sgilmore10 opened a new pull request, #36333:
URL: https://github.com/apache/arrow/pull/36333
### Rationale for this change
We would like to extend the MATLAB interface to support timestamp data.
### What changes are included in this PR?
1. Added a new `arrow.array.MATLABArray` class that can be converted to/from
a MATLAB `datetime` array.
2. Added a new type class called `arrow.type.TimestampType`.
3. Added a new enum class called `arrow.type.TimeUnit`.
**Example**
```matlab
>> dates = datetime(2023, 6, 27) + days(0:3)'
dates =
3×1 datetime array
27-Jun-2023
28-Jun-2023
29-Jun-2023
>> timestampArray = arrow.array.TimestampArray(dates)
timestampArray =
[
2023-06-27 00:00:00.000000,
2023-06-28 00:00:00.000000,
2023-06-29 00:00:00.000000
]
>> fromArrow = datetime(timestampArray)
fromArrow =
3×1 datetime array
27-Jun-2023
28-Jun-2023
29-Jun-2023
>> isequal(dates, fromArrow)
ans =
logical
1
```
`TimestampArray` uses `Microsecond` as the default `TimeUnit.` However, you
can specify the `TimeUnit` via a name-value pair during construction.
```matlab
>> timestampArray = arrow.array.TimestampArray(dates, TimeUnit="second")
timestampArray =
[
2023-06-27 00:00:00,
2023-06-28 00:00:00,
2023-06-29 00:00:00
]
>> timestampArray.Type
ans =
TimestampType with properties:
TimeZone: ""
TimeUnit: Second
ID: Timestamp
BitWidth: 64
NumFields: 0
NumBuffers: 2
```
The `TimestampArray` is timezone-aware if the MATLAB `datetime` array's
TimeZone property is set:
```matlab
>> dates = datetime(2023, 6, 27, TimeZone="America/Anchorage") + days(0:3)';
>> timestampArray = arrow.array.TimestampArray(dates);
>> timestampArray.Type.TimeZone
ans =
"America/Anchorage"
```
Lastly, `arrow.array.TimestampArray` treats `NaT` as null values by default.
However, users can control this behavior by supplying either `InferNulls` or
`Valid` as name-value pairs.
```matlab
>> dates = [datetime(2023, 6, 27) NaT datetime(2023, 6, 29)];
>> timestampArray = arrow.array.TimestampArray(dates)
timestampArray =
[
2023-06-27 00:00:00.000000,
null,
2023-06-29 00:00:00.000000
]
>> timestampArray = arrow.array.TimestampArray(dates, Valid=3)
timestampArray =
[
null,
null,
2023-06-29 00:00:00.000000
]
```
### Are these changes tested?
Added three new test files:
1. `tTimestampArray.m`
2.`tTimestampType.m`
3. `tTimeUnit.m`
### Are there any user-facing changes?
Yes.
1. Added `arrow.array.TimestampArray`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
