This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
Re: Function Cross Reference
- From: Andrew Lees <andrewl at acay dot com dot au>
- To: Jan Staschulat <jans at ida dot ing dot tu-bs dot de>
- Cc: gcc-help at gnu dot org
- Date: Wed, 10 Apr 2002 00:50:18 +1000
- Subject: Re: Function Cross Reference
- Organization: Diamond Consulting Services
- References: <Pine.GSO.4.05.10204081602090.7054-100000@krantor.ida.ing.tu-bs.de>
On Tue, 9 Apr 2002 00:35, you wrote:
> Hi,
>
> we have a large software project written in C. Our goal is to create kind
> of a UML Diagramm of it, that unleashes the structure of it. Especially
> the CALL/CALLEE relationship of the functions. One might also call it
> function cross reference.
>
> Before I start writing such a tool myself, I wonder if such a tool already
> exists? Do you have any hints? Is there an option for the c compiler
> that gives me that information?
>
There are several cross referencing/reference chasing tools, depending on
precisely what is being attempted.
The old cxref does things statically, i.e. provides a printout.
The gnu id tools provide a general identifier cross referencing tool that
integrates interactively with emacs/xemacs - these are general purpose tools
that work well with program sources, particularly C/C++. They can be driven
to obtain various types of information from an id database, so could be
attached to structure drawing tools.
sniff is interactive and nice
The xref-speller tools by Marian Vittek http://www.xref-tech.com integrated
very well with emacs/xemacs, lovely tool and my favourite for most of the
time. This tool is astoundingly good. Interactive "follow the reference"
tool, that understands scopes and can manage multiple main programs in a
project. Not sure how they would interact with a diagramming tool, though.
cbrowser.
And many others that escape my recall at the moment.
The tools that integrate with emacs/xemacs may offer a good basis for what
you want, given the enormous flexibilty of that environment, and the
undoubted pre-existence of structure collapsing tools.
regards,
Andy Lees.