[jira] [Commented] (ARROW-10709) [Python] Difficult to make an efficient zero-copy file reader in Python

["Maarten Breddels (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ARROW-10709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17238106#comment-17238106 ] 

Maarten Breddels commented on ARROW-10709:
------------------------------------------

Pandas also does not like it when .read returns a memoryview. Maybe a better solution is to try to call a read_buffer method on the Python object (with a fallback to read). 

> [Python] Difficult to make an efficient zero-copy file reader in Python
> -----------------------------------------------------------------------
>
>                 Key: ARROW-10709
>                 URL: https://issues.apache.org/jira/browse/ARROW-10709
>             Project: Apache Arrow
>          Issue Type: Improvement
>          Components: Python
>            Reporter: Maarten Breddels
>            Priority: Minor
>              Labels: pull-request-available
>          Time Spent: 40m
>  Remaining Estimate: 0h
>
> There is an option to do efficient data transport using file.read_buffer() using zero memory copies (benchmarking have confirmed that, very nice!).
> However, file.read_buffer() when backed by a Python object (via PythonFile), will call PythonFile.read() via PyReadableFile::Read. A 'normal' file.read() that does memory copying, also calls the PythonFile.read() method, but only allows for a bytes object (PyBytes_Check is used in PyReadableFile::Read).
> This makes it hard to create 1 file object in Python land that supports normal .read() (and thus needs to returns a bytes object) and to also support a zero-copy route where .read() can return a memory view.
> Possibly the strict check on PyBytes_Check can me lifted by also allowing trying PyObject_GetBuffer.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)