public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
* RE: an xref Suggestion
@ 2000-02-01  1:36 William Gacquer
  2000-02-01 11:53 ` Bruce Stephens
  0 siblings, 1 reply; 4+ messages in thread
From: William Gacquer @ 2000-02-01  1:36 UTC (permalink / raw)
  To: kdevelop; +Cc: gcc

well, I use ctags oftenly too but exuberant ctags has some major defaults.
It is not able to deal with define statements and that not a symbol browser
but it is more like a definition finder.
I cannot say if "gcc -fxref" has any default. 

Help me guys from the gcc mailing list! Could you tell us if the fxref
output format is documented anywhere? Could you also tell us if "gcc -fxref"
is too be used or if we should forget about it? 

	William

> -----Original Message-----
> From: Bernd Gehrmann [ mailto:bernd@physik.hu-berlin.de ]
> Sent: Monday, January 31, 2000 9:27 PM
> To: kdevelop@barney.cs.uni-potsdam.de
> Subject: Re: an xref Suggestion
> 
> 
> On Mon, 31 Jan 2000 William Gacquer wrote:
> >Hello,
> >	Kdevelop gets nicer everyday! That's great. But It is seriously
> >missing a symbol/cross-references browser. I have had a look 
> at a few tools
> >that can do that but they are all missing something. I have 
> just found a
> >very good one : gcc.
> >	Yes, gcc is indeed able to to export a table of 
> cross-references and
> >symbols using the -fxref option. (For curious people who 
> will try this
> >feature now, please know that by typing "gcc -c -fxref 
> foo.c", it will
> >create a hidden ".foo.c.gxref" database. Hard to read but 
> very complete!
> 
> Last time I heard about this, someone that the code behind this option
> is not maintained anymore, and it's only documented in the 
> source code.
> Does it still work correctly, escpecially wrt namespaces, templates,
> nested classes and so on?
> 
> Bernd.
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: an xref Suggestion
  2000-02-01  1:36 an xref Suggestion William Gacquer
@ 2000-02-01 11:53 ` Bruce Stephens
  2000-02-01 12:12   ` Curtis R. Anderson
  0 siblings, 1 reply; 4+ messages in thread
From: Bruce Stephens @ 2000-02-01 11:53 UTC (permalink / raw)
  To: William Gacquer; +Cc: kdevelop, gcc

William Gacquer <wgacquer@ubisoft.fr> writes:

[...]

> Help me guys from the gcc mailing list! Could you tell us if the
> fxref output format is documented anywhere? Could you also tell us
> if "gcc -fxref" is too be used or if we should forget about it?

The code was for C++, and suffers from bitrot (I don't think it was
ever intended to work with C, but I well could be wrong).

However, if someone wanted to work on this, then now would be a great
time to start: Mark Mitchell has been making changes to the C++ front
end which should make this much easier, and has been documenting the
necessary stuff, too (in gcc/cp/ir.texi).

[...]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: an xref Suggestion
  2000-02-01 11:53 ` Bruce Stephens
@ 2000-02-01 12:12   ` Curtis R. Anderson
  2000-02-01 12:57     ` Bruce Stephens
  0 siblings, 1 reply; 4+ messages in thread
From: Curtis R. Anderson @ 2000-02-01 12:12 UTC (permalink / raw)
  To: Bruce Stephens; +Cc: William Gacquer, kdevelop, gcc

Bruce Stephens wrote:
> 
> William Gacquer <wgacquer@ubisoft.fr> writes:
> 
> [...]
> 
> > Help me guys from the gcc mailing list! Could you tell us if the
> > fxref output format is documented anywhere? Could you also tell us
> > if "gcc -fxref" is too be used or if we should forget about it?
> 
> The code was for C++, and suffers from bitrot (I don't think it was
> ever intended to work with C, but I well could be wrong).
> 
> However, if someone wanted to work on this, then now would be a great
> time to start: Mark Mitchell has been making changes to the C++ front
> end which should make this much easier, and has been documenting the
> necessary stuff, too (in gcc/cp/ir.texi).
 
This sounds like something I was wanting to work on.

I miss the listing and cross reference generation capabilities of the
compilers from Digital/Compaq. The gcc community would have a lot to
gain if we can build this into the collection.

I want to see the ability to output listings in either plain text or
PostScript format for neatness. Also, all the compilers in gcc should
have uniform listing appearance.
-- 
Curtis R. Anderson, Co-creator of "Gleepy the Hen", SP 2.5?, KoX
Geek geezer with a minimal web page at http://gleepy.homepage.com/ .
I vacuum my cat regularly.             ICQ: 50137888
mailto:gleepy@intelligencia.com        UTM: PS 7036 7315, zone 17

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: an xref Suggestion
  2000-02-01 12:12   ` Curtis R. Anderson
@ 2000-02-01 12:57     ` Bruce Stephens
  0 siblings, 0 replies; 4+ messages in thread
From: Bruce Stephens @ 2000-02-01 12:57 UTC (permalink / raw)
  To: Curtis R. Anderson; +Cc: William Gacquer, kdevelop, gcc

"Curtis R. Anderson" <gleepy@intelligencia.com> writes:

> Bruce Stephens wrote:

[...]

> > However, if someone wanted to work on this, then now would be a great
> > time to start: Mark Mitchell has been making changes to the C++ front
> > end which should make this much easier, and has been documenting the
> > necessary stuff, too (in gcc/cp/ir.texi).
>  
> This sounds like something I was wanting to work on.
> 
> I miss the listing and cross reference generation capabilities of
> the compilers from Digital/Compaq. The gcc community would have a
> lot to gain if we can build this into the collection.

Yes, I think so too.  The ability to do proper source-navigation (with
honorable exceptions, such as cxref, LXR) strikes me as an obvious
hole in what's available in free software (hence, presumably, Cygnus
buying Source Navigator).  It would be great to have the relevant
basic facilities in gcc (as is reasonably common in other compilers).

KDevelop isn't the only potential consumer of such data, obviously:
there's at least gIDE and SDS, and probably there'd be people using it
in Emacs, too.

Go for it, if you're interested.  From my cursory reading of ir.texi,
it doesn't look *that* hard (for C++; I've no idea what would apply to
other languages, if anything).

> I want to see the ability to output listings in either plain text or
> PostScript format for neatness. Also, all the compilers in gcc
> should have uniform listing appearance.

I must admit, I don't care at all about this.  I find the tools
available for pretty-printing (for example, a2ps) perfectly
sufficient.  However, that's quite possibly just me.

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2000-02-01 12:57 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2000-02-01  1:36 an xref Suggestion William Gacquer
2000-02-01 11:53 ` Bruce Stephens
2000-02-01 12:12   ` Curtis R. Anderson
2000-02-01 12:57     ` Bruce Stephens

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).