[SPARK-44717][PYTHON][PS] Respect TimestampNTZ in resampling #42392
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
cc @zhengruifeng and @attilapiros FYI |
HyukjinKwon
force-pushed
the
SPARK-44717
branch
from
cb0f65b
to
758016f
Compare
|
cc @gengliangwang too |
zhengruifeng
approved these changes
Aug 8, 2023
attilapiros
reviewed
Aug 8, 2023
Comment on lines
265
to
266
| def test_series_resample(self): | ||
| self.check_series_resample() |
There was a problem hiding this comment.
We are still depending on the TZ environment setting indirectly. So just by running the test on a different TZ this test (and also the test_dataframe_resample) would simply fail.
We should either guarantee the correct TZ in the beginning of the test or validate the assumption and produce a meaningful error.
There was a problem hiding this comment.
@HyukjinKwon this is still an issue. The old tests still are failing when the TZ is not UTC i.e in America/New_York.
See https://github.com/attilapiros/spark/actions/runs/5884585704/job/15960073184?pr=5
Anyway for production we have the "spark.sql.timestampType=TIMESTAMP_NTZ" settings.
There was a problem hiding this comment.
Yeah ... this PR enables spark.sql.timestampType=TIMESTAMP_NTZ as a workaround for now ..
To fix this, we need a bigger scope of change .. and can be arguable in a way.
HyukjinKwon
commented
Aug 9, 2023
| class ResampleWithTimezoneTests( | ||
| ResampleWithTimezoneMixin, PandasOnSparkTestUtils, TestUtils, ReusedConnectTestCase | ||
| ): | ||
| @unittest.skip("SPARK-44731: Support 'spark.sql.timestampType' in Python Spark Connect client") |
|
Merged to master and branch-3.5. |
HyukjinKwon
added a commit
that referenced
this pull request
Aug 9, 2023
### What changes were proposed in this pull request? This PR proposes to respect `TimestampNTZ` type in resampling at pandas API on Spark. ### Why are the changes needed? It still operates as if the timestamps are `TIMESTAMP_LTZ` even when `spark.sql.timestampType` is set to `TIMESTAMP_NTZ`, which is unexpected. ### Does this PR introduce _any_ user-facing change? This fixes a bug so end users can use exactly same behaviour with pandas with `TimestampNTZType` - pandas does not respect the local timezone with DST. While we might need to follow this even for `TimestampType`, this PR does not address the case as it might be controversial. ### How was this patch tested? Unittest was added. Closes #42392 from HyukjinKwon/SPARK-44717. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Hyukjin Kwon <gurwls223@apache.org> (cherry picked from commit e05959e) Signed-off-by: Hyukjin Kwon <gurwls223@apache.org>
zhengruifeng
pushed a commit
that referenced
this pull request
Aug 11, 2023
…in Python Spark Connect ### What changes were proposed in this pull request? This PR proposes: - Share the namespaces for `to_timestamp_ntz`, `to_timestamp_ltz` and `to_unix_timestamp` in Spark Connect. They were missed. - Adds the support of `TimestampNTZ` for literal handling in Python Spark Connect (by respecting `spark.sql.timestampType`). ### Why are the changes needed? For feature parity, and respect timestamp ntz in resampling in pandas API on Spark ### Does this PR introduce _any_ user-facing change? Yes, this virtually fixes the same bug: #42392 in Spark Connect with Python. ### How was this patch tested? Unittests reenabled. Closes #42445 from HyukjinKwon/SPARK-44731. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Ruifeng Zheng <ruifengz@apache.org>
zhengruifeng
pushed a commit
that referenced
this pull request
Aug 11, 2023
…in Python Spark Connect ### What changes were proposed in this pull request? This PR proposes: - Share the namespaces for `to_timestamp_ntz`, `to_timestamp_ltz` and `to_unix_timestamp` in Spark Connect. They were missed. - Adds the support of `TimestampNTZ` for literal handling in Python Spark Connect (by respecting `spark.sql.timestampType`). ### Why are the changes needed? For feature parity, and respect timestamp ntz in resampling in pandas API on Spark ### Does this PR introduce _any_ user-facing change? Yes, this virtually fixes the same bug: #42392 in Spark Connect with Python. ### How was this patch tested? Unittests reenabled. Closes #42445 from HyukjinKwon/SPARK-44731. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Ruifeng Zheng <ruifengz@apache.org> (cherry picked from commit 73b0376) Signed-off-by: Ruifeng Zheng <ruifengz@apache.org>
hvanhovell
pushed a commit
to hvanhovell/spark
that referenced
this pull request
Aug 13, 2023
…in Python Spark Connect ### What changes were proposed in this pull request? This PR proposes: - Share the namespaces for `to_timestamp_ntz`, `to_timestamp_ltz` and `to_unix_timestamp` in Spark Connect. They were missed. - Adds the support of `TimestampNTZ` for literal handling in Python Spark Connect (by respecting `spark.sql.timestampType`). ### Why are the changes needed? For feature parity, and respect timestamp ntz in resampling in pandas API on Spark ### Does this PR introduce _any_ user-facing change? Yes, this virtually fixes the same bug: apache#42392 in Spark Connect with Python. ### How was this patch tested? Unittests reenabled. Closes apache#42445 from HyukjinKwon/SPARK-44731. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Ruifeng Zheng <ruifengz@apache.org>
valentinp17
pushed a commit
to valentinp17/spark
that referenced
this pull request
Aug 24, 2023
### What changes were proposed in this pull request? This PR proposes to respect `TimestampNTZ` type in resampling at pandas API on Spark. ### Why are the changes needed? It still operates as if the timestamps are `TIMESTAMP_LTZ` even when `spark.sql.timestampType` is set to `TIMESTAMP_NTZ`, which is unexpected. ### Does this PR introduce _any_ user-facing change? This fixes a bug so end users can use exactly same behaviour with pandas with `TimestampNTZType` - pandas does not respect the local timezone with DST. While we might need to follow this even for `TimestampType`, this PR does not address the case as it might be controversial. ### How was this patch tested? Unittest was added. Closes apache#42392 from HyukjinKwon/SPARK-44717. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Hyukjin Kwon <gurwls223@apache.org>
valentinp17
pushed a commit
to valentinp17/spark
that referenced
this pull request
Aug 24, 2023
…in Python Spark Connect ### What changes were proposed in this pull request? This PR proposes: - Share the namespaces for `to_timestamp_ntz`, `to_timestamp_ltz` and `to_unix_timestamp` in Spark Connect. They were missed. - Adds the support of `TimestampNTZ` for literal handling in Python Spark Connect (by respecting `spark.sql.timestampType`). ### Why are the changes needed? For feature parity, and respect timestamp ntz in resampling in pandas API on Spark ### Does this PR introduce _any_ user-facing change? Yes, this virtually fixes the same bug: apache#42392 in Spark Connect with Python. ### How was this patch tested? Unittests reenabled. Closes apache#42445 from HyukjinKwon/SPARK-44731. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Ruifeng Zheng <ruifengz@apache.org>
ragnarok56
pushed a commit
to ragnarok56/spark
that referenced
this pull request
Mar 2, 2024
### What changes were proposed in this pull request? This PR proposes to respect `TimestampNTZ` type in resampling at pandas API on Spark. ### Why are the changes needed? It still operates as if the timestamps are `TIMESTAMP_LTZ` even when `spark.sql.timestampType` is set to `TIMESTAMP_NTZ`, which is unexpected. ### Does this PR introduce _any_ user-facing change? This fixes a bug so end users can use exactly same behaviour with pandas with `TimestampNTZType` - pandas does not respect the local timezone with DST. While we might need to follow this even for `TimestampType`, this PR does not address the case as it might be controversial. ### How was this patch tested? Unittest was added. Closes apache#42392 from HyukjinKwon/SPARK-44717. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Hyukjin Kwon <gurwls223@apache.org>
ragnarok56
pushed a commit
to ragnarok56/spark
that referenced
this pull request
Mar 2, 2024
…in Python Spark Connect ### What changes were proposed in this pull request? This PR proposes: - Share the namespaces for `to_timestamp_ntz`, `to_timestamp_ltz` and `to_unix_timestamp` in Spark Connect. They were missed. - Adds the support of `TimestampNTZ` for literal handling in Python Spark Connect (by respecting `spark.sql.timestampType`). ### Why are the changes needed? For feature parity, and respect timestamp ntz in resampling in pandas API on Spark ### Does this PR introduce _any_ user-facing change? Yes, this virtually fixes the same bug: apache#42392 in Spark Connect with Python. ### How was this patch tested? Unittests reenabled. Closes apache#42445 from HyukjinKwon/SPARK-44731. Authored-by: Hyukjin Kwon <gurwls223@apache.org> Signed-off-by: Ruifeng Zheng <ruifengz@apache.org>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
This PR proposes to respect
TimestampNTZtype in resampling at pandas API on Spark.Why are the changes needed?
It still operates as if the timestamps are
TIMESTAMP_LTZeven whenspark.sql.timestampTypeis set toTIMESTAMP_NTZ, which is unexpected.Does this PR introduce any user-facing change?
This fixes a bug so end users can use exactly same behaviour with pandas with
TimestampNTZType- pandas does not respect the local timezone with DST. While we might need to follow this even forTimestampType, this PR does not address the case as it might be controversial.How was this patch tested?
Unittest was added.