This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [RFC][patch 1/9] initial Python support
- From: Thiago Jung Bauermann <bauerman at br dot ibm dot com>
- To: tromey at redhat dot com
- Cc: thiagoju at linux dot ibm dot com, Daniel Jacobowitz <drow at false dot org>, gdb-patches at sourceware dot org
- Date: Tue, 08 Jul 2008 01:10:51 -0300
- Subject: Re: [RFC][patch 1/9] initial Python support
- References: <20080429155212.444237503@br.ibm.com> <20080429155304.288626880@br.ibm.com> <20080528205921.GA2969@caradoc.them.org> <20080615181833.uxmo25mg0kko40kw@imap.linux.ibm.com> <m3lk0wysfp.fsf@fleche.redhat.com> <m3k5fygby2.fsf@fleche.redhat.com>
On Sun, 2008-07-06 at 11:27 -0600, Tom Tromey wrote:
> >>>>> "Tom" == Tom Tromey <tromey@redhat.com> writes:
>
> >>> > + if (PyErr_Occurred ())
> >>> > + {
> >>> > + PyErr_Print ();
> >>> > + error ("error while executing Python code");
> >>> > + }
> >>> > +}
>
> Thiago> This is stillpending, I think.
>
> FYI -- I fixed this today. I added a new 'maint set python-stack'
> variable that controls whether Python stack traces are printed. By
> default they are. I changed all the new code to use a new
> gdbpy_print_stack which checks this variable. Finally, I set things
> up so that gdb overrides Python's sys.stdout and sys.stderr to print
> things using printf_filtered.
Awesome!
This means that there are only three things pending for this patch to go
in:
1. I'd like someone to verify if my call to bpstat_do_actions in
execute_gdb_command is needed and if I implemented it correctly (I'm not
sure I knew what I was doing, I just wanted something there so I could
post the patch).
2. Testcases
3. Documentation (both user and internal)
I can work on the last two items, but I'd like input on the first. After
these are resolved I'll post a refreshed patch to the list.
--
[]'s
Thiago Jung Bauermann
Software Engineer
IBM Linux Technology Center