[jira] [Created] (ARROW-15853) [Python][Docs] Describe behavior of pyarrow.array()

["David Li (Jira)" <ji...@apache.org>]

David Li created ARROW-15853:
--------------------------------

             Summary: [Python][Docs] Describe behavior of pyarrow.array(<mutable zero-copy source>)
                 Key: ARROW-15853
                 URL: https://issues.apache.org/jira/browse/ARROW-15853
             Project: Apache Arrow
          Issue Type: Improvement
          Components: Documentation, Python
            Reporter: David Li


{noformat}
Python 3.10.0 | packaged by conda-forge | (default, Nov 20 2021, 02:25:18) [GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import numpy as np
>>> import pyarrow as pa
>>> ndarr = np.array(range(10))
>>> arr = pa.array(ndarr)
>>> arr
<pyarrow.lib.Int64Array object at 0x7efdf7974100>
[
  0,
  1,
  2,
  3,
  4,
  5,
  6,
  7,
  8,
  9
]
>>> ndarr[0] = 10
>>> arr
<pyarrow.lib.Int64Array object at 0x7efdf7974100>
[
  10,
  1,
  2,
  3,
  4,
  5,
  6,
  7,
  8,
  9
]
{noformat}
While this behavior makes perfect sense with some consideration, it may surprise people. We should document which sources can be zero-copied, and note that in these cases, modifications to the underlying array will affect the PyArrow array (even though this is normally not the case).



--
This message was sent by Atlassian Jira
(v8.20.1#820001)