[GH-2983] Box3D SQL parser keyword + Geometry→Box3D cast resolution#3016
Open
jiayuasu wants to merge 1 commit into
Open
[GH-2983] Box3D SQL parser keyword + Geometry→Box3D cast resolution#3016jiayuasu wants to merge 1 commit into
jiayuasu wants to merge 1 commit into
Conversation
…tion Adds the parser keyword and analyzer rule that let `CAST(... AS box3d)` resolve, mirroring the Box2D parser/cast support landed in apache#2927. - `SedonaSqlAstBuilder.visitPrimitiveDataType` recognises `BOX3D` alongside the existing `GEOMETRY` and `BOX2D` keywords, returning `Box3DUDT`. Applied uniformly to the spark-3.4 / 3.5 / 4.0 / 4.1 parser variants. - `Box3DCastResolutionRule` rewrites `Cast(geom, Box3DUDT)` → `ST_Box3D(geom)` during analysis (before `CheckAnalysis`), since Spark's stock `Cast.canCast` refuses arbitrary UDT-to-UDT casts. Registered from `SedonaSqlExtensions.injectResolutionRule`. - Only the forward direction (`CAST(geom AS box3d)`) is wired. The inverse cast (`CAST(box3d AS geometry)`) is deferred until Box3D has an `ST_GeomFromBox3D` counterpart driven by a concrete consumer — Box2D shipped both directions because `ST_GeomFromBox2D` already existed, which is not the case here. Tests: - `Box3DCastResolutionRuleSuite` (rule unit test) — covers the forward rewrite, leaves the inverse Cast untouched (out of scope), and leaves unrelated casts alone. - `Box3DCastSuite` per spark-3.4 / 3.5 / 4.0 / 4.1 — DataFrame `.cast(Box3DUDT)` plus SQL `CAST(... AS box3d)` end-to-end, gated on the parser-extension probe (matches `Box2DCastSuite`'s pattern across the same four matrix cells).
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
1 participant
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.
Did you read the Contributor Guide?
Is this PR related to a ticket?
What changes were proposed in this PR?
Adds the parser keyword and analyzer rule that let
CAST(... AS box3d)resolve, mirroring the Box2D parser/cast support landed in #2927.SedonaSqlAstBuilder.visitPrimitiveDataTyperecognisesBOX3Dalongside the existingGEOMETRYandBOX2Dkeywords, returningBox3DUDT. Applied uniformly to the spark-3.4 / 3.5 / 4.0 / 4.1 parser variants.Box3DCastResolutionRulerewritesCast(geom, Box3DUDT)→ST_Box3D(geom)during analysis (beforeCheckAnalysis), since Spark's stockCast.canCastrefuses arbitrary UDT-to-UDT casts. Registered fromSedonaSqlExtensions.injectResolutionRule.CAST(geom AS box3d)) is wired. The inverse cast (CAST(box3d AS geometry)) is deferred until Box3D has anST_GeomFromBox3Dcounterpart driven by a concrete consumer — Box2D shipped both directions becauseST_GeomFromBox2Dalready existed, which is not the case here. The issue tracks this rationale.How was this patch tested?
Box3DCastResolutionRuleSuite(rule unit test inspark/common) — verifies the forward rewrite, that the inverseCast(box3d, Geometry)is left untouched (out of scope), and that unrelated casts are not affected.Box3DCastSuiteadded per spark-3.4 / 3.5 / 4.0 / 4.1 — DataFrame.cast(Box3DUDT)plus SQLCAST(... AS box3d)end-to-end, gated on the parser-extension probe (matches the existingBox2DCastSuitepattern across the same four matrix cells).Did this PR include necessary documentation updates?