CySQL Translation Fixes#76
Merged
Merged
Conversation
…risons - BED-5945
|
Warning Rate limit exceeded
You’ve run out of usage credits. Purchase more in the billing tab. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
WalkthroughImplements Cypher HEAD/TAIL/NODES/RELATIONSHIPS translation and zero-depth (*0..) expansions, with array-slice formatting/walking, projection/renamer updates, temporal type inference, quantifier/property handling, Neo4j query-rewrite, SQL helper, and extensive tests. ChangesFeature Implementation
Sequence Diagram(s)sequenceDiagram
participant Cypher as Cypher Parser
participant Translator as PG Translator
participant TypeInfer as Type Inference
participant Expander as ExpansionBuilder
participant Renamer as FrameBindingRewriter
participant SQL as PostgreSQL Output
Cypher->>Translator: HEAD(arr)/TAIL(arr)/NODES(p)/RELATIONSHIPS(p)
Translator->>TypeInfer: infer array/path element or composite type
TypeInfer-->>Translator: DataType (with CastType hint)
Translator->>SQL: emit array index/slice or casted row-column reference
Cypher->>Expander: expansion with MinDepth=0
Expander->>TypeInfer: infer terminal satisfaction/locality
Expander->>Renamer: rewrite bound endpoint seeds (ArraySlice) as needed
Expander->>SQL: build zero-depth SELECT and union with recursive body
Translator->>Renamer: emit expressions that may be scope-rewritten
Renamer->>SQL: final rewritten AST -> SQL output
🎯 4 (Complex) | ⏱️ ~60 minutes
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Tip 💬 Introducing Slack Agent: The best way for teams to turn conversations into code.Slack Agent is built on CodeRabbit's deep understanding of your code, so your team can collaborate across the entire SDLC without losing context.
Built for teams:
One agent for your entire SDLC. Right inside Slack. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 8
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@cypher/models/pgsql/pgtypes.go`:
- Around line 119-127: IsTemporalType currently treats Date as a temporal
leading to mis-typing of Date ± Interval; remove Date from
DataType.IsTemporalType() so Date is not treated as a generic temporal, then
update OperatorResultType() to explicitly handle date/interval arithmetic: add
clauses that map Date + Interval and Date - Interval to TimestampWithoutTimeZone
(and leave other temporal combos alone), remove the old Date case in
OperatorResultType() that returned Date for those operations, and adjust the
Interval case logic in OperatorResultType() to account for the new explicit
Date+Interval handling so Interval results stay correct.
In `@cypher/models/pgsql/test/translation_cases/multipart.sql`:
- Line 51: The CTEs s2 and s5 include an explicit "one-hop" union arm (the
SELECT ... from s2_seed join edge e0 ... producing depth = 1 and the equivalent
SELECT ... from s5_seed join edge e2 ...) which duplicates the same results
produced by the recursive branch (s2 join lateral ... and s5 join lateral ...
with s?.depth + 1). Remove those explicit depth=1 seed union arms (the clauses
referencing s2_seed join edge e0 and s5_seed join edge e2) so each recursive CTE
only has the initial seed row (depth = 0 or the seed) plus the recursive
union-all step, leaving the lateral recursive logic and path construction
(s2.path, s5.path, e0/e2 handling) intact.
In `@cypher/models/pgsql/translate/expansion.go`:
- Around line 590-606: The zero-hop branch currently uses a local predicate
returned by expansionTerminalSatisfactionLocality even if that predicate
references the traversal edge alias, which is not in scope for depth=0; update
the block handling expansionModel.TerminalNodeSatisfactionProjection so that
after computing localSatisfaction :=
expansionTerminalSatisfactionLocality(s.traversalStep) you check if
referencesIdentifier(localSatisfaction, s.traversalStep.Edge.Identifier) and, if
true, treat localSatisfaction as nil (i.e., skip setting satisfiedExpression and
skip adding zeroDepthNodeJoin entries to fromClause.Joins); ensure all uses in
this block (satisfiedExpression and the two join appends) are conditional on
localSatisfaction being non-nil so edge-dependent terminal predicates do not
drive a zero-depth select.
In `@cypher/models/pgsql/translate/expression.go`:
- Around line 56-60: The code currently returns nil for non-composite
expressions in the default case after calling
s.inferExpressionType(translatedAtom), which silently discards an unsupported
property lookup; instead, detect when
expressionHasCompositeProperties(expressionType) is false and return a clear
error (e.g., "unsupported property lookup on non-composite expression") rather
than nil so the caller sees a direct error; update the default branch in
expression.go to return that error (referencing translatedAtom,
s.inferExpressionType, and expressionHasCompositeProperties) instead of
returning nil.
In `@cypher/models/pgsql/translate/function.go`:
- Around line 134-141: expressionForPath currently only checks s.scope.Lookup
for the identifier and fails on aliased identifiers; update expressionForPath to
resolve identifier aliases before validating the binding type. After obtaining
binding := s.scope.Lookup(typedExpression), if the binding represents an alias
(use the scope's alias/link fields or a ResolveAlias/OriginalIdentifier property
that inferExpressionType already handles), follow the alias to the original
binding (iteratively if needed) and then verify binding.DataType ==
pgsql.PathComposite and call expressionForPathComposite(binding, s.scope); this
ensures WITH p AS q RETURN nodes(q) / relationships(q) resolve correctly.
In `@cypher/models/pgsql/translate/quantifiers.go`:
- Around line 93-97: The code currently hard-fails inside the switch on
quantifierArrayExpression when encountering a *pgsql.BinaryExpression that is
not a property lookup (checked via pgsql.OperatorIsPropertyLookup), returning
"unknown cypher array type" and preventing the later inference path; remove that
early return and allow non-property binary expressions to fall through into the
inference logic (the same change should be applied to the other similar branches
around the other *pgsql.BinaryExpression checks at the noted spots) so that
non-property binaries are inferred instead of rejected.
In `@cypher/models/pgsql/translate/renamer.go`:
- Around line 182-229: The switch only handles the pointer form
(*pgsql.ArraySlice) but not the value form (pgsql.ArraySlice), so value-form
nodes skip rewriting Expression/Lower/Upper and keep stale frame refs; add a new
case for pgsql.ArraySlice that mirrors the logic in the existing
*pgsql.ArraySlice case, calling rewriteIdentifierScopeReference and
rewriteCompoundIdentifierScopeReference with s.scope for
typedExpression.Expression, typedExpression.Lower and typedExpression.Upper so
all identifier/compound-identifier variants are rebound correctly.
In `@integration/testdata/cases/pattern_binding_inline.json`:
- Around line 214-240: The test fixture doesn't exercise the HEAD(r).enforced
branch because the enforced path is length 1 so TAIL(TAIL(NODES(p))) is empty;
add a longer enforced path that contains a later OU with blocksinheritance=true
so NONE(...) would be false without HEAD(r).enforced. Concretely, add a node
like "ou-blocked-2" (kinds ["OU"], properties {"blocksinheritance": true}) and
an edge from "gpo-enforced" to "ou-blocked-2" (kind "GPLink", properties
{"enforced": true}) and another edge from "ou-blocked-2" to "base-enforced"
(kind "Contains"), then update the expected "path_node_ids" to include the
enforced multi-hop path (e.g., ["gpo-enforced","ou-blocked-2","base-enforced"])
and corresponding "path_edge_kinds" (["GPLink","Contains"]) so the test proves
HEAD(r).enforced is required.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 46246fd4-3b4f-4f91-b37e-73dae887e725
📒 Files selected for processing (22)
cypher/models/cypher/functions.gocypher/models/pgsql/format/format.gocypher/models/pgsql/model.gocypher/models/pgsql/pgtypes.gocypher/models/pgsql/pytypes_test.gocypher/models/pgsql/test/translation_cases/multipart.sqlcypher/models/pgsql/test/translation_cases/pattern_binding.sqlcypher/models/pgsql/test/translation_cases/pattern_expansion.sqlcypher/models/pgsql/test/translation_cases/shortest_paths.sqlcypher/models/pgsql/test/translation_test.gocypher/models/pgsql/translate/expansion.gocypher/models/pgsql/translate/expression.gocypher/models/pgsql/translate/function.gocypher/models/pgsql/translate/projection.gocypher/models/pgsql/translate/quantifiers.gocypher/models/pgsql/translate/renamer.gocypher/models/pgsql/translate/with.gocypher/models/walk/walk_pgsql.gointegration/testdata/cases/expansion_inline.jsonintegration/testdata/cases/pattern_binding_inline.jsonintegration/testdata/cases/shortest_paths_inline.jsonintegration/testdata/cases/temporal_inline.json
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
cypher/models/pgsql/translate/renamer.go (1)
97-113:⚠️ Potential issue | 🟠 Major | ⚡ Quick winAdd
pgsql.ArraySlicehandling inpgsql.Projectionand*pgsql.OrderBybranches.The walker visits both value-form and pointer-form
pgsql.ArraySlicenodes, butpgsql.Projection(lines 97–113) and*pgsql.OrderBy(lines 268–276) lack cases for them. When anArraySliceappears directly in either position, the rewrite is lost because the parent does not write back the mutated value. Other parent types likepgsql.CompositeValueandpgsql.FunctionCallalready include explicit write-back for both forms.Suggested fix
case pgsql.Projection: for idx, projection := range typedExpression { switch typedProjection := projection.(type) { case pgsql.Identifier: if rewritten, err := rewriteIdentifierScopeReference(s.scope, typedProjection); err != nil { return err } else { typedExpression[idx] = rewritten } case pgsql.CompoundIdentifier: if rewritten, err := rewriteCompoundIdentifierScopeReference(s.scope, typedProjection); err != nil { return err } else { typedExpression[idx] = rewritten } + + case pgsql.ArraySlice: + if err := s.rewriteArraySlice(&typedProjection); err != nil { + return err + } + typedExpression[idx] = typedProjection + + case *pgsql.ArraySlice: + if err := s.rewriteArraySlice(typedProjection); err != nil { + return err + } } } case *pgsql.OrderBy: switch typedOrderByExpression := typedExpression.Expression.(type) { case pgsql.Identifier: if rewritten, err := rewriteIdentifierScopeReference(s.scope, typedOrderByExpression); err != nil { return err } else { typedExpression.Expression = rewritten } case pgsql.CompoundIdentifier: if rewritten, err := rewriteCompoundIdentifierScopeReference(s.scope, typedOrderByExpression); err != nil { return err } else { typedExpression.Expression = rewritten } + + case pgsql.ArraySlice: + if err := s.rewriteArraySlice(&typedOrderByExpression); err != nil { + return err + } + typedExpression.Expression = typedOrderByExpression + + case *pgsql.ArraySlice: + if err := s.rewriteArraySlice(typedOrderByExpression); err != nil { + return err + } }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@cypher/models/pgsql/translate/renamer.go` around lines 97 - 113, Handle pgsql.ArraySlice in the pgsql.Projection and *pgsql.OrderBy branches: when iterating pgsql.Projection (inside the case for pgsql.Projection) add cases for both pgsql.ArraySlice and *pgsql.ArraySlice, call the appropriate rewrite (same pattern used for Identifier/CompoundIdentifier) and write the rewritten node back into typedExpression (e.g., typedExpression[idx] = rewritten or typedExpression[idx] = &rewritten for pointer form); similarly, in the *pgsql.OrderBy branch add handling for both pgsql.ArraySlice and *pgsql.ArraySlice for the order field and assign the rewritten value back into typedOrderBy.Field (using value or pointer as appropriate) so the parent retains the mutated node.
♻️ Duplicate comments (1)
cypher/models/pgsql/translate/expansion.go (1)
590-605:⚠️ Potential issue | 🟠 Major | ⚡ Quick winDon't carry edge-bound terminal predicates into the zero-hop arm.
expansionTerminalSatisfactionLocalitycan still return a predicate that referencess.traversalStep.Edge.Identifier, but the depth-0 SELECT never has that alias in scope. The new zero-depth branch will therefore emit invalid SQL for relationship-aware terminal predicates. Leave the zero-hop row at its defaultfalsesatisfaction in that case, and only add the node joins when the remaining local predicate is still usable.Suggested fix
if expansionModel.TerminalNodeSatisfactionProjection != nil { localSatisfaction, _ := expansionTerminalSatisfactionLocality(s.traversalStep) - if localSatisfaction == nil { + if localSatisfaction != nil && + referencesIdentifier(localSatisfaction, s.traversalStep.Edge.Identifier) { + localSatisfaction = nil + } else if localSatisfaction == nil { localSatisfaction = pgsql.NewLiteral(true, pgsql.Boolean) } - satisfiedExpression = localSatisfaction - - if seed != nil && referencesIdentifier(localSatisfaction, s.traversalStep.LeftNode.Identifier) { - fromClause.Joins = append(fromClause.Joins, zeroDepthNodeJoin(s.traversalStep.LeftNode.Identifier, rootIDExpression)) - } - - if s.traversalStep.RightNode.Identifier != s.traversalStep.LeftNode.Identifier && - referencesIdentifier(localSatisfaction, s.traversalStep.RightNode.Identifier) { - fromClause.Joins = append(fromClause.Joins, zeroDepthNodeJoin(s.traversalStep.RightNode.Identifier, rootIDExpression)) + if localSatisfaction != nil { + satisfiedExpression = localSatisfaction + + if seed != nil && referencesIdentifier(localSatisfaction, s.traversalStep.LeftNode.Identifier) { + fromClause.Joins = append(fromClause.Joins, zeroDepthNodeJoin(s.traversalStep.LeftNode.Identifier, rootIDExpression)) + } + + if s.traversalStep.RightNode.Identifier != s.traversalStep.LeftNode.Identifier && + referencesIdentifier(localSatisfaction, s.traversalStep.RightNode.Identifier) { + fromClause.Joins = append(fromClause.Joins, zeroDepthNodeJoin(s.traversalStep.RightNode.Identifier, rootIDExpression)) + } } }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@cypher/models/pgsql/translate/expansion.go` around lines 590 - 605, expansionTerminalSatisfactionLocality may return a predicate that references the edge alias (s.traversalStep.Edge.Identifier), which is not in scope for the zero-depth SELECT; change the logic in the TerminalNodeSatisfactionProjection branch so you first compute localSatisfaction via expansionTerminalSatisfactionLocality(s.traversalStep) and then validate that localSatisfaction does NOT reference the edge identifier (using referencesIdentifier(localSatisfaction, s.traversalStep.Edge.Identifier)); if it does reference the edge, discard it and treat zero-hop satisfaction as the default false literal (pgsql.NewLiteral(false,...)) and do NOT add any zero-depth node joins; only append zeroDepthNodeJoin entries to fromClause.Joins when the retained localSatisfaction is usable and still references the relevant node identifiers (use existing referencesIdentifier checks for LeftNode.Identifier and RightNode.Identifier).
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@cypher/models/pgsql/translate/renamer.go`:
- Around line 97-113: Handle pgsql.ArraySlice in the pgsql.Projection and
*pgsql.OrderBy branches: when iterating pgsql.Projection (inside the case for
pgsql.Projection) add cases for both pgsql.ArraySlice and *pgsql.ArraySlice,
call the appropriate rewrite (same pattern used for
Identifier/CompoundIdentifier) and write the rewritten node back into
typedExpression (e.g., typedExpression[idx] = rewritten or typedExpression[idx]
= &rewritten for pointer form); similarly, in the *pgsql.OrderBy branch add
handling for both pgsql.ArraySlice and *pgsql.ArraySlice for the order field and
assign the rewritten value back into typedOrderBy.Field (using value or pointer
as appropriate) so the parent retains the mutated node.
---
Duplicate comments:
In `@cypher/models/pgsql/translate/expansion.go`:
- Around line 590-605: expansionTerminalSatisfactionLocality may return a
predicate that references the edge alias (s.traversalStep.Edge.Identifier),
which is not in scope for the zero-depth SELECT; change the logic in the
TerminalNodeSatisfactionProjection branch so you first compute localSatisfaction
via expansionTerminalSatisfactionLocality(s.traversalStep) and then validate
that localSatisfaction does NOT reference the edge identifier (using
referencesIdentifier(localSatisfaction, s.traversalStep.Edge.Identifier)); if it
does reference the edge, discard it and treat zero-hop satisfaction as the
default false literal (pgsql.NewLiteral(false,...)) and do NOT add any
zero-depth node joins; only append zeroDepthNodeJoin entries to fromClause.Joins
when the retained localSatisfaction is usable and still references the relevant
node identifiers (use existing referencesIdentifier checks for
LeftNode.Identifier and RightNode.Identifier).
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 104ba301-1150-467b-b2b9-8770be3f7d4f
📒 Files selected for processing (15)
cypher/models/pgsql/pgtypes.gocypher/models/pgsql/pytypes_test.gocypher/models/pgsql/test/translation_cases/multipart.sqlcypher/models/pgsql/test/translation_cases/nodes.sqlcypher/models/pgsql/test/translation_cases/pattern_expansion.sqlcypher/models/pgsql/test/translation_cases/quantifiers.sqlcypher/models/pgsql/translate/expansion.gocypher/models/pgsql/translate/expression.gocypher/models/pgsql/translate/quantifiers.gocypher/models/pgsql/translate/renamer.gocypher/models/pgsql/translate/renamer_test.gocypher/models/pgsql/translate/semantic_drift_test.gointegration/testdata/cases/expansion_inline.jsonintegration/testdata/cases/pattern_binding_inline.jsonintegration/testdata/cases/quantifiers_inline.json
✅ Files skipped from review due to trivial changes (1)
- integration/testdata/cases/quantifiers_inline.json
🚧 Files skipped from review as they are similar to previous changes (7)
- cypher/models/pgsql/pytypes_test.go
- cypher/models/pgsql/test/translation_cases/pattern_expansion.sql
- integration/testdata/cases/pattern_binding_inline.json
- cypher/models/pgsql/pgtypes.go
- cypher/models/pgsql/translate/expression.go
- integration/testdata/cases/expansion_inline.json
- cypher/models/pgsql/translate/quantifiers.go
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@drivers/neo4j/query_rewrite.go`:
- Around line 51-55: Add a cheap fast-path guard in rewriteQuery to avoid
parsing queries that cannot match temporal rewrites: implement a helper like
containsTemporalRewriteCandidate(query string) bool (using strings.ToLower and
substr checks for "date(", "time(", "localtime(", "datetime(", "localdatetime(")
and call it at the top of rewriteQuery; if it returns false immediately return
the original query and parameters without calling frontend.ParseCypher. Remember
to import the strings package and keep the existing rewriteQuery signature and
error behavior unchanged.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: e57eae66-c2c1-4925-99d7-2eb14a6426ec
📒 Files selected for processing (11)
cypher/models/pgsql/model.gocypher/models/pgsql/translate/expansion.gocypher/models/pgsql/translate/expansion_test.gocypher/models/pgsql/translate/function.gocypher/models/pgsql/translate/function_test.gocypher/models/pgsql/translate/model.gocypher/models/pgsql/translate/renamer.gocypher/models/pgsql/translate/renamer_test.godrivers/neo4j/query_rewrite.godrivers/neo4j/query_rewrite_internal_test.godrivers/neo4j/transaction.go
🚧 Files skipped from review as they are similar to previous changes (4)
- cypher/models/pgsql/model.go
- cypher/models/pgsql/translate/function.go
- cypher/models/pgsql/translate/expansion.go
- cypher/models/pgsql/translate/renamer.go
…r zero hop traversals and use a more explicit AST tool
…ed to be rewritten
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@cypher/models/pgsql/translate/expansion.go`:
- Around line 1939-1954: The code currently replaces the original primer ∪
recursive set with zeroDepth ∪ recursive when ZeroDepthStatement is present,
dropping the primer branch and changing first-hop behavior; instead keep the
primer branch as the left side of the overall expansion. Modify
ExpansionBuilder.Build so that when s.ZeroDepthStatement != nil you construct
expansionBody as a SetOperation whose LOperand preserves the primer branch (e.g.
a SetExpression combining s.PrimerStatement and *s.ZeroDepthStatement with the
same Operator and s.UseUnionAll) and whose ROperand remains
s.RecursiveStatement, ensuring PrimerStatement, ZeroDepthStatement,
RecursiveStatement and UseUnionAll are used to compose the nested SetExpression
rather than replacing the primer entirely.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 92c1e731-8b6a-408c-acac-35cd1e279ed7
📒 Files selected for processing (15)
cypher/models/pgsql/format/format.gocypher/models/pgsql/format/format_test.gocypher/models/pgsql/functions.gocypher/models/pgsql/model.gocypher/models/pgsql/test/translation_cases/shortest_paths.sqlcypher/models/pgsql/test/validation_integration_test.gocypher/models/pgsql/translate/expansion.gocypher/models/pgsql/translate/expansion_test.gocypher/models/pgsql/translate/renamer.gocypher/models/pgsql/translate/renamer_test.gocypher/models/walk/walk_pgsql.godrivers/neo4j/query_rewrite.godrivers/neo4j/query_rewrite_internal_test.godrivers/pg/query/sql/schema_down.sqldrivers/pg/query/sql/schema_up.sql
✅ Files skipped from review due to trivial changes (1)
- cypher/models/pgsql/format/format_test.go
🚧 Files skipped from review as they are similar to previous changes (4)
- drivers/neo4j/query_rewrite_internal_test.go
- drivers/neo4j/query_rewrite.go
- cypher/models/pgsql/translate/renamer.go
- cypher/models/pgsql/translate/renamer_test.go
15 tasks
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.
Description
Adds PostgreSQL Cypher translation fixes for recent query-shape defects: zero-hop variable-length expansions, temporal arithmetic comparisons against relationship properties, path list functions in expansion predicates, and shortest path aggregation alias preservation. Adds translation and integration coverage for the reported shapes.
Resolves:
BP-2518Type of Change
Testing
make test_allwithCONNECTION_STRINGset)Screenshots (if appropriate):
N/A
Driver Impact
drivers/pg)drivers/neo4j)Checklist
go.mod/go.sumare up to date if dependencies changedSummary by CodeRabbit
New Features
Behavioral Improvements
Tests