++ and
- the other dialects, but I need to know just how bad they are. For
- example, if I take an average body of CFRONT-compilable code, what
- percentage of it might be rejected by g++, in the sense of causing
- a fatal error?
-
- I would greatly appreciate the benefit of experience from anyone
- who's ported to g++ from a different compiler.
-
- (*To be more precise, I won't actually be generating executables from
- the source code. I'll be using g++ as a front-end to analyze
- existing source code and produce an intermediate representation which
- describes it - essentially an attributed syntax tree.)
-
- Thanks!
-
- - Dave Martin
-