| commit | fd57c3493af7b73390d52aee9d2817f1266d8347 | [log] [tgz] |
|---|---|---|
| author | Kent Yao <yao@apache.org> | Wed May 01 15:33:27 2024 -0700 |
| committer | Dongjoon Hyun <dhyun@apple.com> | Wed May 01 15:33:27 2024 -0700 |
| tree | 81da527893273e75043f8ffc2e6edc02613e8bcf | |
| parent | 69ea082fc69aa61b0e62f68301dfefef518bc6e7 [diff] |
[SPARK-47911][SQL] Introduces a universal BinaryFormatter to make binary output consistent
### What changes were proposed in this pull request?
This PR introduces a universal BinaryFormatter to make binary output consistent
across all clients, such as `beeline`, `spark-sql`, and `spark-shell`, for both primitive and nested binaries.
Considering we already have different styles and compatibility with Apache Hive through the
beeline or Hive JDBC driver, we need a configuration to control the binary output format.
#### Problem statement
Currently, the binary output format is inconsistent across different clients. For example,
- Hive beeline(spark thriftserver)
- For primitive binaries, we pass the binary directly to the clients, and then the clients will convert the binary based on the option`convertBinaryArrayToString`
- when convertBinaryArrayToString is true, results in UTF8 encoded strings, `[83, 112, 97, 114, 107] -> "Spark"`
- when convertBinaryArrayToString is false, Hive3-beeline results in comma-separated byte strings, `[83, 112, 97, 114, 107] -> [83, 112, 97, 114, 107]` Hive4-beeline, a base64 encoded string, `[83, 112, 97, 114, 107] -> U3Bhcmsg`
- For nested binaries, we pass the binary as UTF8 encoded strings
- Spark SQL CLI
- For both primitive and nested binaries, we print UTF8-encoded strings
- Spark Shell / pyspark shell
- We do a special `cast` to convert the binary to a string in space-separated hexadecimal format, `[83, 112, 97, 114, 107] -> "[53 70 61 72 6b]"`
**Given that no two behaviors are compatible or consistent, this could take you a lot of time to digest.**
Besides Apache Hive, other modern databases like Postgres, and MySQL, also support different binary output formats. The hexadecimal format is the most recommended format for binary output. `[83, 112, 97, 114, 107] -> "(0x)537061726b"`
FYI
- https://issues.apache.org/jira/browse/HIVE-14786
- https://issues.apache.org/jira/browse/HIVE-23856
- https://cwiki.apache.org/confluence/display/Hive/HiveServer2+Clients
- https://www.postgresql.org/docs/current/datatype-binary.html#DATATYPE-BINARY-BYTEA-HEX-FORMAT
- https://dev.mysql.com/doc/refman/8.3/en/mysql-command-options.html#option_mysql_binary-as-hex
### Why are the changes needed?
- A universal BinaryFormatter for consistensy
- A configuration for flexibility to align with Hive or other systems.
### Does this PR introduce _any_ user-facing change?
Yes, we have introduced a new configuration spark.sql.binaryOutputStyle but the AS-IS behavior is kept.
### How was this patch tested?
new tests
### Was this patch authored or co-authored using generative AI tooling?
no
Closes #46133 from yaooqinn/SPARK-47911.
Authored-by: Kent Yao <yao@apache.org>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>
Spark is a unified analytics engine for large-scale data processing. It provides high-level APIs in Scala, Java, Python, and R, and an optimized engine that supports general computation graphs for data analysis. It also supports a rich set of higher-level tools including Spark SQL for SQL and DataFrames, pandas API on Spark for pandas workloads, MLlib for machine learning, GraphX for graph processing, and Structured Streaming for stream processing.
You can find the latest Spark documentation, including a programming guide, on the project web page. This README file only contains basic setup instructions.
Spark is built using Apache Maven. To build Spark and its example programs, run:
./build/mvn -DskipTests clean package
(You do not need to do this if you downloaded a pre-built package.)
More detailed documentation is available from the project site, at “Building Spark”.
For general development tips, including info on developing Spark using an IDE, see “Useful Developer Tools”.
The easiest way to start using Spark is through the Scala shell:
./bin/spark-shell
Try the following command, which should return 1,000,000,000:
scala> spark.range(1000 * 1000 * 1000).count()
Alternatively, if you prefer Python, you can use the Python shell:
./bin/pyspark
And run the following command, which should also return 1,000,000,000:
>>> spark.range(1000 * 1000 * 1000).count()
Spark also comes with several sample programs in the examples directory. To run one of them, use ./bin/run-example <class> [params]. For example:
./bin/run-example SparkPi
will run the Pi example locally.
You can set the MASTER environment variable when running examples to submit examples to a cluster. This can be spark:// URL, “yarn” to run on YARN, and “local” to run locally with one thread, or “local[N]” to run locally with N threads. You can also use an abbreviated class name if the class is in the examples package. For instance:
MASTER=spark://host:7077 ./bin/run-example SparkPi
Many of the example programs print usage help if no params are given.
Testing first requires building Spark. Once Spark is built, tests can be run using:
./dev/run-tests
Please see the guidance on how to run tests for a module, or individual tests.
There is also a Kubernetes integration test, see resource-managers/kubernetes/integration-tests/README.md
Spark uses the Hadoop core library to talk to HDFS and other Hadoop-supported storage systems. Because the protocols have changed in different versions of Hadoop, you must build Spark against the same version that your cluster runs.
Please refer to the build documentation at “Specifying the Hadoop Version and Enabling YARN” for detailed guidance on building for a particular distribution of Hadoop, including building for particular Hive and Hive Thriftserver distributions.
Please refer to the Configuration Guide in the online documentation for an overview on how to configure Spark.
Please review the Contribution to Spark guide for information on how to get started contributing to the project.
