feat(spark): SparkSource query+path and pre-computed offline read for BatchFeatureView by abhijeet-dhumal · Pull Request #6440 · feast-dev/feast

abhijeet-dhumal · 2026-05-27T11:59:35Z

What this PR does / why we need it

get_historical_features() on a BatchFeatureView re-runs the full UDF on raw data every call. For embedding pipelines, that's 20–40 min of compute per training run even though features already exist from the last materialize.

Fix: Route get_historical_features() to read pre-computed parquet from batch_source.path instead of re-executing the UDF.

To support this, SparkSource now accepts query + path together:

query — raw data read during materialize()
path — write-back target and pre-computed read source for get_historical_features()

SparkSource(
    query="SELECT id, text, event_timestamp FROM bronze.documents",
    path="s3://my-bucket/feast/features/document_embeddings/",
)

Also allows BatchFeatureView with online=False, offline=True (offline-only) to skip the online validation check in get_historical_features(), so it can be used purely for training data without configuring an online store.

Falls back to live query if path doesn't exist yet (first run before any materialization).

Which issue(s) this PR fixes

N/A. Enables efficient training data retrieval for BatchFeatureView embedding pipelines without re-running UDFs.

Checks

I've made sure the tests are passing.
My commits are signed off (git commit -s)
My PR title follows conventional commits format

Testing Strategy

Unit tests — offline path routing, SparkSource constraint, graceful fallback
Manual tests — get_historical_features() reads from parquet, not UDF, after materialization

devin-ai-integration

Devin Review found 1 potential issue.

View 5 additional findings in Devin Review.

…orical_features The function and its call were removed in this PR but the replacement (_apply_bfv_transformations_for_historical) lives in a separate PR (feast-dev#6440). Removing it here would silently return raw untransformed features for any BatchFeatureView with a Python UDF via the standard get_historical_features() API path (FeatureStore → passthrough_provider → SparkOfflineStore). Restoring the function and its call until feast-dev#6440 lands. Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

abhijeet-dhumal · 2026-06-01T07:48:49Z

@ntkathole May I request your review here too ?

jyejare

This PR adds support for SparkSource with combined query+path configuration and pre-computed offline reads for BatchFeatureView. The changes enable reading from materialized offline stores to avoid expensive UDF re-execution. While the feature is useful, there are several security vulnerabilities and error handling gaps that need attention.

jyejare · 2026-06-01T12:36:52Z

+            file_format = fv.batch_source.file_format or "parquet"
+            try:
+                df = spark_session.read.format(file_format).load(fv.batch_source.path)
+                df.createOrReplaceTempView(tmp_view)
+                ctx = replace(ctx, table_subquery=tmp_view)
+                new_contexts.append(ctx)


[Critical] Path traversal vulnerability in file loading

The code directly loads files from fv.batch_source.path without any validation. This creates a path traversal security vulnerability where malicious paths could access unauthorized files on the system.

Suggested:

Suggested change

file_format = fv.batch_source.file_format or "parquet"

try:

df = spark_session.read.format(file_format).load(fv.batch_source.path)

df.createOrReplaceTempView(tmp_view)

ctx = replace(ctx, table_subquery=tmp_view)

new_contexts.append(ctx)

+ # Validate and sanitize the path

+ import os

+ normalized_path = os.path.normpath(fv.batch_source.path)

+ if '..' in normalized_path or normalized_path.startswith('/'):

+ warnings.warn(f"Invalid path '{fv.batch_source.path}' for '{ctx.name}'", RuntimeWarning)

+ new_contexts.append(ctx)

+ continue

+ try:

+ df = spark_session.read.format(file_format).load(normalized_path)

+ df.createOrReplaceTempView(tmp_view)

+ ctx = replace(ctx, table_subquery=tmp_view)

+ new_contexts.append(ctx)

+ continue

Thanks for flagging this. fv.batch_source.path is a trusted configuration value set by the feature store admin at definition time — it's not runtime user input flowing from an API boundary, so path traversal in the traditional web security sense doesn't apply here.

Also, the suggested fix using os.path.normpath() would silently corrupt S3/GCS URIs (s3://bucket/path - s3:/bucket/path) since normpath collapses double slashes. Happy to add a lightweight guard for local paths only if you'd like, but I'd keep it separate from object storage paths.

Make Sense.

jyejare · 2026-06-01T12:36:52Z

+                continue
+            except Exception:
+                warnings.warn(
+                    f"Offline path '{fv.batch_source.path}' not readable for "
+                    f"'{ctx.name}'; falling back to source query.",
+                    RuntimeWarning,


[Warning] Overly broad exception handling masks specific errors

Catching all exceptions with 'except Exception:' is too broad and could mask important errors like permission issues, file corruption, or configuration problems. This makes debugging difficult and could hide security issues.
Suggested:

Suggested change

continue

except Exception:

warnings.warn(

f"Offline path '{fv.batch_source.path}' not readable for "

f"'{ctx.name}'; falling back to source query.",

RuntimeWarning,

+ except (FileNotFoundError, PermissionError) as e:

+ warnings.warn(

+ f"Offline path '{fv.batch_source.path}' not accessible for "

+ f"'{ctx.name}': {str(e)}; falling back to source query.",

+ RuntimeWarning,

+ stacklevel=2,

+ )

+ except Exception as e:

+ # Log unexpected errors but continue with fallback

+ import logging

+ logging.warning(f"Unexpected error loading '{fv.batch_source.path}': {e}")

+ warnings.warn(

+ f"Failed to load offline path for '{ctx.name}'; falling back to source query.",

+ RuntimeWarning,

+ stacklevel=2,

+ )

Good catch — updated to catch FileNotFoundError and PermissionError explicitly for the expected fallback cases, with a separate except Exception block emitting a distinct warning so unexpected errors aren't silently swallowed.

jyejare · 2026-06-01T12:36:52Z

+            if udf is not None:
+                temp_view_name = f"__feast_bfv_{ctx.name}_{uuid.uuid4().hex[:8]}"
+                spark_session.conf.set("spark.sql.runSQLOnFiles", "true")
+                raw_df = spark_session.sql(f"SELECT * FROM {ctx.table_subquery}")


[Warning] UDF execution without sandboxing poses security risks

Executing user-defined functions without proper sandboxing or validation could allow arbitrary code execution. This is a significant security risk in production environments.

Suggested:

Suggested change

if udf is not None:

temp_view_name = f"__feast_bfv_{ctx.name}_{uuid.uuid4().hex[:8]}"

spark_session.conf.set("spark.sql.runSQLOnFiles", "true")

raw_df = spark_session.sql(f"SELECT * FROM {ctx.table_subquery}")

+ # Add UDF validation and execution controls

+ if not hasattr(udf, '__call__'):

+ raise ValueError(f"Invalid UDF for {ctx.name}")

+ raw_df = spark_session.sql(f"SELECT * FROM {ctx.table_subquery}")

+ # Consider adding timeout and resource limits here

+ transformed_df = udf(raw_df)

+ transformed_df.createOrReplaceTempView(temp_view_name)

The UDF here is feature_transformation.udf registered in the Feast registry by the ML engineer at feature view definition time — same function that runs during materialize(). It's not arbitrary user input. The hasattr(udf, '__call__') check doesn't add a meaningful boundary since any Python object can implement __call__. Executor sandboxing is a cluster-level concern (Spark resource limits, executor isolation) — outside Feast's scope.

ntkathole · 2026-06-03T08:14:28Z


-        query_context = _apply_bfv_transformations(
-            spark_session, feature_views, query_context
+        query_context = _apply_bfv_transformations_for_historical(


so previous _apply_bfv_transformations helper not removed from code?

Good catch 🙌
the previous version had a separate _apply_bfv_transformations_for_historical that duplicated UDF execution logic. Removed it and folded the pre-computed path shortcut directly into _apply_bfv_transformations as the first branch. The call site now uses the single function.
Please review !

ntkathole · 2026-06-03T08:15:50Z

+                )
+
+        if (
+            hasattr(fv, "feature_transformation")


Use has_transformation() and get_transformation_function() instead

Done ✅
The UDF detection now uses has_transformation() and get_transformation_function() from feast.feature_view_utils, same as the existing _apply_bfv_transformations already did. The raw hasattr/getattr chains for feature_transformation are gone.

ntkathole · 2026-06-03T08:21:00Z

    max_date_partition: str


+def _apply_bfv_transformations_for_historical(


I think instead of writing new helper, better approach is to extend the existing _apply_bfv_transformations with a pre-computed-path shortcut (adding the "read from parquet if offline=True and path exists" branch at the top), rather than creating a parallel function that reimplements UDF execution.

Agreed 🙌
Refactored exactly as you suggested. _apply_bfv_transformations_for_historical is deleted. _apply_bfv_transformations now has two branches:

Pre-computed path shortcut (offline=True + batch_source.path) which reads parquet directly, falls back on error

UDF execution via has_transformation() / get_transformation_function() which is unchanged from upstream

No duplication now.. Please review !

…orical_features The function and its call were removed in this PR but the replacement (_apply_bfv_transformations_for_historical) lives in a separate PR (feast-dev#6440). Removing it here would silently return raw untransformed features for any BatchFeatureView with a Python UDF via the standard get_historical_features() API path (FeatureStore → passthrough_provider → SparkOfflineStore). Restoring the function and its call until feast-dev#6440 lands. Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

codecov-commenter · 2026-07-19T12:31:00Z

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

❌ Patch coverage is 30.43478% with 16 lines in your changes missing coverage. Please review.
✅ Project coverage is 45.96%. Comparing base (0da4546) to head (16798e4).
⚠️ Report is 1 commits behind head on master.

Files with missing lines	Patch %	Lines
...ffline_stores/contrib/spark_offline_store/spark.py	26.66%	11 Missing ⚠️
...stores/contrib/spark_offline_store/spark_source.py	50.00%	1 Missing and 2 partials ⚠️
sdk/python/feast/feature_store.py	0.00%	2 Missing ⚠️
❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #6440      +/-   ##
==========================================
- Coverage   45.97%   45.96%   -0.01%     
==========================================
  Files         412      412              
  Lines       48793    48814      +21     
  Branches     6904     6908       +4     
==========================================
+ Hits        22431    22437       +6     
- Misses      24813    24826      +13     
- Partials     1549     1551       +2

Flag	Coverage Δ
go-feature-server	`30.58% <ø> (ø)`
python-unit	`47.26% <30.43%> (-0.01%)`	⬇️

Files with missing lines	Coverage Δ
sdk/python/feast/feature_store.py	`41.03% <0.00%> (-0.03%)`	⬇️
...stores/contrib/spark_offline_store/spark_source.py	`59.89% <50.00%> (+<0.01%)`	⬆️
...ffline_stores/contrib/spark_offline_store/spark.py	`36.16% <26.66%> (-0.36%)`	⬇️

Continue to review full report in Codecov by Harness.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0da4546...16798e4. Read the comment docs.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

SparkSource previously required exactly one of table/query/path. This relaxes the constraint to allow query + path together: - query: used for reading raw data during materialization - path: used for offline write-back (offline=True) and as pre-computed read source in get_historical_features Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

… get_historical_features Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

Catch FileNotFoundError and PermissionError separately for the expected fallback cases (path not yet materialized, or no access). Unexpected errors now emit a distinct RuntimeWarning instead of being silently swallowed by a bare except Exception. Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

The 1.5x speedup assertion for convert_response_to_dict is consistently flaky on macOS CI runners (getting 1.26-1.34x) due to variable load. 1.2x is still a meaningful regression guard without being brittle. Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

…review Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

# [0.65.0](v0.64.0...v0.65.0) (2026-07-20) ### Bug Fixes * add debug logging for FIPS mode detection fallback ([6c1b24e](6c1b24e)) * Build embedded UI from local source ([#6525](#6525)) ([3500349](3500349)) * Bump decommissioned Snowflake Python UDF runtime from 3.9 to 3.10 ([#6606](#6606)) ([#6608](#6608)) ([10341e4](10341e4)) * configure FIPS-compliant gRPC cipher suites for offline server ([6bc80a2](6bc80a2)) * Correct Flink PyArrow dependency constraints ([#6604](#6604)) ([70a9751](70a9751)) * Fix ValueError in signal handling for Trino worker threads ([#6428](#6428)) ([506d919](506d919)) * Fixed monitoring page issues ([7946018](7946018)) * Make pytest config compatible with newer pytest ([#5779](#5779)) ([a57ea33](a57ea33)) * Replace comma with space in DynamoDB-incompatible label tag value ([51e3a16](51e3a16)) * Resolve UI build warnings ([#6529](#6529)) ([abe92af](abe92af)) * Unblock nightly UI build ([#6570](#6570)) ([f296d4b](f296d4b)) * Use LONGBLOB for SQL registry proto columns on MySQL ([#6566](#6566)) ([7e4beb2](7e4beb2)) ### Features * Add click-to-zoom lightbox for blog post images ([#6575](#6575)) ([1cb23fd](1cb23fd)) * Add dark mode support to website and blog ([#6589](#6589)) ([7358fb8](7358fb8)) * Add OnlineStore for Aerospike ([#6532](#6532)) ([9cd35e1](9cd35e1)) * Add OpenLineage Consumer to Feast - receive, store, and visualize cross-producer lineage ([#6549](#6549)) ([a834126](a834126)) * Add registry list feature views by updated since ([#6092](#6092)) ([#6093](#6093)) ([006c606](006c606)) * Add ScyllaDB online store with vector search ([#6508](#6508)) ([1669661](1669661)) * Added compute and jobs UI ([ba2c05c](ba2c05c)) * Added Iceberg REST Catalog data source support ([e0a8573](e0a8573)) * Bring Your Own Spark - SparkApplication ([#6550](#6550)) ([dcd496f](dcd496f)) * **cassandra:** Add multi-DC support via per-datacenter execution profiles ([#6434](#6434)) ([0de9196](0de9196)) * Enhanced data source creation as a visual catalog with type-specific forms ([#6557](#6557)) ([d6acbba](d6acbba)) * Enhanced datasets UI functionality ([de11152](de11152)) * Implement RegistryServer.Proto RPC with RBAC-filtered response ([#6558](#6558)) ([#6552](#6552)) ([0d02614](0d02614)) * New zoned timestamp feature type ([#6536](#6536)) ([#6537](#6537)) ([eb042f0](eb042f0)) * **operator:** Auto-create RBAC for spark_application batch engine ([#6597](#6597)) ([f487b37](f487b37)) * **operator:** integrate cluster TLS profile for OCP 5.0 compliance ([43263a6](43263a6)) * Permissions CRUD UI and OIDC auth integration in UI ([6511da1](6511da1)) * Retrieve historical features from BigQuery without entity_df ([#6569](#6569)) ([cd5f6bb](cd5f6bb)), closes [#6558](#6558) [#6552](#6552) * **spark:** SparkSource query+path and pre-computed offline read for BatchFeatureView ([#6440](#6440)) ([4dc8757](4dc8757)) ### BREAKING CHANGES * total_timeout_ms is renamed to batch_total_timeout_ms. Config files using the old name must be updated. No default value change. Docs updated (reference + perf-tuning guide) with a short explainer on the per-attempt vs total deadline distinction. Two new unit tests pin the policy wiring: socket_timeout_ms propagates to all three scopes, and is omitted (not injected as None) when unset. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * refactor(aerospike): use MAP_KEY_ORDERED, KEY_DIGEST, and instance-scoped client Cheap-win cleanups flagged in review, all touching the same small patch of write-path and lifecycle code. * Map CDTs are now created with MAP_KEY_ORDERED. map_get_by_key / map_remove_by_key on an ordered map are O(log N) in the map size instead of O(N); matters on reads of wide feature views and on the update() background scan (which walks every record in the project's set). * Writes drop POLICY_KEY_SEND and rely on the client default (POLICY_KEY_DIGEST). The serialized entity key is no longer stored alongside each record, saving per-record storage the read path never consumes (batch_operate preserves request order; results are paired back by zip in online_read). * _client moves from a class attribute to an instance attribute (set in __init__). Previously two AerospikeOnlineStore instances could share the cached client through class state until one wrote self._client. With the instance attribute the state is always per-instance from construction. * Drop MongoDB references from class docstrings and comments (they referred to how the storage layout was derived rather than documenting current behavior). Also rewrite the _build_batch_writes docstring to describe the policies applied on the write path. Unit test assertions for the write-path record are updated: bw.policy is now None (client default applies) and map ops carry map_policy={'map_order': MAP_KEY_ORDERED}. All three docker-backed integration tests still pass end-to-end (cross-FV upsert, update() background scan, full feature-store round-trip), so the read/write shape survives the ordering and policy changes against a real server. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * feat(aerospike): add per-FV namespace/set overrides and prewriting hook Adds three configuration knobs to AerospikeOnlineStoreConfig: - namespace_overrides: pin individual feature views to a different Aerospike namespace (e.g. RAM-only vs. SSD-backed) without splitting the project across stores. - set_overrides: place a feature view in its own set so admin ops on it (truncate, scan-based deletes during `feast apply`) do not touch records of other views. - prewriting_hook: import-string-resolved callable invoked once per online_write_batch with the rows about to be written, returning the rows that actually go on the wire. Resolved and cached on first use; returning [] short-circuits the wire call. Read, write, update and teardown paths all honour the per-FV ns/set resolution. update() groups dropped feature views by their resolved (ns, set) pair and issues one background scan per group. teardown() truncates every unique (ns, set) pair the project may have written to, including the store-level default. Adds 22 unit tests for the new behaviour and updates 3 existing call sites of _build_batch_writes for the new namespace= parameter. Adds a sample hook module under examples/online_store/aerospike_overrides_and_hooks/ and corresponding sections in docs/reference/online-stores/aerospike.md. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * test: update aerospike image tag Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * chore: sync README template and secrets baseline after master merge Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * chore: fix secrets baseline line number for v1 operator types Adding aerospike to the feast-operator enum shifted the allowlisted SecretRef entry in api/v1/featurestore_types.go by one line. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * docs: update aerospike docs Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * fix(aerospike): wire batch max_retries and fix empty projection handling Copilot review feedback on PR #6532: - Add max_retries to the batch client policy (batch_operate/batch_write path) - Treat empty projected feature maps as present FV slots (is not None) - Return {} from _normalize_projected_features([]) instead of None - Fix projection unit test mock/assertions - Correct prewriting_hook config docstring Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com> Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * style(aerospike): format online_read docs assignment for ruff Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * chore: update pixi.lock for aerospike optional extra Regenerate the v6 lockfile with Pixi v0.63.1 after adding the aerospike extra to pyproject.toml. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * fix(aerospike): add client init lock and batch chunking Guard lazy client creation with a lock to avoid connection leaks under concurrent first use, and chunk batch reads/writes by batch_max_records so large materializations stay under Aerospike server batch limits. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com>

… BatchFeatureView (feast-dev#6440) * feat: allow query + path in SparkSource for offline materialization SparkSource previously required exactly one of table/query/path. This relaxes the constraint to allow query + path together: - query: used for reading raw data during materialization - path: used for offline write-back (offline=True) and as pre-computed read source in get_historical_features Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> * feat: read from offline path in get_historical_features for BFVs Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> * fix: graceful fallback when offline path is not readable Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> * style: ruff format spark.py and spark_source.py Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> * fix: allow offline-only BatchFeatureView to skip online validation in get_historical_features Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> * fix(spark): narrow exception handling in offline path fallback Catch FileNotFoundError and PermissionError separately for the expected fallback cases (path not yet materialized, or no access). Unexpected errors now emit a distinct RuntimeWarning instead of being silently swallowed by a bare except Exception. Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> * fix(test): lower performance benchmark threshold from 1.5x to 1.2x The 1.5x speedup assertion for convert_response_to_dict is consistently flaky on macOS CI runners (getting 1.26-1.34x) due to variable load. 1.2x is still a meaningful regression guard without being brittle. Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> * refactor: fold pre-computed path into _apply_bfv_transformations per review Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com> --------- Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

# [0.65.0](feast-dev/feast@v0.64.0...v0.65.0) (2026-07-20) ### Bug Fixes * add debug logging for FIPS mode detection fallback ([6c1b24e](feast-dev@6c1b24e)) * Build embedded UI from local source ([feast-dev#6525](feast-dev#6525)) ([3500349](feast-dev@3500349)) * Bump decommissioned Snowflake Python UDF runtime from 3.9 to 3.10 ([feast-dev#6606](feast-dev#6606)) ([feast-dev#6608](feast-dev#6608)) ([10341e4](feast-dev@10341e4)) * configure FIPS-compliant gRPC cipher suites for offline server ([6bc80a2](feast-dev@6bc80a2)) * Correct Flink PyArrow dependency constraints ([feast-dev#6604](feast-dev#6604)) ([70a9751](feast-dev@70a9751)) * Fix ValueError in signal handling for Trino worker threads ([feast-dev#6428](feast-dev#6428)) ([506d919](feast-dev@506d919)) * Fixed monitoring page issues ([7946018](feast-dev@7946018)) * Make pytest config compatible with newer pytest ([feast-dev#5779](feast-dev#5779)) ([a57ea33](feast-dev@a57ea33)) * Replace comma with space in DynamoDB-incompatible label tag value ([51e3a16](feast-dev@51e3a16)) * Resolve UI build warnings ([feast-dev#6529](feast-dev#6529)) ([abe92af](feast-dev@abe92af)) * Unblock nightly UI build ([feast-dev#6570](feast-dev#6570)) ([f296d4b](feast-dev@f296d4b)) * Use LONGBLOB for SQL registry proto columns on MySQL ([feast-dev#6566](feast-dev#6566)) ([7e4beb2](feast-dev@7e4beb2)) ### Features * Add click-to-zoom lightbox for blog post images ([feast-dev#6575](feast-dev#6575)) ([1cb23fd](feast-dev@1cb23fd)) * Add dark mode support to website and blog ([feast-dev#6589](feast-dev#6589)) ([7358fb8](feast-dev@7358fb8)) * Add OnlineStore for Aerospike ([feast-dev#6532](feast-dev#6532)) ([9cd35e1](feast-dev@9cd35e1)) * Add OpenLineage Consumer to Feast - receive, store, and visualize cross-producer lineage ([feast-dev#6549](feast-dev#6549)) ([a834126](feast-dev@a834126)) * Add registry list feature views by updated since ([feast-dev#6092](feast-dev#6092)) ([feast-dev#6093](feast-dev#6093)) ([006c606](feast-dev@006c606)) * Add ScyllaDB online store with vector search ([feast-dev#6508](feast-dev#6508)) ([1669661](feast-dev@1669661)) * Added compute and jobs UI ([ba2c05c](feast-dev@ba2c05c)) * Added Iceberg REST Catalog data source support ([e0a8573](feast-dev@e0a8573)) * Bring Your Own Spark - SparkApplication ([feast-dev#6550](feast-dev#6550)) ([dcd496f](feast-dev@dcd496f)) * **cassandra:** Add multi-DC support via per-datacenter execution profiles ([feast-dev#6434](feast-dev#6434)) ([0de9196](feast-dev@0de9196)) * Enhanced data source creation as a visual catalog with type-specific forms ([feast-dev#6557](feast-dev#6557)) ([d6acbba](feast-dev@d6acbba)) * Enhanced datasets UI functionality ([de11152](feast-dev@de11152)) * Implement RegistryServer.Proto RPC with RBAC-filtered response ([feast-dev#6558](feast-dev#6558)) ([feast-dev#6552](feast-dev#6552)) ([0d02614](feast-dev@0d02614)) * New zoned timestamp feature type ([feast-dev#6536](feast-dev#6536)) ([feast-dev#6537](feast-dev#6537)) ([eb042f0](feast-dev@eb042f0)) * **operator:** Auto-create RBAC for spark_application batch engine ([feast-dev#6597](feast-dev#6597)) ([f487b37](feast-dev@f487b37)) * **operator:** integrate cluster TLS profile for OCP 5.0 compliance ([43263a6](feast-dev@43263a6)) * Permissions CRUD UI and OIDC auth integration in UI ([6511da1](feast-dev@6511da1)) * Retrieve historical features from BigQuery without entity_df ([feast-dev#6569](feast-dev#6569)) ([cd5f6bb](feast-dev@cd5f6bb)), closes [feast-dev#6558](feast-dev#6558) [feast-dev#6552](feast-dev#6552) * **spark:** SparkSource query+path and pre-computed offline read for BatchFeatureView ([feast-dev#6440](feast-dev#6440)) ([4dc8757](feast-dev@4dc8757)) ### BREAKING CHANGES * total_timeout_ms is renamed to batch_total_timeout_ms. Config files using the old name must be updated. No default value change. Docs updated (reference + perf-tuning guide) with a short explainer on the per-attempt vs total deadline distinction. Two new unit tests pin the policy wiring: socket_timeout_ms propagates to all three scopes, and is omitted (not injected as None) when unset. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * refactor(aerospike): use MAP_KEY_ORDERED, KEY_DIGEST, and instance-scoped client Cheap-win cleanups flagged in review, all touching the same small patch of write-path and lifecycle code. * Map CDTs are now created with MAP_KEY_ORDERED. map_get_by_key / map_remove_by_key on an ordered map are O(log N) in the map size instead of O(N); matters on reads of wide feature views and on the update() background scan (which walks every record in the project's set). * Writes drop POLICY_KEY_SEND and rely on the client default (POLICY_KEY_DIGEST). The serialized entity key is no longer stored alongside each record, saving per-record storage the read path never consumes (batch_operate preserves request order; results are paired back by zip in online_read). * _client moves from a class attribute to an instance attribute (set in __init__). Previously two AerospikeOnlineStore instances could share the cached client through class state until one wrote self._client. With the instance attribute the state is always per-instance from construction. * Drop MongoDB references from class docstrings and comments (they referred to how the storage layout was derived rather than documenting current behavior). Also rewrite the _build_batch_writes docstring to describe the policies applied on the write path. Unit test assertions for the write-path record are updated: bw.policy is now None (client default applies) and map ops carry map_policy={'map_order': MAP_KEY_ORDERED}. All three docker-backed integration tests still pass end-to-end (cross-FV upsert, update() background scan, full feature-store round-trip), so the read/write shape survives the ordering and policy changes against a real server. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * feat(aerospike): add per-FV namespace/set overrides and prewriting hook Adds three configuration knobs to AerospikeOnlineStoreConfig: - namespace_overrides: pin individual feature views to a different Aerospike namespace (e.g. RAM-only vs. SSD-backed) without splitting the project across stores. - set_overrides: place a feature view in its own set so admin ops on it (truncate, scan-based deletes during `feast apply`) do not touch records of other views. - prewriting_hook: import-string-resolved callable invoked once per online_write_batch with the rows about to be written, returning the rows that actually go on the wire. Resolved and cached on first use; returning [] short-circuits the wire call. Read, write, update and teardown paths all honour the per-FV ns/set resolution. update() groups dropped feature views by their resolved (ns, set) pair and issues one background scan per group. teardown() truncates every unique (ns, set) pair the project may have written to, including the store-level default. Adds 22 unit tests for the new behaviour and updates 3 existing call sites of _build_batch_writes for the new namespace= parameter. Adds a sample hook module under examples/online_store/aerospike_overrides_and_hooks/ and corresponding sections in docs/reference/online-stores/aerospike.md. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * test: update aerospike image tag Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * chore: sync README template and secrets baseline after master merge Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * chore: fix secrets baseline line number for v1 operator types Adding aerospike to the feast-operator enum shifted the allowlisted SecretRef entry in api/v1/featurestore_types.go by one line. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * docs: update aerospike docs Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * fix(aerospike): wire batch max_retries and fix empty projection handling Copilot review feedback on PR feast-dev#6532: - Add max_retries to the batch client policy (batch_operate/batch_write path) - Treat empty projected feature maps as present FV slots (is not None) - Return {} from _normalize_projected_features([]) instead of None - Fix projection unit test mock/assertions - Correct prewriting_hook config docstring Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com> Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * style(aerospike): format online_read docs assignment for ruff Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * chore: update pixi.lock for aerospike optional extra Regenerate the v6 lockfile with Pixi v0.63.1 after adding the aerospike extra to pyproject.toml. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com> * fix(aerospike): add client init lock and batch chunking Guard lazy client creation with a lock to avoid connection leaks under concurrent first use, and chunk batch reads/writes by batch_max_records so large materializations stay under Aerospike server batch limits. Signed-off-by: Valentyn Kahamlyk <valentin.kagamlyk@gmail.com>

abhijeet-dhumal changed the title ~~Feat/spark bfv offline historical features~~ feat(spark): SparkSource query+path and pre-computed offline read for BatchFeatureView May 27, 2026

abhijeet-dhumal force-pushed the feat/spark-bfv-offline-historical-features branch from a98e23b to 57b2489 Compare May 27, 2026 14:50

abhijeet-dhumal marked this pull request as ready for review May 28, 2026 06:23

abhijeet-dhumal requested a review from a team as a code owner May 28, 2026 06:23

devin-ai-integration Bot reviewed May 28, 2026

View reviewed changes

Comment thread sdk/python/feast/feature_store.py

abhijeet-dhumal force-pushed the feat/spark-bfv-offline-historical-features branch from 57b2489 to e30e146 Compare May 29, 2026 08:57

abhijeet-dhumal force-pushed the feat/spark-bfv-offline-historical-features branch from e30e146 to 4316349 Compare June 1, 2026 07:48

jyejare reviewed Jun 1, 2026

View reviewed changes

abhijeet-dhumal force-pushed the feat/spark-bfv-offline-historical-features branch from 4316349 to 5312075 Compare June 2, 2026 12:45

abhijeet-dhumal requested a review from jyejare June 2, 2026 13:38

jyejare approved these changes Jun 2, 2026

View reviewed changes

ntkathole force-pushed the feat/spark-bfv-offline-historical-features branch from 2f6910e to 9eb3d29 Compare June 3, 2026 08:03

ntkathole added the ok-to-test label Jun 3, 2026

ntkathole reviewed Jun 3, 2026

View reviewed changes

abhijeet-dhumal force-pushed the feat/spark-bfv-offline-historical-features branch from 9eb3d29 to e7fc883 Compare June 9, 2026 07:35

abhijeet-dhumal force-pushed the feat/spark-bfv-offline-historical-features branch from e7fc883 to fce2b7c Compare July 19, 2026 12:27

abhijeet-dhumal requested review from jyejare and ntkathole July 19, 2026 12:39

ntkathole approved these changes Jul 20, 2026

View reviewed changes

abhijeet-dhumal added 4 commits July 20, 2026 14:15

feat: read from offline path in get_historical_features for BFVs

3535a8c

Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

fix: graceful fallback when offline path is not readable

5dc7b6d

Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

style: ruff format spark.py and spark_source.py

7639c2a

Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

abhijeet-dhumal added 4 commits July 20, 2026 14:15

fix: allow offline-only BatchFeatureView to skip online validation in…

b2c76ff

… get_historical_features Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

refactor: fold pre-computed path into _apply_bfv_transformations per …

16798e4

…review Signed-off-by: abhijeet-dhumal <abhijeetdhumal652@gmail.com>

ntkathole force-pushed the feat/spark-bfv-offline-historical-features branch from fce2b7c to 16798e4 Compare July 20, 2026 08:45

ntkathole merged commit 4dc8757 into feast-dev:master Jul 20, 2026
20 of 22 checks passed

-            if udf is not None:
-                temp_view_name = f"__feast_bfv_{ctx.name}_{uuid.uuid4().hex[:8]}"
-                spark_session.conf.set("spark.sql.runSQLOnFiles", "true")
-                raw_df = spark_session.sql(f"SELECT * FROM {ctx.table_subquery}")
++                # Add UDF validation and execution controls
++                if not hasattr(udf, '__call__'):
++                    raise ValueError(f"Invalid UDF for {ctx.name}")
++                raw_df = spark_session.sql(f"SELECT * FROM {ctx.table_subquery}")
++                # Consider adding timeout and resource limits here
++                transformed_df = udf(raw_df)
++                transformed_df.createOrReplaceTempView(temp_view_name)

		max_date_partition: str


		def _apply_bfv_transformations_for_historical(

Uh oh!

Conversation

abhijeet-dhumal commented May 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What this PR does / why we need it

Which issue(s) this PR fixes

Checks

Testing Strategy

Uh oh!

devin-ai-integration Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

abhijeet-dhumal commented Jun 1, 2026

Uh oh!

jyejare left a comment

Choose a reason for hiding this comment

Uh oh!

jyejare Jun 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jyejare Jun 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jyejare Jun 1, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ntkathole Jun 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

codecov-commenter commented Jul 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

abhijeet-dhumal commented May 27, 2026 •

edited

Loading

jyejare Jun 1, 2026 •

edited

Loading

jyejare Jun 1, 2026 •

edited

Loading

jyejare Jun 1, 2026 •

edited

Loading

ntkathole Jun 3, 2026 •

edited

Loading

codecov-commenter commented Jul 19, 2026 •

edited

Loading