feat: unify to_timestamp/try_to_timestamp, honor ANSI, drop legacy UDF#2100
Open
davidlghellin wants to merge 3 commits into
Open
feat: unify to_timestamp/try_to_timestamp, honor ANSI, drop legacy UDF#2100davidlghellin wants to merge 3 commits into
to_timestamp/try_to_timestamp, honor ANSI, drop legacy UDF#2100davidlghellin wants to merge 3 commits into
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
Unifies Spark-compatible to_timestamp / try_to_timestamp (and _ntz) planning and execution by routing parsing through a single ANSI-aware SparkTimestamp UDF, removing the legacy SparkTryToTimestamp implementation. This aligns timestamp parsing behavior with spark.sql.ansi.enabled while keeping cluster-mode serialization working via updated physical plan codec/proto.
Changes:
- Added new BDD coverage for
to_timestamp(strict/ANSI-aware) andtry_to_timestamp(NULL-on-failure) including timezone and per-row format cases. - Refactored planner logic to share
to_timestamp/try_to_timestampimplementation and consistently passansi_modeintoSparkTimestamp. - Removed legacy
SparkTryToTimestampUDF and extended remote execution serialization (codec.rs+physical.proto) to carryansi_mode.
Reviewed changes
Copilot reviewed 11 out of 11 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| python/pysail/tests/spark/function/features/try_to_timestamp.feature | New BDD scenarios for safe timestamp parsing semantics. |
| python/pysail/tests/spark/function/features/to_timestamp.feature | New BDD scenarios for strict/ANSI-aware parsing + timezone/ntz behavior. |
| crates/sail-plan/src/resolver/expression/cast.rs | Passes ANSI mode into string→timestamp casting via SparkTimestamp. |
| crates/sail-plan/src/function/scalar/datetime.rs | Unifies planner path for to_timestamp/try_to_timestamp and uses SparkTimestamp for parsing. |
| crates/sail-plan/src/function/scalar/conversion.rs | Threads ANSI mode into cast_to_timestamp UDF planning. |
| crates/sail-plan/src/function/scalar/conditional.rs | Updates coalesce temporal coercion to use ANSI-strict SparkTimestamp. |
| crates/sail-function/src/scalar/datetime/spark_try_to_timestamp.rs | Removes the legacy UDF implementation. |
| crates/sail-function/src/scalar/datetime/spark_timestamp.rs | Extends SparkTimestamp to handle ANSI mode, try_* variants, and optional per-row formats. |
| crates/sail-function/src/scalar/datetime/mod.rs | Drops module export for removed legacy UDF. |
| crates/sail-execution/src/codec.rs | Updates physical plan codec to serialize/deserialize ansi_mode for SparkTimestamp. |
| crates/sail-execution/proto/sail/plan/physical.proto | Adds ansi_mode field to SparkTimestampUdf for cluster execution parity. |
Comment on lines
+9
to
+13
| When query | ||
| """ | ||
| SELECT to_timestamp('2024-01-15 10:30:45') AS result | ||
| """ | ||
| Then query result |
Comment on lines
+8
to
+13
| When query | ||
| """ | ||
| SELECT try_to_timestamp('2024-01-15 10:30:45') AS result | ||
| """ | ||
| Then query result | ||
| | result | |
Contributor
Author
There was a problem hiding this comment.
Fixed — reindented the docstring (""")
| Err(PlanError::invalid( | ||
| "try_to_timestamp requires 1 or 2 arguments", | ||
| )) | ||
| Err(PlanError::invalid("to_timestamp requires 1 or 2 arguments")) |
Comment on lines
+228
to
+240
| let mut coerced = arg_types.to_vec(); | ||
| if let Some(format) = arg_types.get(1) { | ||
| match format { | ||
| DataType::Utf8 | DataType::LargeUtf8 | DataType::Utf8View => {} | ||
| // A NULL format yields a NULL result; coerce it to a Utf8 null. | ||
| DataType::Null => coerced[1] = DataType::Utf8, | ||
| other => { | ||
| return Err(unsupported_data_type_exec_err(self.name(), "STRING", other)); | ||
| } | ||
| } | ||
| } | ||
| Ok(coerced) | ||
| } |
Contributor
Author
There was a problem hiding this comment.
Fixed — but I went with the "reject non-string types" option rather than coercing to Utf8.
Spark 3.5.7 Test ReportCommit Information
Test Summary
Test DetailsError CountsPassed Tests Diff(empty) Failed Tests |
Spark 4.1.1 Test ReportCommit Information
Test Summary
Test DetailsError CountsPassed Tests Diff(empty) Failed Tests(truncated) |
Codecov Report❌ Patch coverage is @@ Coverage Diff @@
## main #2100 +/- ##
==========================================
- Coverage 77.17% 76.03% -1.15%
==========================================
Files 1004 1004
Lines 162436 162505 +69
==========================================
- Hits 125365 123565 -1800
- Misses 37071 38940 +1869
*This pull request uses carry forward flags. Click here to find out more.
... and 51 files with indirect coverage changes 🚀 New features to boost your workflow:
|
Ibis Test ReportCommit Information
Test Summary
Test DetailsError CountsPassed Tests Diff--- before.txt 2026-06-18 18:27:01.670252618 +0000
+++ after.txt 2026-06-18 18:27:01.866254582 +0000
@@ -502,0 +503 @@
+ibis/backends/tests/test_export.py::test_table_to_csv[pyspark]Failed Tests |
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
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.
No description provided.