Re: Creating C++ objects from PyRoot/GaudiPython vs. from C++

From: Brett Viren <bv_at_bnl.gov>
Date: Fri, 25 Feb 2011 12:29:18 -0800

Hi Wim,

<wlavrijsen_at_lbl.gov> writes:

> A couple. If, as I understand your message, all these particular C++ objects
> that you have participate in the C++-side reference counting and should never
> be managed by python, then do:
>
> from GaudiPython import gbl
> gbl.DayaBay.__init__._creates = False
> ud = gbl.DayaBay.UserDataHeader()

The "DayaBay.__init__" doesn't apparently have a "_creates" member:

In [1]: from GaudiPython import gbl
In [2]: gbl.DayaBay.__init__._creates

---

AttributeError                            Traceback (most recent call last)

/home/bviren/work/dayabay/offline/NuWa-trunk/dybgaudi/RootIO/RootIOTest/cmt/<ipython console> in <module>()

AttributeError: 'wrapper_descriptor' object has no attribute '_creates'

---

I couldn't find this "wrapper_descriptor".

However, I find that the class object does have this member. See the following ipython session. However, the underlying C++ object can not be kept alive after the Python object is del'ed:

---

In [4]: gbl.DayaBay.UserDataHeader.__init__._creates Destroying UserDataHeader (0x945beb0) #0 @Thu, 01 Jan 1970 00:00:00 +0000 (GMT) + 0 nsec

Out[4]: 0
In [5]: gbl.DayaBay.UserDataHeader.__init__._creates = False
In [6]: udh = gbl.DayaBay.UserDataHeader()
In [7]: del(udh)
Destroying UserDataHeader (0x93d7df8) #0 @Thu, 01 Jan 1970 00:00:00 +0000 (GMT) +        0 nsec
In [8]: udh = gbl.DayaBay.UserDataHeader()
In [9]: udh.addRef()

Out[9]: 1L
In [10]: del(udh)
Destroying UserDataHeader (0x93d7df8) #0 @Thu, 01 Jan 1970 00:00:00 +0000 (GMT) + 0 nsec

---

FYI, that "Destroying..." message is instrumented in the ~UserDataHeader() destructor and dumps the C++ pointer (followed by an event number and event time stamp, both zero in this test).

> Additional options are to tell the function that stores the object into the
> store that it should take ownership rights, or setting ownership on the
> python object directly.

I can explore this. Do you have some pointers?

> Note that GaudiPython has a different memory policy by default then
> does ROOT.

Okay. Is this something I need to understand to solve this problem?

Thanks!
-Brett.

Received on Fri Feb 25 2011 - 21:29:26 CET