From: Tom Tromey <tom@tromey.com>
To: Pedro Alves <palves@redhat.com>
Cc: Tom Tromey <tom@tromey.com>, gdb-patches@sourceware.org
Subject: Re: [RFA 01/14] Introduce py-ref.h
Date: Sat, 12 Nov 2016 17:31:00 -0000 [thread overview]
Message-ID: <8760nsiozp.fsf@tromey.com> (raw)
In-Reply-To: <7df8b22f-cd05-c0d6-5ede-e9dc29bca6e2@redhat.com> (Pedro Alves's message of "Thu, 10 Nov 2016 23:48:49 +0000")
>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:
Pedro> I think this should say something about supporting explicitly
Pedro> wrapping a NULL PyObject *, and how to check whether there's
Pedro> a managed object.
[... other review comments...]
Here's a new version of this patch, which I think addresses all the
comments, from both you and others.
Tom
diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 99edafc..da23621 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,7 @@
+2016-11-10 Tom Tromey <tom@tromey.com>
+
+ * python/py-ref.h: New file.
+
2016-11-11 Yao Qi <yao.qi@linaro.org>
* spu-tdep.c (spu_software_single_step): Don't call
diff --git a/gdb/python/py-ref.h b/gdb/python/py-ref.h
new file mode 100644
index 0000000..ae3b5cc
--- /dev/null
+++ b/gdb/python/py-ref.h
@@ -0,0 +1,158 @@
+/* Python reference-holding class
+
+ Copyright (C) 2016 Free Software Foundation, Inc.
+
+ This file is part of GDB.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
+
+#ifndef GDB_PYTHON_REF_H
+#define GDB_PYTHON_REF_H
+
+/* An instance of this class either holds a reference to a PyObject,
+ or is "NULL". If it holds a reference, then when the object is
+ destroyed, the PyObject is decref'd.
+
+ Normally an instance is constructed using a PyObject*. This sort
+ of initialization lets this class manage the lifetime of that
+ reference.
+
+ Assignment and copy construction will make a new reference as
+ appropriate. Assignment from a plain PyObject* is disallowed to
+ avoid confusion about whether this acquires a new reference;
+ instead use the "reset" method -- which, like the PyObject*
+ constructor, transfers ownership.
+*/
+class gdbpy_ref
+{
+ public:
+
+ /* Create a new NULL instance. */
+ gdbpy_ref ()
+ : m_obj (NULL)
+ {
+ }
+
+ /* Create a new instance. OBJ is a reference, management of which
+ is now transferred to this class. */
+ explicit gdbpy_ref (PyObject *obj)
+ : m_obj (obj)
+ {
+ }
+
+ /* Copy another instance. */
+ gdbpy_ref (const gdbpy_ref &other)
+ : m_obj (other.m_obj)
+ {
+ if (m_obj != NULL)
+ Py_INCREF (m_obj);
+ }
+
+ /* Transfer ownership from OTHER. */
+ gdbpy_ref (gdbpy_ref &&other)
+ : m_obj (other.m_obj)
+ {
+ other.m_obj = NULL;
+ }
+
+ /* Destroy this instance. */
+ ~gdbpy_ref ()
+ {
+ Py_XDECREF (m_obj);
+ }
+
+ /* Copy another instance. */
+ gdbpy_ref &operator= (const gdbpy_ref &other)
+ {
+ /* Do nothing on self-assignment. */
+ if (this != &other)
+ {
+ Py_XDECREF (m_obj);
+ m_obj = other.m_obj;
+ if (m_obj != NULL)
+ Py_INCREF (m_obj);
+ }
+ return *this;
+ }
+
+ /* Transfer ownership from OTHER. */
+ gdbpy_ref &operator= (gdbpy_ref &&other)
+ {
+ /* Note that this handles self-assignment properly. */
+ std::swap (m_obj, other.m_obj);
+ return *this;
+ }
+
+ /* Change this instance's referent. OBJ is a reference, management
+ of which is now transferred to this class. */
+ void reset (PyObject *obj)
+ {
+ Py_XDECREF (m_obj);
+ m_obj = obj;
+ }
+
+ /* Return this instance's referent. In Python terms this is a
+ borrowed pointer. */
+ PyObject *get () const
+ {
+ return m_obj;
+ }
+
+ /* Return this instance's referent, and stop managing this
+ reference. The caller is now responsible for the ownership of
+ the reference. */
+ PyObject *release ()
+ {
+ PyObject *result = m_obj;
+
+ m_obj = NULL;
+ return result;
+ }
+
+ private:
+
+ PyObject *m_obj;
+};
+
+inline bool operator== (const gdbpy_ref &self, const gdbpy_ref &other)
+{
+ return self.get () == other.get ();
+}
+
+inline bool operator== (const gdbpy_ref &self, const PyObject *other)
+{
+ return self.get () == other;
+}
+
+inline bool operator== (const PyObject *self, const gdbpy_ref &other)
+{
+ return self == other.get ();
+}
+
+inline bool operator!= (const gdbpy_ref &self, const gdbpy_ref &other)
+{
+ return self.get () != other.get ();
+}
+
+inline bool operator!= (const gdbpy_ref &self, const PyObject *other)
+{
+ return self.get () != other;
+}
+
+inline bool operator!= (const PyObject *self, const gdbpy_ref &other)
+{
+ return self != other.get ();
+}
+
+#endif /* GDB_PYTHON_REF_H */
next prev parent reply other threads:[~2016-11-12 17:31 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-07 6:00 [RFA 00/14] add a smart pointer for PyObject* Tom Tromey
2016-11-07 5:48 ` [RFA 06/14] Use gdbpy_reference in gdbpy_inferiors Tom Tromey
2016-11-07 5:48 ` [RFA 13/14] Use gdbpy_reference in py-value.c Tom Tromey
2016-11-07 5:48 ` [RFA 07/14] Use gdbpy_reference in gdbpy_breakpoints Tom Tromey
2016-11-07 5:48 ` [RFA 12/14] Use gdbpy_reference in python.c Tom Tromey
2016-11-07 5:48 ` [RFA 10/14] Use gdbpy_reference in call_doc_function Tom Tromey
2016-11-07 5:48 ` [RFA 14/14] Use gdbpy_reference in gdbpy_lookup_symbol Tom Tromey
2016-11-07 5:48 ` [RFA 08/14] Use gdbpy_reference in py-framefilter.c Tom Tromey
2016-11-07 5:48 ` [RFA 11/14] Use gdbpy_reference in py-prettyprint.c Tom Tromey
2016-11-07 5:48 ` [RFA 05/14] Use gdbpy_reference in py-function.c Tom Tromey
2016-11-07 5:48 ` [RFA 03/14] Use gdbpy_reference in py-type.c Tom Tromey
2016-11-07 5:48 ` [RFA 01/14] Introduce py-ref.h Tom Tromey
2016-11-07 7:45 ` Jan Kratochvil
2016-11-07 15:48 ` Tom Tromey
2016-11-10 23:48 ` Pedro Alves
2016-11-12 17:31 ` Tom Tromey [this message]
2016-11-15 14:32 ` Pedro Alves
2016-11-16 23:18 ` Tom Tromey
2016-11-16 23:34 ` Pedro Alves
2016-11-07 5:48 ` [RFA 04/14] Use gdbpy_reference in gdbpy_string_to_argv Tom Tromey
2016-11-07 5:56 ` [RFA 02/14] Change event code to use gdbpy_reference Tom Tromey
2016-11-11 0:09 ` Pedro Alves
2016-11-11 0:51 ` Pedro Alves
2016-11-07 5:57 ` [RFA 09/14] Use gdbpy_reference in py-linetable.c Tom Tromey
2016-11-08 4:07 ` [RFA 00/14] add a smart pointer for PyObject* Tom Tromey
2016-11-11 1:18 ` Pedro Alves
2016-11-11 3:34 ` Tom Tromey
2016-11-11 4:03 ` Pedro Alves
2016-11-11 5:49 ` Tom Tromey
2016-11-12 17:11 ` Tom Tromey
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=8760nsiozp.fsf@tromey.com \
--to=tom@tromey.com \
--cc=gdb-patches@sourceware.org \
--cc=palves@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