Mirror of the gdb mailing list
 help / color / mirror / Atom feed
From: Daniel Jacobowitz <drow@mvista.com>
To: Bradford Chamberlain <brad@cs.washington.edu>
Cc: gdb@sources.redhat.com
Subject: Re: QUERY: Pretty-printing C++ classes in gdb
Date: Fri, 08 Mar 2002 18:58:00 -0000	[thread overview]
Message-ID: <20020308215829.A14794@nevyn.them.org> (raw)
In-Reply-To: <20020308165335.L25327-100000@gnocchi.cs.washington.edu>

On Fri, Mar 08, 2002 at 04:56:43PM -0800, Bradford Chamberlain wrote:
> 
> (I hope the following is not a FAQ.  I didn't find one on the gdb page.)
> 
> I am curious whether gdb has hooks that would allow me to provide a means
> of pretty-printing C++ classes without modifying the gdb source code.  
> For example, imagine that I am supplying you with a C++ class library, and
> want to save you the hassle of inspecting the guts of my classes and
> understanding their inner workings when debugging your code that uses
> them.  Rather, I'd like to cause "p instanceOfClassFoo" to give you a
> clean view of the class that doesn't require understanding its
> implementation.
> 
> One way to do this would be to hack the source, perhaps treating my
> library as a new language.  However, I'd like to avoid changing the gdb
> source both for simplicity and so that I don't have to ship users a new
> copy of gdb.  So is there a less intrusive way that I can hook my pretty
> printing in?  For example, I could imagine that a well-defined method name
> or operator<< overload might be utilized by gdb if it exists.  Or that
> perhaps I can give you a dot-file that will be used at load time to
> configure the behavior.
> 
> Any information on whether such a feature exists, or other possible
> workarounds would be appreciated.

The standard way to do this is:

  - Provide a .gdbinit.  Placed in the current working directory or in
$HOME, it lets you specify commands to be executed.

  - Provide utility functions or methods to print each class however
you wish.

  - Provide aliases in the former to call the latter.

GDB will probably never do this by default for some method name or
operator overload; it is not appropriate (what if the functions affect
state?).  Be sure to write the print functions carefully.


-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer


      reply	other threads:[~2002-03-09  2:58 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-03-08 16:56 Bradford Chamberlain
2002-03-08 18:58 ` Daniel Jacobowitz [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20020308215829.A14794@nevyn.them.org \
    --to=drow@mvista.com \
    --cc=brad@cs.washington.edu \
    --cc=gdb@sources.redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox