[
https://issues.apache.org/jira/browse/ARROW-2052?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16346436#comment-16346436
]
ASF GitHub Bot commented on ARROW-2052:
---------------------------------------
pitrou commented on issue #1534: ARROW-2052: [C++ / Python] Rework OwnedRef,
remove ScopedRef
URL: https://github.com/apache/arrow/pull/1534#issuecomment-361860680
> I need to take a closer look at our various usages of OwnedRef -- there
are places where we have OwnedRef as member variables of classes, so the risk
is that an OwnedRef might be destructed at the same time that a PyAcquireGIL
lock is being destructed
The PR already takes care of that, by using OwnedRefNoGIL in those places
(perhaps I've missed some?).
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> Unify OwnedRef and ScopedRef
> ----------------------------
>
> Key: ARROW-2052
> URL: https://issues.apache.org/jira/browse/ARROW-2052
> Project: Apache Arrow
> Issue Type: Task
> Components: Python
> Affects Versions: 0.8.0
> Reporter: Antoine Pitrou
> Assignee: Antoine Pitrou
> Priority: Trivial
> Labels: pull-request-available
>
> Currently {{OwnedRef}} and {{ScopedRef}} have similar semantics with small
> differences. Furtheremore, the naming distinction isn't obvious.
> I propose to unify them as a single {{OwnedRef}} class with the following
> characteristics:
> - doesn't take the GIL automatically
> - has a {{release()}} method that decrefs the pointer (and sets the internal
> copy to NULL) before returning it
> - has a {{detach()}} method that returns the pointer (and sets the internal
> copy to NULL) without decrefing it
> For the rare situations where an {{OwnedRef}} may be destroyed with the GIL
> released, a {{OwnedRefNoGIL}} derived class would also be proposed (the
> naming scheme follows Cython here).
> Opinions / comments?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
