Support split eval function

[ahkcs]

Summary

This PR implements the split eval function for PPL, enabling users to split strings into multivalue arrays based on a delimiter.

Examples

Basic split with semicolon

source=people | eval result = split('a;b;c', ';')

Result: ['a', 'b', 'c']

---

Split into individual characters (empty delimiter)

source=people | eval result = split('abcd', '')

Result: ['a', 'b', 'c', 'd']

---

Multi-character delimiter

source=people | eval result = split('name::value', '::')

Result: ['name', 'value']

---

Split field value

source=people | eval words = split(employer, ' ')

Splits the employer field on spaces.

---

Check List

• New functionality includes testing.
• New functionality has been documented.
• New functionality has javadoc added.
• New functionality has a user manual doc added.
• New PPL command checklist all confirmed.
• API changes companion pull request created.
• Commits are signed per the DCO using --signoff or -s.
• Public documentation issue/PR created.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[dai-chen]

Can this be reused? https://issues.apache.org/jira/browse/CALCITE-6951

[ahkcs]

https://issues.apache.org/jira/browse/CALCITE-6951

Hi @dai-chen , currently we are using SPLIT, which is also supported by Calcite. And we did some special handling for empty delimiter on top of that.

Here's the documentation for Calcite SPLIT function:

SPLIT(string [, delimiter ]) 

Returns the string array of string split at delimiter (if omitted, default is comma). If the string is empty it returns an empty array, otherwise, if the delimiter is empty, it returns an array containing the original string.

[dai-chen]

https://issues.apache.org/jira/browse/CALCITE-6951

Hi @dai-chen , currently we are using SPLIT, which is also supported by Calcite. And we did some special handling for empty delimiter on top of that.

Here's the documentation for Calcite SPLIT function:

SPLIT(string [, delimiter ]) 

Returns the string array of string split at delimiter (if omitted, default is comma). If the string is empty it returns an empty array, otherwise, if the delimiter is empty, it returns an array containing the original string.

I see. So the only reason of SplitFunctionImp is special handling for delimiter="", right?
Can the extract function below help?

• Case 1: Delimiter is not empty string, translate split to

SELECT SPLIT('a;b;c;d', ';');
+--------------+
|    EXPR$0    |
+--------------+
| [a, b, c, d] |
+--------------+

• Case 2: Delimiter is empty string, translate split to:

SELECT REGEXP_EXTRACT_ALL('abcd', '.');
+--------------+
|    EXPR$0    |
+--------------+
| [a, b, c, d] |
+--------------+

[ahkcs]

@dai-chen Thanks for the suggestion! I think it makes sense. I have updated the PR to move the implementation to PPLFuncImpTable

[dai-chen]

Thanks for the changes!