[SPARK-51590][SQL] Disable TIME in builtin file-based datasources #50358
Conversation
This file contains hidden or 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
MaxGekk
changed the title
MaxGekk
requested review from
gengliangwang,
yaooqinn,
dongjoon-hyun and
HyukjinKwon
HyukjinKwon
approved these changes
Mar 24, 2025
|
Merged to master. |
SauronShepherd
pushed a commit
to SauronShepherd/spark
that referenced
this pull request
Mar 25, 2025
### What changes were proposed in this pull request? In the PR, I propose to modify the `supportDataType()` method of builtin file based datasources, and disable the new data type `TIME` till it is fully supported in such datasources. In particular, disable it in the following V1 and V2 datasources: - Text - JSON/CSV - XML - Parquet - ORC: native and Hive - Avro ### Why are the changes needed? Before the changes if we don't disable the unsupported data type `TIME` in datasources, users might get exceptions from external libs like: ORC: ```java java.lang.IllegalArgumentException: Can't parse category at 'time^(6)' at org.apache.orc.impl.ParserUtils.parseCategory(ParserUtils.java:67) ``` Avro: ```java org.apache.spark.sql.avro.IncompatibleSchemaException: Unexpected type TimeType(6). at org.apache.spark.sql.avro.SchemaConverters$.toAvroType(SchemaConverters.scala:369) ``` ### Does this PR introduce _any_ user-facing change? Yes. After the changes, users get more friendly `AnalysisException`: ``` [UNSUPPORTED_DATA_TYPE_FOR_DATASOURCE] The ORC datasource doesn't support the column `t` of the type "TIME(6)". SQLSTATE: 0A000 ``` ### How was this patch tested? By running the modified test suites: ``` $ build/sbt "test:testOnly *FileBasedDataSourceSuite" $ build/sbt -Phive "test:testOnle *HiveOrcQuerySuite" ``` ### Was this patch authored or co-authored using generative AI tooling? No. Closes apache#50358 from MaxGekk/time-off-fs-ds-2. Authored-by: Max Gekk <[email protected]> Signed-off-by: Hyukjin Kwon <[email protected]>
MaxGekk
added a commit
that referenced
this pull request
Mar 26, 2025
### What changes were proposed in this pull request? In the PR, I propose to support the new data type `TIME` in the Parquet datasource, and store it as the logical type: ``` TimeType(isAdjustedToUTC = false, unit = MICROS) ``` see https://github.com/apache/parquet-format/blob/master/LogicalTypes.md#time where - `isAdjustedToUTC` = **false** means the `TIME` values represent a local time, regardless of the local time zone in effect. - `unit` = **MICROS** - stores the number of microseconds after midnight. At the writer side, recognize `int64` values of the logical type `TimeType(isAdjustedToUTC = false, unit = MICROS)` and convert them to Spark's type `TIME(6)`. ### Why are the changes needed? To allow Spark SQL users reading of parquet files with `TIME` values created by other systems. ### Does this PR introduce _any_ user-facing change? Yes. Before the changes, reading and writing Dataset to/from parquet failed with the error `UNSUPPORTED_DATA_TYPE_FOR_DATASOURCE` because the PR #50358 disabled `TIME` in the Parquet datasource. ### How was this patch tested? By running new tests: ``` $ build/sbt "test:testOnly *ParquetSchemaSuite" $ build/sbt "test:testOnly *ParquetIOSuite" $ build/sbt "test:testOnly *FileBasedDataSourceSuite" ``` ### Was this patch authored or co-authored using generative AI tooling? No. Closes #50389 from MaxGekk/time-parquet. Authored-by: Max Gekk <[email protected]> Signed-off-by: Max Gekk <[email protected]>
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?
In the PR, I propose to modify the
supportDataType()method of builtin file based datasources, and disable the new data typeTIMEtill it is fully supported in such datasources. In particular, disable it in the following V1 and V2 datasources:Why are the changes needed?
Before the changes if we don't disable the unsupported data type
TIMEin datasources, users might get exceptions from external libs like:ORC:
Avro:
Does this PR introduce any user-facing change?
Yes. After the changes, users get more friendly
AnalysisException:How was this patch tested?
By running the modified test suites:
Was this patch authored or co-authored using generative AI tooling?
No.