public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Tom Tromey <tromey@redhat.com>
To: sami wagiaalla <swagiaal@redhat.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [patch] Support inferior events in python
Date: Thu, 03 Feb 2011 21:42:00 -0000	[thread overview]
Message-ID: <m3sjw4dccv.fsf@fleche.redhat.com> (raw)
In-Reply-To: <4D4ADA8A.40507@redhat.com> (sami wagiaalla's message of "Thu, 03	Feb 2011 11:40:42 -0500")

>>>>> "Sami" == sami wagiaalla <swagiaal@redhat.com> writes:

Sami> Added. I also, documented StopEvent. In events which inherit from
Sami> others I added a statement asking the reader to refer to the
Sami> parent. Do you think that is enough, or should the inherited
Sami> attributes be restated ?

I think it is sufficient to mention the superclass.

Sami> +int
Sami> +emit_stop_event (struct bpstats *bs, enum target_signal stop_signal)
Sami> +{
Sami> +  PyObject *stop_event_obj = NULL; /* Appease GCC warning.  */
Sami> +
Sami> +  if (evregpy_no_listeners_p (gdb_py_events.stop))
Sami> +    return 0;

Sami> +  if (bs && bs->breakpoint_at
Sami> +      && bs->breakpoint_at->type == bp_breakpoint)

I think this logic is too restrictive.
What if we change what breakpoints are reflected to Python?
That is, I think the check for bp_breakpoint should be removed.
Instead, you could just have it check ->py_bp_object.
But if you are doing that you should either rearrange, or just get rid
of gdbpy_breakpoint_from_bpstats, to either preserve the abstraction or
remove it.

Sami> +      PyObject *breakpoint = gdbpy_breakpoint_from_bpstats (bs);
Sami> +      if (breakpoint != NULL)
Sami> +      stop_event_obj =
Sami> +	  create_breakpoint_event_object (breakpoint);
Sami> +      if (!stop_event_obj)
Sami> +	goto fail;

This could result in a "fail" without an error being set.
But the above change should eliminate that.

Sami> +PyObject *
Sami> +create_thread_event_object (PyTypeObject *py_type)
Sami> +{
Sami> +  PyObject *thread_event_obj = create_event_object (py_type);
Sami> +  PyObject *thread = get_event_thread();

Missing space.

Sami> +
Sami> +  if (!thread_event_obj || !thread)
Sami> +    goto fail;

You have to do error checks after each Python call, you can't batch them.

Either: create_event_object can fail, in which case it is not ok to call
get_event_thread on failure; or, create_event_object can return NULL but
not set a Python exception, in which case you must set one explicitly.

Tom

  parent reply	other threads:[~2011-02-03 21:42 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-01-04 15:54 sami wagiaalla
2011-01-04 18:22 ` Eli Zaretskii
2011-01-04 20:09 ` Tom Tromey
2011-01-17 22:59   ` sami wagiaalla
2011-01-19 16:42     ` Tom Tromey
2011-01-21 23:06       ` sami wagiaalla
2011-01-28 16:21         ` Tom Tromey
2011-02-02 21:04           ` sami wagiaalla
2011-02-02 21:35             ` Tom Tromey
2011-02-03 16:41               ` sami wagiaalla
2011-02-03 18:26                 ` Eli Zaretskii
2011-02-03 19:45                   ` sami wagiaalla
2011-02-03 21:42                 ` Tom Tromey [this message]
2011-02-04 20:07                   ` sami wagiaalla
2011-02-04 20:29                     ` Tom Tromey
2011-02-04 20:35                       ` sami wagiaalla
2011-02-04 23:00                         ` Paul Pluzhnikov
2011-02-05  5:44                           ` Hui Zhu
2011-02-07 15:22                             ` sami wagiaalla
2011-02-07 15:24                               ` Tom Tromey
2011-02-07 15:34                                 ` Paul Pluzhnikov
2011-02-07 16:01                                   ` sami wagiaalla
2011-02-07 15:39                                 ` sami wagiaalla
2011-04-20 20:26                         ` Patch for non-stop remote assertion (was: RE: [patch] Support inferior events in python) Marc Khouzam
2011-04-25 18:12                           ` Patch for non-stop remote assertion Tom Tromey
2011-04-25 18:31                             ` Marc Khouzam
2011-05-16 15:41                               ` Marc Khouzam
2011-05-19 18:38                               ` Tom Tromey
2011-02-09  7:55                     ` [patch] Support inferior events in python Jan Kratochvil
2011-02-09 16:19                       ` sami wagiaalla
2011-02-09 16:30                         ` Jan Kratochvil
2011-02-11 15:28                           ` sami wagiaalla
2011-02-11 15:55                             ` Joel Brobecker
2011-02-11 19:19                               ` sami wagiaalla
2011-02-11 19:46                                 ` Jan Kratochvil
2011-02-11 15:57                             ` Pedro Alves
2011-02-14 17:36                               ` sami wagiaalla
2011-02-16 11:48                                 ` Jan Kratochvil
2011-07-06 19:42                                   ` Jan Kratochvil
2011-07-07 13:51                                     ` sami wagiaalla
2011-07-07 14:03                                       ` Jan Kratochvil
2011-09-13 21:45                                       ` Jan Kratochvil

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=m3sjw4dccv.fsf@fleche.redhat.com \
    --to=tromey@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --cc=swagiaal@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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).