[SEA-NodeJS] (3/3) INTERVAL type parity + operation-lifecycle depth by msrathore-db · Pull Request #411 · databricks/databricks-sql-nodejs

msrathore-db · 2026-06-01T08:31:09Z

Third of three stacked PRs (base: 2/3 execution + results). Completes the SEA M0 foundation.

ArrowResultConverter — INTERVAL parity: formats Arrow Interval[YearMonth]/Interval[DayTime] and Duration (→Int64 via SeaArrowIpcDurationFix) into canonical Thrift strings (Y-M / D HH:mm:ss.fffffffff), byte-identical to the Thrift path.
Exhaustive operation-lifecycle coverage: seaCancel/seaClose/seaFinished idempotency, flag-set-before-await (cancel-mid-fetch), kernel-error mapping, neutral OperationStatus callback shape.
SeaIntervalParity tests build real Arrow IPC batches via flatbuffers and assert the formatted strings.

With this, SEA reaches M0 parity with Thrift. Replaces the single 8/8 #383.

Stack: 1/3 → 2/3 → 3/3

This pull request and its description were written by Isaac.

…rity, docs Validated each finding against a live pecotesting warehouse first; the headline INTERVAL story turned out to be split-artifact, not breakage. - F7: getResultMetadata stored the *unpatched* Duration IPC bytes in meta.arrowSchema while advertising ArrowBased — store the patched bytes so an ArrowBased consumer doesn't hit `Unrecognized type "Duration" (18)`. - F3: fetchChunk now honors the `isClosed` cooperative-cancel probe (parity with ThriftOperationBackend) at its yield points. - F6: on a fetch error, best-effort close the statement (napi contract: stream is unspecified after Err) and surface a typed kernel error via decodeNapiKernelError. - F9: cancel-after-fetch now throws the canonical OperationStateError(Canceled) ("The operation was canceled by a client") — byte-matches the Thrift message. - F10: typed HiveDriverError (not raw Error) in the schema/fetchNextBatch guards. - F1: corrected SeaArrowIpcDurationFix docs — on this layer the rewriter only makes Duration *decodable* (raw Int64); the duration_unit formatter lands in #411 (verified live: byte-identical to Thrift). - F5: documented that nested Duration is a SHARED apache-arrow@13 limitation — verified the Thrift backend throws the identical error, so SEA matches parity. - F2: added a live e2e that drives a real Arrow Duration column through the rewriter (asserts no "Duration (18)" crash + raw-Int64 on this layer). - F8: pinned the no-`Failed` invariant in status() (failures reject at submit). - F12: renamed SeaResultsProvider's SeaStatementHandle → SeaFetchHandle (was a name collision with the lifecycle interface of a different shape). - F13: dropped the no-op await on the synchronous statement.schema(). - F14: fixed the Float-precision comment (Precision enum, not bit-width). - F15: SeaResultsProvider.prime loops instead of self-recursing on empty batches. Deferred (noted on the PR): F4 (per-batch triple-decode perf) and F11 (hasResultSet() hard-coded true for M0). Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

…erage Validated every interval edge case (null, multi-row, negative sub-year, sibling-survives) against a live pecotesting warehouse first — all byte-identical to Thrift. The findings were layering/dead-code/coverage, not runtime bugs. - F1: corrected the DURATION_UNIT_METADATA_KEY doc — the interval/duration branches are SEA-gated by construction (Thrift maps INTERVAL → ArrowString and never reaches them), NOT "reused by thrift" as the old comment claimed. - F3/F6/F10: formatArrowInterval now handles YEAR_MONTH only (typed `Interval` + `IntervalUnit.YEAR_MONTH`, no magic `=== 0`) and THROWS on any other unit. The old non-exhaustive default silently misread MONTH_DAY_NANO/undefined as [days,ms]; and the native Interval[DayTime] branch was dead+broken (the kernel emits DAY-TIME as Duration, handled separately) — removed it. - F2: exported the metadata key + added a test pinning it equal to the SEA-side declaration (guards against a silent rename drift). - F8: dropped the dead `_unit` param from formatDayTimeFromTotal. - F9: removed the dead duration check in the BigNum branch (rewritten Int64s arrive as raw bigint) and fixed the false "Int32Array instanceof Uint8Array" comment. - F7: added a YEAR-MONTH sub-year-negative unit test (-1 month → "-0-1"). - F4: added live e2e for null INTERVAL → null and multi-row batches. - F5: e2e before() now probes getSeaNative() and skips (not errors) when the binding is absent; dropped the flaky wall-clock latency assertions (assert behavior — cancel resolved, callback fired once). Deferred (low/informational, noted): F11 (consolidate test makeContext helpers), F12 (tighten instanceOf assertions), F13 (per-operation interval-representation breadcrumb — a per-value log would be spam). Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

gopalldb · 2026-06-02T10:59:08Z

Change naming to Kernel.. instead of Sea.. everywhere

gopalldb · 2026-06-02T11:16:06Z

🟡 P2 (Minor )

toNanoseconds default treats unknown units as nanoseconds — SEA-only and currently impossible (rewriter enumerates exactly the four units), but the asymmetry (rewriter defaults MICROSECOND, converter
defaults NANOSECOND) is a latent footgun if they ever drift. → consider throwing on unrecognized units, mirroring formatArrowInterval's fail-loud stance.
formatArrowInterval throws mid-result on native non-YEAR_MONTH intervals — reachable only on SEA if the kernel's "DAY-TIME always arrives as Duration" assumption breaks. Fail-loud is defensible (the old
non-exhaustive default emitted confidently-wrong [days, ms]), and it can't affect Thrift.
Test gaps — no dedicated test asserts the converter throws on a native non-YEAR_MONTH Interval, and no SEA test exercises MILLISECOND/SECOND duration units specifically

…erage Validated every interval edge case (null, multi-row, negative sub-year, sibling-survives) against a live pecotesting warehouse first — all byte-identical to Thrift. The findings were layering/dead-code/coverage, not runtime bugs. - F1: corrected the DURATION_UNIT_METADATA_KEY doc — the interval/duration branches are SEA-gated by construction (Thrift maps INTERVAL → ArrowString and never reaches them), NOT "reused by thrift" as the old comment claimed. - F3/F6/F10: formatArrowInterval now handles YEAR_MONTH only (typed `Interval` + `IntervalUnit.YEAR_MONTH`, no magic `=== 0`) and THROWS on any other unit. The old non-exhaustive default silently misread MONTH_DAY_NANO/undefined as [days,ms]; and the native Interval[DayTime] branch was dead+broken (the kernel emits DAY-TIME as Duration, handled separately) — removed it. - F2: exported the metadata key + added a test pinning it equal to the SEA-side declaration (guards against a silent rename drift). - F8: dropped the dead `_unit` param from formatDayTimeFromTotal. - F9: removed the dead duration check in the BigNum branch (rewritten Int64s arrive as raw bigint) and fixed the false "Int32Array instanceof Uint8Array" comment. - F7: added a YEAR-MONTH sub-year-negative unit test (-1 month → "-0-1"). - F4: added live e2e for null INTERVAL → null and multi-row batches. - F5: e2e before() now probes getSeaNative() and skips (not errors) when the binding is absent; dropped the flaky wall-clock latency assertions (assert behavior — cancel resolved, callback fired once). Deferred (low/informational, noted): F11 (consolidate test makeContext helpers), F12 (tighten instanceOf assertions), F13 (per-operation interval-representation breadcrumb — a per-value log would be spam). Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

…t gaps Addresses the P2 review comment on #411 (the interval/duration layer) and cascades the #410 fixes underneath (this branch was rebased onto the updated #410 tip). Validated against a live warehouse (interval-duration + interval-edge e2e) and unit tests. - ArrowResultConverter.toNanoseconds: throw on an unrecognized Arrow duration unit instead of silently defaulting to NANOSECOND. The four units are exactly what SeaArrowIpcDurationFix stamps, so an unknown unit means the two sides drifted — fail loud, matching formatArrowInterval. - SeaIntervalParity tests: add DAY-TIME via Duration(MILLISECOND) and Duration(SECOND) (only MICROSECOND/NANOSECOND were exercised), and a test asserting the converter throws (HiveDriverError) on a native non-YEAR_MONTH Arrow Interval rather than misreading it. - interval-duration-e2e: flip the stale "raw Int64 on this layer" assertion to expect the formatted thrift string "1 02:03:04.000000000". That test was written on #410 (pre-formatter) and explicitly noted "#411 flips this to the formatted string" — now that #411's formatter is wired, the value is the byte-identical thrift DAY-TIME string. Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

msrathore-db · 2026-06-02T14:14:20Z

Addressed the P2 items in 54dbcff (validated against a live warehouse + unit tests):

toNanoseconds fail-loud: now throws HiveDriverError on an unrecognized Arrow duration unit instead of defaulting to NANOSECOND. The four units are exactly what SeaArrowIpcDurationFix stamps, so an unknown unit means the two sides drifted — fail loud, matching formatArrowInterval's stance.
formatArrowInterval mid-result throw: kept as-is (fail-loud is the intended contract, as you noted — it can't affect Thrift).
Test gaps: added DAY-TIME via Duration(MILLISECOND) and Duration(SECOND) cases (only MICROSECOND/NANOSECOND were exercised), plus a test asserting the converter throws on a native non-YEAR_MONTH Arrow Interval rather than misreading it as [days, ms].

The "rename Sea.. → Kernel.. everywhere" comment is broader (it touches code already on main and the public useSEA option) — following up separately on scope.

msrathore-db · 2026-06-02T14:19:58Z

Re: "Change naming to Kernel.. instead of Sea.. everywhere" — deferring this to a separate dedicated PR rather than folding it into this stack. The SEA backend (the lib/sea/ files + the public useSEA connection option) is already on main, and a ~349-occurrence rename across both #410/#411 and the merged code would diverge from main, churn the public API, and make these feature PRs hard to review. It'll be cleaner as one self-contained rename PR once the stack lands.

…erage Validated every interval edge case (null, multi-row, negative sub-year, sibling-survives) against a live pecotesting warehouse first — all byte-identical to Thrift. The findings were layering/dead-code/coverage, not runtime bugs. - F1: corrected the DURATION_UNIT_METADATA_KEY doc — the interval/duration branches are SEA-gated by construction (Thrift maps INTERVAL → ArrowString and never reaches them), NOT "reused by thrift" as the old comment claimed. - F3/F6/F10: formatArrowInterval now handles YEAR_MONTH only (typed `Interval` + `IntervalUnit.YEAR_MONTH`, no magic `=== 0`) and THROWS on any other unit. The old non-exhaustive default silently misread MONTH_DAY_NANO/undefined as [days,ms]; and the native Interval[DayTime] branch was dead+broken (the kernel emits DAY-TIME as Duration, handled separately) — removed it. - F2: exported the metadata key + added a test pinning it equal to the SEA-side declaration (guards against a silent rename drift). - F8: dropped the dead `_unit` param from formatDayTimeFromTotal. - F9: removed the dead duration check in the BigNum branch (rewritten Int64s arrive as raw bigint) and fixed the false "Int32Array instanceof Uint8Array" comment. - F7: added a YEAR-MONTH sub-year-negative unit test (-1 month → "-0-1"). - F4: added live e2e for null INTERVAL → null and multi-row batches. - F5: e2e before() now probes getSeaNative() and skips (not errors) when the binding is absent; dropped the flaky wall-clock latency assertions (assert behavior — cancel resolved, callback fired once). Deferred (low/informational, noted): F11 (consolidate test makeContext helpers), F12 (tighten instanceOf assertions), F13 (per-operation interval-representation breadcrumb — a per-value log would be spam). Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

…t gaps Addresses the P2 review comment on #411 (the interval/duration layer) and cascades the #410 fixes underneath (this branch was rebased onto the updated #410 tip). Validated against a live warehouse (interval-duration + interval-edge e2e) and unit tests. - ArrowResultConverter.toNanoseconds: throw on an unrecognized Arrow duration unit instead of silently defaulting to NANOSECOND. The four units are exactly what SeaArrowIpcDurationFix stamps, so an unknown unit means the two sides drifted — fail loud, matching formatArrowInterval. - SeaIntervalParity tests: add DAY-TIME via Duration(MILLISECOND) and Duration(SECOND) (only MICROSECOND/NANOSECOND were exercised), and a test asserting the converter throws (HiveDriverError) on a native non-YEAR_MONTH Arrow Interval rather than misreading it. - interval-duration-e2e: flip the stale "raw Int64 on this layer" assertion to expect the formatted thrift string "1 02:03:04.000000000". That test was written on #410 (pre-formatter) and explicitly noted "#411 flips this to the formatted string" — now that #411's formatter is wired, the value is the byte-identical thrift DAY-TIME string. Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

…rity, docs Validated each finding against a live pecotesting warehouse first; the headline INTERVAL story turned out to be split-artifact, not breakage. - F7: getResultMetadata stored the *unpatched* Duration IPC bytes in meta.arrowSchema while advertising ArrowBased — store the patched bytes so an ArrowBased consumer doesn't hit `Unrecognized type "Duration" (18)`. - F3: fetchChunk now honors the `isClosed` cooperative-cancel probe (parity with ThriftOperationBackend) at its yield points. - F6: on a fetch error, best-effort close the statement (napi contract: stream is unspecified after Err) and surface a typed kernel error via decodeNapiKernelError. - F9: cancel-after-fetch now throws the canonical OperationStateError(Canceled) ("The operation was canceled by a client") — byte-matches the Thrift message. - F10: typed HiveDriverError (not raw Error) in the schema/fetchNextBatch guards. - F1: corrected SeaArrowIpcDurationFix docs — on this layer the rewriter only makes Duration *decodable* (raw Int64); the duration_unit formatter lands in #411 (verified live: byte-identical to Thrift). - F5: documented that nested Duration is a SHARED apache-arrow@13 limitation — verified the Thrift backend throws the identical error, so SEA matches parity. - F2: added a live e2e that drives a real Arrow Duration column through the rewriter (asserts no "Duration (18)" crash + raw-Int64 on this layer). - F8: pinned the no-`Failed` invariant in status() (failures reject at submit). - F12: renamed SeaResultsProvider's SeaStatementHandle → SeaFetchHandle (was a name collision with the lifecycle interface of a different shape). - F13: dropped the no-op await on the synchronous statement.schema(). - F14: fixed the Float-precision comment (Precision enum, not bit-width). - F15: SeaResultsProvider.prime loops instead of self-recursing on empty batches. Deferred (noted on the PR): F4 (per-batch triple-decode perf) and F11 (hasResultSet() hard-coded true for M0). Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

* feat(sea): SEA execution + result fetching [2/3] Second of three stacked PRs (base: [1/3] connect + auth). Wires the statement-execution + result-read path: - SeaSessionBackend.executeStatement: real implementation — runs SQL via the napi Connection and returns a SeaOperationBackend (replaces [1/3]'s stub). - SeaOperationBackend: fetch pipeline (napi Statement.fetchNextBatch → SeaResultsProvider → ArrowResultConverter → ResultSlicer) plus operation cancel/close/finished via the SeaOperationLifecycle helpers. - SeaResultsProvider / SeaArrowIpc / SeaArrowIpcDurationFix: Arrow IPC decode for inline + CloudFetch result batches (the duration-fix pre-processor rewrites Arrow Duration → Int64 so apache-arrow@13 can read it). - ArrowResultConverter: constructor now takes the neutral { schema? } shape so both the Thrift and SEA backends construct it without an adapter. - flatbuffers pinned to 23.5.26 to match apache-arrow@13's nested copy. Tests: executeStatement + openSession forwarding, M0 datatype round-trip through the shared converter (primitives + ARRAY/MAP/STRUCT), multi-batch streaming, and the neutral-metadata converter contract. Full INTERVAL-type value parity + exhaustive operation-lifecycle coverage land in [3/3]. Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com> * fix(sea): address #410 review — fetch cleanup, cooperative cancel, parity, docs Validated each finding against a live pecotesting warehouse first; the headline INTERVAL story turned out to be split-artifact, not breakage. - F7: getResultMetadata stored the *unpatched* Duration IPC bytes in meta.arrowSchema while advertising ArrowBased — store the patched bytes so an ArrowBased consumer doesn't hit `Unrecognized type "Duration" (18)`. - F3: fetchChunk now honors the `isClosed` cooperative-cancel probe (parity with ThriftOperationBackend) at its yield points. - F6: on a fetch error, best-effort close the statement (napi contract: stream is unspecified after Err) and surface a typed kernel error via decodeNapiKernelError. - F9: cancel-after-fetch now throws the canonical OperationStateError(Canceled) ("The operation was canceled by a client") — byte-matches the Thrift message. - F10: typed HiveDriverError (not raw Error) in the schema/fetchNextBatch guards. - F1: corrected SeaArrowIpcDurationFix docs — on this layer the rewriter only makes Duration *decodable* (raw Int64); the duration_unit formatter lands in #411 (verified live: byte-identical to Thrift). - F5: documented that nested Duration is a SHARED apache-arrow@13 limitation — verified the Thrift backend throws the identical error, so SEA matches parity. - F2: added a live e2e that drives a real Arrow Duration column through the rewriter (asserts no "Duration (18)" crash + raw-Int64 on this layer). - F8: pinned the no-`Failed` invariant in status() (failures reject at submit). - F12: renamed SeaResultsProvider's SeaStatementHandle → SeaFetchHandle (was a name collision with the lifecycle interface of a different shape). - F13: dropped the no-op await on the synchronous statement.schema(). - F14: fixed the Float-precision comment (Precision enum, not bit-width). - F15: SeaResultsProvider.prime loops instead of self-recursing on empty batches. Deferred (noted on the PR): F4 (per-batch triple-decode perf) and F11 (hasResultSet() hard-coded true for M0). Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com> * fix(sea): address #410 review — error fidelity, fetch perf, validation, tests Addresses the 8 review threads on PR #410. Validated against a live warehouse (results-e2e parity gate + interval-duration-e2e + execution-e2e all pass) plus new/updated unit tests. - SeaOperationLifecycle: rethrowKernelError now delegates to the canonical decodeNapiKernelError, so cancel/close errors get the same fidelity as fetch errors — the sqlState remap (envelope field is `sqlState`, the old code read `sqlstate` and dropped it), the kernelMetadata namespace, and the strict `startsWith` sentinel match (was a loose `indexOf >= 0`). - SeaArrowIpc: replace decodeIpcBatch (full RecordBatchReader materialization just to sum row counts) with countRowsInIpc, which reads RecordBatch header `length` via MessageReader and skips bodies — no vector decode. Removes ~2x Arrow decode CPU + transient allocation on the fetch hot path (the converter still re-decodes for values). SeaResultsProvider switched to it. - SeaArrowIpc: hermetic unit tests (tests/unit/sea/SeaArrowIpc.test.ts) for the framing walk, no-op/garbage rewrite paths, the row-count path, and the empty-schema guard. (The Duration-positive rewrite stays covered by the live e2e — apache-arrow@13 can't construct a Duration column hermetically.) - SeaOperationBackend: on a fetch-error cleanup close() that also fails, log the failure at warn (statement may leak) instead of fully swallowing it — the original fetch error is still surfaced. - SeaSessionBackend: reject queryTags / useLZ4Compression / stagingAllowedLocalPath with M0-style errors instead of silently ignoring them (+ unit tests). Silent no-ops are the worst failure mode for callers. - SeaArrowIpc: throw a typed HiveDriverError (not a raw TypeError) when an IPC payload carries no schema. - SeaArrowIpcDurationFix.readMessageAt: fail-closed guards for negative metadataLength / bodyLength (was relying on subarray clamping). - Fix stale `tests/integration/sea/...` doc refs → `tests/e2e/sea/`. Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com> * test(telemetry): de-flake FeatureFlagCache placeholder test (network seam) The "fetchFeatureFlag should return false as placeholder implementation" test called the real fetchFeatureFlag, which makes an HTTP call via fetchWithRetry (10s timeout) to a bogus host. Under mocha's 2s default it only passed when the DNS failure happened to resolve quickly — flaky across runners and Node versions (it timed out on Node 14/16/18 in CI, fail-fast-canceling the rest of the matrix; Node 20 passed). Stub the fetchWithRetry network seam so the test deterministically exercises the behavior under test (fetchFeatureFlag resolves to false) with no real network call. No production code change. Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com> --------- Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

Third of three stacked PRs (base: [2/3] execution + results). Completes the SEA foundation: - ArrowResultConverter: INTERVAL parity. Formats Arrow Interval[YearMonth] / Interval[DayTime] and Duration (rewritten to Int64 by SeaArrowIpcDurationFix) into the canonical Thrift strings ("Y-M" / "D HH:mm:ss.fffffffff"), byte- identical to the Thrift path. Threads the Arrow field through convertArrowTypes so the duration-unit metadata is available at value-conversion time. - Exhaustive operation-lifecycle coverage: seaCancel / seaClose / seaFinished idempotency, flag-set-before-await ordering (cancel-mid-fetch), kernel-error mapping, and the neutral OperationStatus callback shape. - SeaIntervalParity tests build real Arrow IPC batches via flatbuffers and assert the formatted strings. With this, SEA reaches M0 parity with Thrift (connect/auth → execute → fetch → operation lifecycle → INTERVAL types). Replaces the single 8/8 PR #383. Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

…erage Validated every interval edge case (null, multi-row, negative sub-year, sibling-survives) against a live pecotesting warehouse first — all byte-identical to Thrift. The findings were layering/dead-code/coverage, not runtime bugs. - F1: corrected the DURATION_UNIT_METADATA_KEY doc — the interval/duration branches are SEA-gated by construction (Thrift maps INTERVAL → ArrowString and never reaches them), NOT "reused by thrift" as the old comment claimed. - F3/F6/F10: formatArrowInterval now handles YEAR_MONTH only (typed `Interval` + `IntervalUnit.YEAR_MONTH`, no magic `=== 0`) and THROWS on any other unit. The old non-exhaustive default silently misread MONTH_DAY_NANO/undefined as [days,ms]; and the native Interval[DayTime] branch was dead+broken (the kernel emits DAY-TIME as Duration, handled separately) — removed it. - F2: exported the metadata key + added a test pinning it equal to the SEA-side declaration (guards against a silent rename drift). - F8: dropped the dead `_unit` param from formatDayTimeFromTotal. - F9: removed the dead duration check in the BigNum branch (rewritten Int64s arrive as raw bigint) and fixed the false "Int32Array instanceof Uint8Array" comment. - F7: added a YEAR-MONTH sub-year-negative unit test (-1 month → "-0-1"). - F4: added live e2e for null INTERVAL → null and multi-row batches. - F5: e2e before() now probes getSeaNative() and skips (not errors) when the binding is absent; dropped the flaky wall-clock latency assertions (assert behavior — cancel resolved, callback fired once). Deferred (low/informational, noted): F11 (consolidate test makeContext helpers), F12 (tighten instanceOf assertions), F13 (per-operation interval-representation breadcrumb — a per-value log would be spam). Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

…t gaps Addresses the P2 review comment on #411 (the interval/duration layer) and cascades the #410 fixes underneath (this branch was rebased onto the updated #410 tip). Validated against a live warehouse (interval-duration + interval-edge e2e) and unit tests. - ArrowResultConverter.toNanoseconds: throw on an unrecognized Arrow duration unit instead of silently defaulting to NANOSECOND. The four units are exactly what SeaArrowIpcDurationFix stamps, so an unknown unit means the two sides drifted — fail loud, matching formatArrowInterval. - SeaIntervalParity tests: add DAY-TIME via Duration(MILLISECOND) and Duration(SECOND) (only MICROSECOND/NANOSECOND were exercised), and a test asserting the converter throws (HiveDriverError) on a native non-YEAR_MONTH Arrow Interval rather than misreading it. - interval-duration-e2e: flip the stale "raw Int64 on this layer" assertion to expect the formatted thrift string "1 02:03:04.000000000". That test was written on #410 (pre-formatter) and explicitly noted "#411 flips this to the formatted string" — now that #411's formatter is wired, the value is the byte-identical thrift DAY-TIME string. Co-authored-by: Isaac Signed-off-by: Madhavendra Rathore <madhavendra.rathore@databricks.com>

msrathore-db force-pushed the msrathore/sea-execution-results branch from 1ab03b7 to 33252ab Compare June 1, 2026 13:43

msrathore-db force-pushed the msrathore/sea-operation-lifecycle branch from e79405d to d8bcc30 Compare June 1, 2026 13:43

msrathore-db force-pushed the msrathore/sea-execution-results branch from 33252ab to 59fa37a Compare June 1, 2026 17:57

msrathore-db force-pushed the msrathore/sea-operation-lifecycle branch from d8bcc30 to dde8efa Compare June 1, 2026 17:58

msrathore-db force-pushed the msrathore/sea-operation-lifecycle branch from dde8efa to fc2aee5 Compare June 1, 2026 20:42

msrathore-db requested a review from gopalldb June 2, 2026 09:10

msrathore-db force-pushed the msrathore/sea-operation-lifecycle branch from e2abd38 to 54dbcff Compare June 2, 2026 14:13

msrathore-db force-pushed the msrathore/sea-operation-lifecycle branch from 54dbcff to 456046c Compare June 2, 2026 14:40

msrathore-db force-pushed the msrathore/sea-execution-results branch from 8d60efd to 1991121 Compare June 2, 2026 14:54

msrathore-db added 2 commits June 2, 2026 15:00

msrathore-db changed the base branch from msrathore/sea-execution-results to main June 2, 2026 15:01

msrathore-db force-pushed the msrathore/sea-operation-lifecycle branch from 456046c to 92f2ee8 Compare June 2, 2026 15:01

msrathore-db temporarily deployed to azure-prod June 2, 2026 15:01 — with GitHub Actions Inactive

gopalldb approved these changes Jun 2, 2026

View reviewed changes

msrathore-db merged commit de04e19 into main Jun 2, 2026
8 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[SEA-NodeJS] (3/3) INTERVAL type parity + operation-lifecycle depth#411

[SEA-NodeJS] (3/3) INTERVAL type parity + operation-lifecycle depth#411
msrathore-db merged 3 commits into
mainfrom
msrathore/sea-operation-lifecycle

msrathore-db commented Jun 1, 2026

Uh oh!

gopalldb commented Jun 2, 2026

Uh oh!

gopalldb commented Jun 2, 2026

Uh oh!

msrathore-db commented Jun 2, 2026

Uh oh!

msrathore-db commented Jun 2, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

msrathore-db commented Jun 1, 2026

Uh oh!

gopalldb commented Jun 2, 2026

Uh oh!

gopalldb commented Jun 2, 2026

Uh oh!

msrathore-db commented Jun 2, 2026

Uh oh!

msrathore-db commented Jun 2, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants