[
https://issues.apache.org/jira/browse/ARROW-7?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15172266#comment-15172266
]
Wes McKinney commented on ARROW-7:
----------------------------------
"In fact this is modern standard to communicate between Python and C." -- this
is not true at all in the scientific Python / PyData ecosystem. As you will see
when I get the patches put together, the extent of the low-level
interoperability required to make this work well goes significantly beyond what
is possible with CFFI. In particular, significant use of the Python and NumPy C
APIs.
> Add Python library build toolchain
> ----------------------------------
>
> Key: ARROW-7
> URL: https://issues.apache.org/jira/browse/ARROW-7
> Project: Apache Arrow
> Issue Type: New Feature
> Components: Python
> Reporter: Wes McKinney
> Assignee: Wes McKinney
>
> I will be working on a patch to make the initial Arrow C++ {{libarrow.so}}
> library callable from Cython (http://cython.org/) extensions. For the
> uninitiated, Cython is the modern "gold standard" for interoperability with
> C++ libraries. I have used it recently to create Kudu's Python client
> (https://github.com/apache/incubator-kudu/tree/master/python/kudu).
> A significant amount of Python "glue code" will be needed to interoperate
> with pandas, NumPy, and other standard Python libraries (in addition to
> Python's built-in scalar data types), but these will be the subject of many
> follow up JIRAs.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
