Parity changelog

How the feature-parity map changed across releases — generated by diffing the parity tables between consecutive v* tags.

Current (latest-eec3e8c): 138 🟢 Real · 9 🟡 Emulated · 3 🟠 BYO-engine · 8 🔴 Not implemented.

v0.1.0 → v0.3.0

• Key Vault references in connections: 🔴 → 🟢
• Blob surface: 🔴 → 🟢
• Delta commits (put-if-absent atomicity): 🔴 → 🟢
• Shortcuts (OneLake → OneLake): 🔴 → 🟢
• notebookutils / mssparkutils (fs, credentials, getSecret, lakehouse, runtime): 🔴 → 🟢
• Spark session / statement / batch via the Livy API: 🔴 → 🟢
• Notebook cell execution: 🔴 → 🟢
• Livy High-Concurrency (5-REPL) sessions: 🔴 → 🟢
• SQL-analytics-endpoint semantics over lakehouse Delta: 🔴 → 🟢
• T-SQL over TDS + Entra FedAuth: 🔴 → 🟢
• Lakehouse SQL analytics endpoint — Delta → engine: 🔴 → 🟢
• Warehouse — read-write T-SQL: 🔴 → 🟢
• Per-item isolation (each item = its own SQL Server database): 🔴 → 🟢
• RBAC → SQL permissions: 🔴 → 🟢
• information_schema / sys. introspection: 🔴 → 🟢
• Per-column type fidelity (real SQL types over the wire): 🔴 → 🟢
• Connection by item name (vs GUID): 🔴 → 🟢
• Data Pipeline control flow (If / ForEach / Until / Switch / Filter / Fail, expression language, dependsOn): 🔴 → 🟢
• Pipeline → notebook activity (TridentNotebook): 🔴 → 🟢
• queryactivityruns detail: 🔴 → 🟢
• Copy activity — OneLake → OneLake: 🔴 → 🟢
• Lookup activity — OneLake CSV/JSON/Parquet/Delta: 🔴 → 🟢
• GetMetadata activity — OneLake path: 🔴 → 🟢
• Web / external-connector leaves: 🔴 → 🟡
• Mirroring — Mirrored Database (mirroring/): 🟡 → 🟢
• Power BI — Semantic Model query (executeQueries): 🔴 → 🟢
• Power BI — Reports / rendering; full DAX; SemPy over XMLA: 🔴 → 🟡
• Fabric CLI fab (Microsoft): 🔴 → 🟢
• deltalake (delta-rs): 🔴 → 🟢
• azure-storage-file-datalake + Blob SDK: 🔴 → 🟢
• azcopy (Microsoft): 🔴 → 🟢
• DuckDB: 🔴 → 🟢
• PySpark behind the Livy API: 🔴 → 🟢
• notebookutils: 🔴 → 🟢
• go-mssqldb: 🔴 → 🟢
• dbt-fabricspark (Microsoft): 🔴 → 🟢
• dbt-fabric (Microsoft): 🔴 → 🟢
• abfss://…@onelake.dfs.fabric.microsoft.com/… production paths: added 🟢
• Delta write/read/append; SQL over temp views: added 🟢
• Time travel option(“versionAsOf”, n): added 🟢
• MERGE INTO: added 🟢
• createDataFrame(local_rows): added 🟢
• sc / RDD API / spark._jvm: added 🔴
• DML row-count envelopes (INSERT/MERGE counts): added 🟡
• Structured streaming, OPTIMIZE/VACUUM, CDF, Java/Scala UDFs, spark.jars: added 🔴
• Concurrent Delta writers to one table: added 🔴
• Fabric SQL Database (database/) — OLTP + OneLake mirror: added 🟢
• Per-activity policy — retry + backoff + timeout: added 🟢
• ForEach sequential / parallel (isSequential, batchCount): added 🟢
• List pagination (continuationToken): added 🟢
• Invoke pipeline (ExecutePipeline): added 🟢
• Script / SqlServerStoredProcedure activities: added 🟢
• Fabric SQL Database (database/), Graph (graph/), Real-Time Hub, Copilot / IQ (iq/), Embed, Workload Dev Kit: added 🔴
• Azure Synapse dedicated SQL pool (dbt-synapse): added 🟢
• Azure Data Lake Analytics — U-SQL / SCOPE (dbt-scope): added 🟠
• ADLS Gen1: added
• ADLS Gen2 (standalone storage account): added 🟢

v0.3.0 → v0.4.0

• Purview / sensitivity labels (governance/): added 🔴
• Lineage (catalog graph): added 🟢
• Purview / lineage / sensitivity labels (governance/): removed

v0.4.0 → v0.4.1

No parity changes.

v0.4.1 → v0.5.0

• Shortcuts to external targets (S3 / ADLS Gen2 / Dataverse): 🔴 → 🟢
• Dataflow Gen2 (Power Query M engine): 🔴 → 🟡
• Data Science — ML models / experiments / MLflow (data-science/): 🔴 → 🟢
• dbt-fabricspark (Microsoft): 🟢 → 🟠
• Environments: added 🟢
• Spark Job Definitions: added 🟢
• Structured streaming, OPTIMIZE/VACUUM, Java/Scala UDFs: added 🔴
• CDF options, spark.jars: added 🔴
• Concurrent Delta overwrite writers: added 🟢
• Activity-level lineage: added 🟢
• Apache Airflow Job: added 🟢
• Deployment pipelines — model, assignment, item pairing, Deploy Stage Content (D0–D2): added 🟢
• Deployment pipelines — role-assignment CRUD (D3): added 🟢
• Fabric Data Engineering VS Code extension 1.18.1 contract (Microsoft): added 🟢
• Apache Airflow 2.10.5: added 🟢
• MLflow 3 + dbt-duckdb: added 🟢
• Environments, Spark Job Definitions: removed
• Structured streaming, OPTIMIZE/VACUUM, CDF, Java/Scala UDFs, spark.jars: removed
• Concurrent Delta writers to one table: removed
• Deployment pipelines: removed

v0.5.0 → v0.6.0

• Real-Time Intelligence — Eventhouse / KQL Database (real-time-intelligence/): added 🟡
• Real-Time Intelligence — Eventstream (real-time-intelligence/event-streams/): added 🟡
• azure-kusto-data (Microsoft) + raw Kusto REST, over kustainer (Microsoft’s own KQL engine): added 🟠
• Real-Time Intelligence — Eventhouse / KQL DB / Eventstream (real-time-intelligence/): removed

v0.6.0 → v0.7.0

• Governance domains (/v1/admin/domains): added 🟢
• Tenant settings / audit (activityevents) / other admin-portal APIs: added 🔴
• CopyJob, KQLDashboard, KQLQueryset, Reflex (Data Activator), WarehouseSnapshot: added 🟡
• Tenant settings / audit / admin-portal APIs: removed

v0.7.0 → v0.7.1

No parity changes.

v0.7.1 → v0.8.0

• Activity log / audit (GET /v1.0/myorg/admin/activityevents): added 🟢
• Tenant settings / other admin-portal APIs: added 🔴
• Graph (graph/), Real-Time Hub, Copilot / IQ (iq/), Embed, Workload Dev Kit: added 🔴
• Tenant settings / audit (activityevents) / other admin-portal APIs: removed
• Fabric SQL Database (database/), Graph (graph/), Real-Time Hub, Copilot / IQ (iq/), Embed, Workload Dev Kit: removed

v0.8.0 → v0.9.0

• Items CRUD + typed collections: added 🟢
• Tenant settings (GET /v1/admin/tenantsettings): added 🟢
• Tenant-wide workspace admin (GET /v1/admin/workspaces): added 🟢
• Tenant-wide item admin (GET /v1/admin/items): added 🟢
• Capacity tenant-setting overrides (GET /v1/admin/capacities/delegatedTenantSettingOverrides, POST …/{capacityId}/delegatedTenantSettingOverrides/{name}/update): added 🟢
• Sensitivity labels (bulkSetLabels / bulkRemoveLabels): added 🟢
• Purview scanning / classification (governance/): added 🔴
• Structured streaming: added 🔴
• OPTIMIZE / VACUUM (Delta maintenance): added 🔴
• Java/Scala UDFs, spark.jars: added 🔴
• Items CRUD + 12 typed collections: removed
• Tenant settings / other admin-portal APIs: removed
• Purview / sensitivity labels (governance/): removed
• Structured streaming, OPTIMIZE/VACUUM, Java/Scala UDFs: removed

v0.9.0 → v0.10.0

• OPTIMIZE / VACUUM (Delta maintenance): 🔴 → 🟢
• dbt-fabricspark (Microsoft): 🟠 → 🟢
• Time travel — DataFrame option and SQL VERSION AS OF: added 🟢
• Change Data Feed: added 🟢
• spark.jars: added 🔴
• Time travel option(“versionAsOf”, n): removed
• CDF options, spark.jars: removed

v0.10.0 → v0.11.0

No parity changes.

v0.11.0 → v0.11.1

No parity changes.

v0.11.1 → v0.12.0

• List Item Job Instances (GET …/jobs/instances): added 🟢
• Item Job Scheduler (…/items/{id}/jobs/{jobType}/schedules): added 🟢
• Sensitivity labels → catalog: added 🟢
• Nested CTEs (WITH x AS (WITH y AS …)): added 🟢
• CTAS (CREATE TABLE … AS SELECT): added 🟢
• Class B strict mode (-tsql-strict): added 🟢
• CopyJob, KQLDashboard, KQLQueryset, WarehouseSnapshot: added 🟡
• Event triggers — Reflex (Data Activator) on OneLake events: added 🟢
• CopyJob, KQLDashboard, KQLQueryset, Reflex (Data Activator), WarehouseSnapshot: removed

v0.12.0 → v0.13.0

• Data-flow lineage beyond OneLake: added 🟢

v0.13.0 → v0.13.1

No parity changes.

v0.13.1 → v0.14.0

No parity changes.

v0.14.0 → v0.14.1

No parity changes.

v0.14.1 → v0.15.0

No parity changes.

v0.15.0 → v0.15.1

No parity changes.

v0.15.1 → v0.15.2

No parity changes.

v0.15.2 → v0.15.3

No parity changes.

v0.15.3 → v0.16.0

No parity changes.

v0.16.0 → v0.16.1

No parity changes.

v0.16.1 → v0.17.0

• Environments: 🟢 → 🟡

v0.17.0 → v0.18.0

• Environments: 🟡 → 🟢
• DML row-count envelopes (INSERT/MERGE counts): 🟡 → 🟢
• notebookutils.notebook.runMultiple — DAG structure and ordering: added 🟢
• notebookutils.notebook.run — exit values: added 🟢
• runMultiple — failure contract: added 🟢
• Reference-run lakehouse rule: added 🟢
• runMultiple — retry and timeouts: added 🟢
• runMultiple — concurrency: added 🟡
• runMultiple — child Spark sessions: added 🟡
• Web / WebHook activity: added 🟢
• REST connector — RestSource and RestSink in a Copy, with pagination: added 🟢
• Salesforce — SalesforceV2Source and SalesforceV2Sink in a Copy: added 🟢
• External-connector leaves (ServiceNow, …): added 🟡
• Web / external-connector leaves: removed

v0.18.0 → v0.19.0

• Item job execution (jobs/instances): 🟡 → 🟢
• Purview Data Map ({endpoint}/datamap/api/atlas/v2/…) — split out of the old single “Purview scanning / classification” row, because its three parts have different ceilings: this one is reachable and moving, scanning is reachable in-family, and the built-in classifiers can never move. One 🔴 made them look equally stuck, and became actively wrong the moment this landed: added 🟡
• Purview scanning (scan/): added 🔴
• Purview system classifiers (the ~200 built-ins): added 🔴
• Materialized lake views: added 🟢
• Delete data activity: added 🟢
• Script / SqlServerStoredProcedure / SqlPoolStoredProcedure activities: added 🟢
• HDInsight Spark activity (HDInsightSpark): added 🟢
• Azure Databricks activities (DatabricksNotebook, DatabricksSparkPython, DatabricksSparkJar): added 🟢
• Azure Batch activity (ADF Custom): added 🟠
• Azure ML activities (AzureMLExecutePipeline, AzureMLBatchExecution, AzureMLUpdateResource): added 🔴
• Validation activity: added 🟢
• Azure Data Explorer Command activity (AzureDataExplorerCommand): added 🟢
• HDInsight Hive/Pig/MapReduce/Streaming, U-SQL, SSIS package activities: added 🔴
• Functions activity (AzureFunctionActivity): added 🟢
• Web and WebHook activities: added 🟢
• ServiceNow via RestSource: added 🟢
• CopyJob: added 🟢
• KQLDashboard, KQLQueryset, WarehouseSnapshot: added 🟡
• Power BI — Report items: added 🟢
• Power BI — DAX beyond the bounded subset: added 🟡
• Power BI — SemPy / semantic-link-labs over XMLA: added 🔴
• Purview scanning / classification (governance/): removed
• Script / SqlServerStoredProcedure activities: removed
• Web / WebHook activity: removed
• CopyJob, KQLDashboard, KQLQueryset, WarehouseSnapshot: removed
• Power BI — Reports / rendering; full DAX; SemPy over XMLA: removed

v0.19.0 → v0.20.0

No parity changes.

v0.20.0 → v0.21.0

• Purview Data Map ({endpoint}/datamap/api/atlas/v2/…) — split out of the old single “Purview scanning / classification” row, because its three parts have different ceilings: this one is reachable and moving, scanning is reachable in-family, and the built-in classifiers can never move. One 🔴 made them look equally stuck, and became actively wrong the moment this landed: 🟡 → 🟢
• Power BI — SemPy / semantic-link-labs over XMLA: 🔴 → 🟢
• Asynchronous outcomes forced (FABRIC_FORCE_LRO): added 🟢
• Variable Library consumed by a Data Pipeline (libraryVariables, @pipeline().libraryVariables.): added 🟢

v0.21.0 → v0.22.0

• Deleted display names held (FABRIC_NAME_RESERVATION): added 🟢
• notebookutils.notebook.run — exit values OVER REST: added 🟡
• Spark Job Definition and Invoke copy job activities (SparkJobDefinition, InvokeCopyJob): added 🟢

v0.22.0 → v0.23.0

• sc / RDD API / spark._jvm: 🔴 → 🟡

v0.23.0 → v0.24.0

• ARM-created capacities (FABRIC_ARM_URL): added 🟢

v0.24.0 → v0.25.0

• Structured streaming: 🔴 → 🟢
• Azure Batch activity (ADF Custom): 🟠 → 🟢
• Real-Time Intelligence — Eventhouse / KQL Database (real-time-intelligence/): 🟡 → 🟢
• Real-Time Intelligence — Eventstream (real-time-intelligence/event-streams/): 🟡 → 🟢
• Catalog Search (POST /v1/catalog/search): added 🟢
• Fabric Core MCP Server (POST /v1/mcp/core): added 🟢
• read.json(multiLine=True): added 🟢
• Eventstream — Lakehouse destination: added 🟢
• Eventstream — Reflex destination: added 🟢
• Eventstream — Eventhouse destination: added 🟢
• Eventstream — operators: added 🟢
• microsoft/fabric Terraform provider (Microsoft): added 🟢
• Apache Kafka (apache/kafka KRaft) + Fabric Eventstream notebook API on Sail and JVM: added 🟠

v0.25.0 → v0.26.0

No parity changes.

v0.26.0 → v0.27.0

No parity changes.

v0.27.0 → v0.28.0

No parity changes.

v0.28.0 → v0.29.0

No parity changes.

v0.29.0 → v0.30.0

No parity changes.

v0.30.0 → v0.31.0

No parity changes.

v0.31.0 → v0.32.0

No parity changes.

v0.32.0 → v0.33.0

• OneLake security roles (dataAccessRoles): added 🟢
• Row- and column-level security, applied by the engine: added 🟢
• Two-context execution — the user context holds no service credential: added 🟢
• Direct path access blocked for a narrowed principal: added 🟢
• Row-level security (T-SQL) — CREATE SECURITY POLICY … FILTER PREDICATE: added 🟢
• Column-level security (T-SQL) — GRANT SELECT ON t(col): added 🟢
• Dynamic data masking — ALTER TABLE … MASKED WITH: added 🟢

v0.33.0 → latest-eec3e8c

No parity changes.