Uh oh!
There was an error while loading. Please reload this page.
fix: isEmpty function when using postgresql dialect - #2703
Merged
Conversation
Contributor
ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughThis PR fixes PostgreSQL's ChangesPostgreSQL Array Length Dimension
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
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 |
sanny-io
commented
Jun 9, 2026
ContributorAuthor
Context: For https://discord.com/channels/1035538056146595961/1513844077383258122/1513844077383258122 |
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
packages/orm/src/client/crud/dialects/postgresql.ts (1)
415-422: ⚡ Quick winRemove commented-out dead code.
The commented-out enum handling block (lines 417-419) is now dead code since the logic below handles all cases including enums. Leaving commented-out code creates confusion about intent and clutters the codebase.
♻️ Proposed cleanup
override buildArrayValue(values: Expression<unknown>[], elemType: string): AliasableExpression<unknown> { const arr = sql`ARRAY[${sql.join(values, sql.raw(','))}]`; - // if (isEnum(this.schema, elemType)) {- // return this.eb.cast(arr, sql`${sql.id(elemType)}[]`);- // } const mappedType = this.getSqlType(elemType); return this.eb.cast(arr, sql`${sql.id(mappedType)}[]`); }🤖 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 `@packages/orm/src/client/crud/dialects/postgresql.ts` around lines 415 - 422, Remove the commented-out enum handling inside buildArrayValue: delete the commented lines that call isEnum(this.schema, elemType) and the associated eb.cast fallback so the function only contains the active implementation using getSqlType and eb.cast; leave buildArrayValue, getSqlType and eb.cast unchanged and ensure no other references to the removed commented code remain.
🤖 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.
Nitpick comments:
In `@packages/orm/src/client/crud/dialects/postgresql.ts`:
- Around line 415-422: Remove the commented-out enum handling inside
buildArrayValue: delete the commented lines that call isEnum(this.schema,
elemType) and the associated eb.cast fallback so the function only contains the
active implementation using getSqlType and eb.cast; leave buildArrayValue,
getSqlType and eb.cast unchanged and ensure no other references to the removed
commented code remain.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 8dd49a34-6242-43f4-99d0-766ff6cbed89
📒 Files selected for processing (2)
packages/orm/src/client/crud/dialects/base-dialect.tspackages/orm/src/client/crud/dialects/postgresql.ts
sanny-io
marked this pull request as draft
June 9, 2026 15:49
isEmpty function and in expressions with enums using postgresql dialectisEmpty function when using postgresql dialectsanny-io
commented
Jun 9, 2026
ContributorAuthor
I believe the issue may only happen when using access policies, so I have relocated the test. |
sanny-io
marked this pull request as ready for review
June 9, 2026 17:40
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
tests/e2e/orm/policy/isempty-function-pg.test.ts (1)
4-36: ⚡ Quick winConsider adding test cases for edge conditions.
The current test validates that
isEmpty(roles)works when the array contains one element. For more comprehensive coverage, consider adding test cases for:
- Empty array: Verify that creating a user with
roles: []is denied by the policy- Multiple roles: Confirm behavior with
roles: ['AUTHOR', 'EDITOR']- Implicit vs explicit behavior: Document whether the test relies on
@@allow('all', true)to handle the non-empty case, or if the deny-only policy is sufficientThese additions would help prevent regressions and clarify the expected policy behavior.
📋 Suggested additional test cases
it('denies creation when roles is empty',async()=>{constschema=`model User { id Int `@id` `@default`(autoincrement()) roles Role[] @@allow('all', true) @@deny('create', isEmpty(roles))}enum Role { AUTHOR EDITOR} `;constclient=awaitcreatePolicyTestClient(schema,{usePrismaPush: true,provider: 'postgresql',});awaitexpect(client.user.create({data: {id: 2,roles: [],},})).rejects.toThrow();});it('allows creation with multiple roles',async()=>{// ... test with roles: ['AUTHOR', 'EDITOR']});🤖 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 `@tests/e2e/orm/policy/isempty-function-pg.test.ts` around lines 4 - 36, Add edge-case tests around the isEmpty(roles) policy: add an it block that constructs the same schema (with @@deny('create', isEmpty(roles))) and asserts that client.user.create with roles: [] is rejected, and add another it block that asserts client.user.create with roles: ['AUTHOR','EDITOR'] resolves and returns the created user; use the same createPolicyTestClient call and client.user.create calls already present in the file (referencing createPolicyTestClient and client.user.create) and keep the existing schema (or add a variant that omits @@allow('all', true) if you want to assert behavior without the allow rule) so tests cover empty-array denial and multi-role acceptance.
🤖 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 `@packages/orm/src/client/crud/dialects/postgresql.ts`:
- Around line 502-504: Add an e2e test for Postgres that verifies enum-array
membership operators work when enums are mapped to text (so the cast to text[]
in buildArrayFilter preserves semantics): create a schema with enum Role and
field roles Role[], insert records with various role combinations, then run
queries using the array filters has, hasSome, and hasEvery against roles and
assert expected matches for provider: 'postgresql' (mirror the style of
tests/e2e/orm/policy/isempty-function-pg.test.ts); ensure the test targets
behavior in postgresql.ts (isEnum) and base-dialect.ts (buildArrayFilter) by
asserting correct runtime results for each operator.
---
Nitpick comments:
In `@tests/e2e/orm/policy/isempty-function-pg.test.ts`:
- Around line 4-36: Add edge-case tests around the isEmpty(roles) policy: add an
it block that constructs the same schema (with @@deny('create', isEmpty(roles)))
and asserts that client.user.create with roles: [] is rejected, and add another
it block that asserts client.user.create with roles: ['AUTHOR','EDITOR']
resolves and returns the created user; use the same createPolicyTestClient call
and client.user.create calls already present in the file (referencing
createPolicyTestClient and client.user.create) and keep the existing schema (or
add a variant that omits @@allow('all', true) if you want to assert behavior
without the allow rule) so tests cover empty-array denial and multi-role
acceptance.
🪄 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: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 23ea969d-e58f-4475-8cc2-ebc2a46f0133
📒 Files selected for processing (2)
packages/orm/src/client/crud/dialects/postgresql.tstests/e2e/orm/policy/isempty-function-pg.test.ts
Contributor
There was a problem hiding this comment.
Caution
Inline review comments failed to post. This is likely due to GitHub's internal server error or limits when posting large numbers of comments. If you are seeing this consistently it is likely a permissions issue. Please check "Moderation" -> "Code review limits" under your organization settings.
Actionable comments posted: 1
🧹 Nitpick comments (1)
tests/e2e/orm/policy/isempty-function-pg.test.ts (1)
4-36: ⚡ Quick winConsider adding test cases for edge conditions.
The current test validates that
isEmpty(roles)works when the array contains one element. For more comprehensive coverage, consider adding test cases for:
- Empty array: Verify that creating a user with
roles: []is denied by the policy- Multiple roles: Confirm behavior with
roles: ['AUTHOR', 'EDITOR']- Implicit vs explicit behavior: Document whether the test relies on
@@allow('all', true)to handle the non-empty case, or if the deny-only policy is sufficientThese additions would help prevent regressions and clarify the expected policy behavior.
📋 Suggested additional test cases
it('denies creation when roles is empty',async()=>{constschema=`model User { id Int `@id` `@default`(autoincrement()) roles Role[] @@allow('all', true) @@deny('create', isEmpty(roles))}enum Role { AUTHOR EDITOR} `;constclient=awaitcreatePolicyTestClient(schema,{usePrismaPush: true,provider: 'postgresql',});awaitexpect(client.user.create({data: {id: 2,roles: [],},})).rejects.toThrow();});it('allows creation with multiple roles',async()=>{// ... test with roles: ['AUTHOR', 'EDITOR']});🤖 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 `@tests/e2e/orm/policy/isempty-function-pg.test.ts` around lines 4 - 36, Add edge-case tests around the isEmpty(roles) policy: add an it block that constructs the same schema (with @@deny('create', isEmpty(roles))) and asserts that client.user.create with roles: [] is rejected, and add another it block that asserts client.user.create with roles: ['AUTHOR','EDITOR'] resolves and returns the created user; use the same createPolicyTestClient call and client.user.create calls already present in the file (referencing createPolicyTestClient and client.user.create) and keep the existing schema (or add a variant that omits @@allow('all', true) if you want to assert behavior without the allow rule) so tests cover empty-array denial and multi-role acceptance.
🤖 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 `@packages/orm/src/client/crud/dialects/postgresql.ts`:
- Around line 502-504: Add an e2e test for Postgres that verifies enum-array
membership operators work when enums are mapped to text (so the cast to text[]
in buildArrayFilter preserves semantics): create a schema with enum Role and
field roles Role[], insert records with various role combinations, then run
queries using the array filters has, hasSome, and hasEvery against roles and
assert expected matches for provider: 'postgresql' (mirror the style of
tests/e2e/orm/policy/isempty-function-pg.test.ts); ensure the test targets
behavior in postgresql.ts (isEnum) and base-dialect.ts (buildArrayFilter) by
asserting correct runtime results for each operator.
---
Nitpick comments:
In `@tests/e2e/orm/policy/isempty-function-pg.test.ts`:
- Around line 4-36: Add edge-case tests around the isEmpty(roles) policy: add an
it block that constructs the same schema (with @@deny('create', isEmpty(roles)))
and asserts that client.user.create with roles: [] is rejected, and add another
it block that asserts client.user.create with roles: ['AUTHOR','EDITOR']
resolves and returns the created user; use the same createPolicyTestClient call
and client.user.create calls already present in the file (referencing
createPolicyTestClient and client.user.create) and keep the existing schema (or
add a variant that omits @@allow('all', true) if you want to assert behavior
without the allow rule) so tests cover empty-array denial and multi-role
acceptance.
🪄 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: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 23ea969d-e58f-4475-8cc2-ebc2a46f0133
📒 Files selected for processing (2)
packages/orm/src/client/crud/dialects/postgresql.tstests/e2e/orm/policy/isempty-function-pg.test.ts
🛑 Comments failed to post (1)
packages/orm/src/client/crud/dialects/postgresql.ts (1)
502-504:
⚠️ Potential issue | 🟡 Minor🧩 Analysis chain
🏁 Script executed:
#!/bin/bash# Search for enum-related tests, especially with arrays rg -n -C5 --type=ts 'enum.*\[\]|Role\[\]|\.create.*roles:' tests/e2e/ # Search for enum usage in dialect code ast-grep --pattern $'isEnum($$$) { $$$}'Repository: zenstackhq/zenstack
Length of output: 5342
🏁 Script executed:
#!/bin/bashset -euo pipefail # 1) Inspect getSqlType around the reported lines FILE="packages/orm/src/client/crud/dialects/postgresql.ts"echo"== getSqlType context ==" python3 - <<'PY'import itertools,syspath="packages/orm/src/client/crud/dialects/postgresql.ts"start=470; end=530with open(path,'r',encoding='utf-8') as f: for i,line in enumerate(f, start=1): if start <= i <= end: print(f"{i}:{line.rstrip()}") if i > end: breakPY# 2) Find actual PG enum schema definitions/usages in e2e testsecho"== e2e: schema enum definitions ==" rg -n --hidden --no-ignore -S --type=ts "enum\\s+[A-Za-z0-9_]+" tests/e2e/ ||trueecho"== e2e: enum array field patterns ==" rg -n --hidden --no-ignore -S --type=ts "[A-Za-z0-9_]+\\s*\\[\\]" tests/e2e/ -g'*.test.ts'| rg -n -S "enum|Enum"||trueecho"== e2e: postgres enum keywords ==" rg -n --hidden --no-ignore -S --type=ts "postgres.*enum|pg.*enum|::.*\\w+\\[\\]" tests/e2e/ ||true# 3) Search for enum usage tied to postgresql dialectecho"== code: isEnum usages (broader) ==" rg -n "isEnum\\(" packages/orm/src/client/crud/dialects/postgresql.ts packages/orm/src/client/crud/dialects -S ||trueRepository: zenstackhq/zenstack
Length of output: 6401
🏁 Script executed:
#!/bin/bashset -euo pipefail # Inspect the enum-array test that we know exists FILE="tests/e2e/orm/policy/isempty-function-pg.test.ts"echo"== ${FILE} ==" nl -ba "$FILE"| sed -n '1,220p'# Find all tests mentioning Role[] (enum array) in e2eecho"== rg 'Role\\[\\]' in e2e ==" rg -n --type=ts "Role\\[\\]" tests/e2e/ ||true# Find tests that define enums and then use them as arrays (simple heuristic)echo"== rg: enum ... then later ... [] usage (limited window) ==" rg -n --type=ts "enum\\s+[A-Za-z0-9_]+" tests/e2e/orm -S | head -n 50 # Inspect typing/typecheck.ts around enum array mention FILE2="tests/e2e/orm/schemas/typing/typecheck.ts"echo"== ${FILE2} (around enum array) ==" nl -ba "$FILE2"| sed -n '60,120p'# Search for isEmpty implementation usage (best-effort; may be TS in repo)echo"== code search: isEmpty( ... ) ==" rg -n --type=ts "isEmpty\\(" packages/orm/src tests/e2e/orm | head -n 50Repository: zenstackhq/zenstack
Length of output: 179
🏁 Script executed:
#!/bin/bashset -euo pipefail FILE="tests/e2e/orm/policy/isempty-function-pg.test.ts"echo"== ${FILE} (first 200 lines with numbers) ==" cat -n "$FILE"| sed -n '1,220p'echo"== rg 'Role\\[\\]' in tests/e2e (first matches) ==" rg -n --type=ts "Role\\[\\]" tests/e2e/ | head -n 50 ||trueecho"== rg for enum definitions in e2e (first 100 lines of matches) ==" rg -n --type=ts "enum\\s+[A-Za-z0-9_]+" tests/e2e/orm | head -n 100 echo"== typing/typecheck.ts around 'enum array' ==" FILE2="tests/e2e/orm/schemas/typing/typecheck.ts" cat -n "$FILE2"| sed -n '60,130p'echo"== code search: isEmpty implementation (best-effort) ==" rg -n --type=ts "isEmpty\\(" packages/orm/src | head -n 80 ||true rg -n --type=ts "function\\s+isEmpty|isEmpty\\s*:" packages/orm/src | head -n 80 ||true rg -n --type=ts "build.*Empty|empty.*predicate|Array.*Empty|isEmpty.*predicate" packages/orm/src | head -n 80 ||trueRepository: zenstackhq/zenstack
Length of output: 7293
🏁 Script executed:
#!/bin/bashset -euo pipefail # 1) Read isEmpty implementation FILE="packages/orm/src/client/functions.ts"echo"== ${FILE}: around isEmpty ==" rg -n "export const isEmpty""$FILE" LINE=$(rg -n "export const isEmpty""$FILE"| head -n1 | cut -d: -f1) start=$((LINE-60)); end=$((LINE+120)) python3 - <<PYimport itertoolspath="$FILE"start=$start; end=$endwith open(path,'r',encoding='utf-8') as f: for i,line in enumerate(f, start=1): if start<=i<=end: print(f"{i}:{line.rstrip()}")PY# 2) Follow any dialect helper referenced by isEmptyecho"== code search: buildCollectionEmptyPredicate / empty predicate ==" rg -n "build.*Empty|Empty.*Predicate|isEmpty" packages/orm/src/client/crud/dialects packages/orm/src/client | head -n 80 # 3) Look for dialect method definitions that match what isEmpty calls# (we’ll search for the most likely helper names found above)Repository: zenstackhq/zenstack
Length of output: 6628
🏁 Script executed:
#!/bin/bashset -euo pipefail # 1) Base dialect isEmpty handling FILE="packages/orm/src/client/crud/dialects/base-dialect.ts"echo"== ${FILE} around isEmpty switch ==" rg -n "case 'isEmpty'""$FILE" LINE=$(rg -n "case 'isEmpty'""$FILE"| head -n1 | cut -d: -f1) start=$((LINE-60)); end=$((LINE+120)) python3 - <<PYpath="$FILE"start=$start; end=$endwith open(path,'r',encoding='utf-8') as f: for i,line in enumerate(f, start=1): if start<=i<=end: print(f"{i}:{line.rstrip()}") if i>end: breakPY# 2) PostgreSQL dialect array helpers FILE2="packages/orm/src/client/crud/dialects/postgresql.ts"echo"== code search in postgresql.ts: buildArrayLength/buildArrayContains/buildArrayHas* ==" rg -n "buildArrayLength|buildArrayContains|buildArrayHasEvery|buildArrayHasSome|getSqlType\\(""$FILE2"# Print likely relevant sections around those methodsforpatin"buildArrayLength""buildArrayContains""buildArrayHasEvery""buildArrayHasSome""getSqlType";do LINE=$(rg -n "$pat""$FILE2"| head -n1 | cut -d: -f1 || true)if [ -n"${LINE:-}" ];then start=$((LINE-40)); end=$((LINE+120))echo"== ${FILE2} around ${pat} (lines ${start}-${end}) ==" python3 - <<PYpath="$FILE2"start=$start; end=$endwith open(path,'r',encoding='utf-8') as f: for i,line in enumerate(f, start=1): if start<=i<=end: print(f"{i}:{line.rstrip()}") if i>end: breakPYfidone# 3) E2E: find has/hasEvery/hasSome involving enum arrays with Postgresecho"== e2e: search for hasEvery/hasSome/has with enum arrays ==" rg -n --type=ts "(hasEvery|hasSome|\\bhas\\b)" tests/e2e/orm/policy tests/e2e/orm/client-api | head -n 200 # specifically within Role[]/enum contextecho"== e2e: search for Role[] + has/hasEvery/hasSome ==" rg -n --type=ts "Role\\[\\].{0,200}(hasEvery|hasSome|\\bhas\\b)" tests/e2e/orm -S ||trueecho"== e2e: search for enum arrays 'Status.ACTIVE' has/hasEvery/hasSome ==" rg -n --type=ts "Status\\.ACTIVE.*(hasEvery|hasSome|\\bhas\\b)" tests/e2e/orm | head -n 200 ||trueRepository: zenstackhq/zenstack
Length of output: 50375
🏁 Script executed:
#!/bin/bashset -euo pipefail # Find where Role[]/roles is used with list predicate operators rg -n --type=ts "roles\\s*Role\\[\\]|roles:|where:.*roles|\\bhasEvery\\b|\\bhasSome\\b|\\bhas\\b" tests/e2e/orm/policy/ -S # Inspect collection-predicate test around the Role/roles definition and nearby assertions FILE="tests/e2e/orm/policy/collection-predicate.test.ts"echo"== ${FILE} (snippet around Role definition) ==" sed -n '110,220p'"$FILE"echo"== ${FILE} (snippet around later Role definition/uses) ==" sed -n '220,380p'"$FILE"echo"== ${FILE} (snippet around more Role definitions/uses) ==" sed -n '380,560p'"$FILE"# Confirm provider usage in the policy tests (best-effort: look for createPolicyTestClient options)echo"== ${FILE}: createPolicyTestClient provider config ==" rg -n "createPolicyTestClient\\(|provider:\\s*'postgresql'|postgresql""$FILE"||true# Also search for other e2e files using "roles Role[]" with has/hasSome/hasEvery rg -n --type=ts "roles\\s*Role\\[\\]" tests/e2e/orm -S rg -n --type=ts "roles.*\\{.*(hasEvery|hasSome|has)\\b" tests/e2e/orm -S ||trueRepository: zenstackhq/zenstack
Length of output: 15886
🏁 Script executed:
#!/bin/bashset -euo pipefail # 1) Find places where enum arrays are used with has/hasSome/hasEvery rg -n --type=ts "\\bhasEvery\\b|\\bhasSome\\b|\\bhas\\s*:" tests/e2e/orm -S > /tmp/has_ops.txt ||true# Reduce to lines that mention enum-like identifiers (heuristic)# Status.ACTIVE/Role/AUTHOR/EDITOR are common in these schemas rg -n --type=ts "(Status\\.|Role\\b|AUTHOR|EDITOR|ACTIVE).*\\b(hasEvery|hasSome|has)\\b|\\b(hasEvery|hasSome|has)\\b.*(Status\\.|Role\\b|AUTHOR|EDITOR|ACTIVE)" /tmp/has_ops.txt ||true# 2) Specifically look for enum value identifiers with has/hasSome/hasEvery rg -n --type=ts "(Status\\.|Role\\.)\\s*.*\\b(hasEvery|hasSome|has)\\b|\\b(hasEvery|hasSome|has)\\b.*(Status\\.|Role\\.)" tests/e2e/orm -S ||true# 3) Look for array filters on a known enum-array field pattern: "<fieldName>: { has: ... }" rg -n --type=ts "[a-zA-Z0-9_]+\\s*:\\s*\\{[^}]*\\b(hasEvery|hasSome|has)\\b" tests/e2e/orm -S | head -n 200 # 4) Identify any schemas in e2e that define enums and also declare an enum array field.# e.g. "enum Status { ... }" and later "status Status[]" rg -n --type=ts "enum\\s+[A-Za-z0-9_]+\\s*\\{|[A-Za-z0-9_]+\\s+[A-Za-z0-9_]+\\[\\]" tests/e2e/orm -S > /tmp/enum_defs_and_arrays.txt ||true head -n 200 /tmp/enum_defs_and_arrays.txt # 5) If we found candidate files, inspect them quickly (only if present)# (We’ll pick the isEmpty-function-pg file plus any file that contains "enum" and "hasSome/hasEvery/has:".) candidates="$(rg -l --type=ts "enum\\s+[A-Za-z0-9_]+\\s*\\{" tests/e2e/orm | tr '\n''')"forfin$(rg -l --type=ts "\\bhasEvery\\b|\\bhasSome\\b|\\bhas\\s*:" tests/e2e/orm -S);doif rg -q --type=ts "enum\\s+[A-Za-z0-9_]+\\s*\\{""$f";thenecho"== candidate with enum + has ops: $f ==" sed -n '1,220p'"$f"fidoneRepository: zenstackhq/zenstack
Length of output: 39300
🏁 Script executed:
#!/bin/bashset -euo pipefail # Look for enum definitions that are likely used with array fields rg -n --type=ts "enum\\s+[A-Za-z0-9_]+" tests/e2e/orm -S > /tmp/e2e_enum_defs.txt ||true# Extract candidate enum namesecho"== First 80 enum definitions ==" head -n 80 /tmp/e2e_enum_defs.txt # Find enum array field declarations (e.g., status Status[]) rg -n --type=ts "\\b[A-Za-z0-9_]+\\s+[A-Za-z0-9_]+\\[\\]" tests/e2e/orm -S > /tmp/e2e_array_fields.txt ||trueecho"== First 80 array-typed field declarations ==" head -n 80 /tmp/e2e_array_fields.txt # Specifically look for enum arrays that use has/hasSome/hasEvery rg -n --type=ts "\\b(hasEvery|hasSome|\\bhas\\b)\\b" tests/e2e/orm -S | head -n 200 rg -n --type=ts "status\\s*\\{\\s*[^}]*\\b(hasEvery|hasSome|\\bhas\\b)" tests/e2e/orm -S ||true rg -n --type=ts "status\\s*:\\s*\\{[^}]*\\b(hasEvery|hasSome|\\bhas\\b)" tests/e2e/orm -S ||true rg -n --type=ts "\\benum\\s+Status\\b|\\bStatus\\s*\\[\\]" tests/e2e/orm -S ||true# Look for runtime tests (non typecheck) involving Status.ACTIVE in where filters rg -n --type=ts "Status\\.ACTIVE" tests/e2e/orm -S ||trueRepository: zenstackhq/zenstack
Length of output: 24846
Confirm PG enum→
textmapping is intentional, and add runtime coverage for enum-array membership operatorsif (isEnum(this.schema, zmodelType)) { // reduce enum to text for type compatibility return 'text';
packages/orm/src/client/crud/dialects/postgresql.tsintentionally maps enum types to'text'(“reduce enum to text for type compatibility”).packages/orm/src/client/crud/dialects/base-dialect.tsfurther casts enum arrays totext[]inbuildArrayFilterfor array comparisons.tests/e2e/orm/policy/isempty-function-pg.test.tscoversenum Role+roles Role[]withprovider: 'postgresql', but it only exercisesisEmpty(roles)(array-length), not enum-array membership (has/hasSome/hasEvery).- Existing
enum ... has/hasEveryreferences found intests/e2e/orm/schemas/typing/typecheck.tsare typechecking-only, not runtime assertions.Add a Postgres e2e test that exercises
rolesenum-array filters withhas/hasSome/hasEveryto ensure the enum→text[]casting preserves semantics.🤖 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 `@packages/orm/src/client/crud/dialects/postgresql.ts` around lines 502 - 504, Add an e2e test for Postgres that verifies enum-array membership operators work when enums are mapped to text (so the cast to text[] in buildArrayFilter preserves semantics): create a schema with enum Role and field roles Role[], insert records with various role combinations, then run queries using the array filters has, hasSome, and hasEvery against roles and assert expected matches for provider: 'postgresql' (mirror the style of tests/e2e/orm/policy/isempty-function-pg.test.ts); ensure the test targets behavior in postgresql.ts (isEnum) and base-dialect.ts (buildArrayFilter) by asserting correct runtime results for each operator.
ymc9
approved these changes
Jun 12, 2026
Uh oh!
There was an error while loading. Please reload this page.
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 freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Summary by CodeRabbit
Bug Fixes
Tests