Typed Microsoft Fabric item discovery — fabric_typed

These shortcuts cover an intentional subset of Microsoft Fabric item types; they are not an exhaustive list of the items that fabric_items() can discover. Each helper requests one exact type and has a corresponding FabricWorkspace method. Most retrieve workload connection details by default. Semantic Model and GraphQL helpers default to lightweight discovery because their executable targets are derived from list-level IDs and workspace fields. User Data Functions default to lightweight discovery because Microsoft limits detail retrieval to delegated user identities. Set detail = TRUE when the workload and identity support it

Usage

fabric_lakehouses(workspace, detail = TRUE, ...)

fabric_warehouses(workspace, detail = TRUE, ...)

fabric_warehouse_snapshots(workspace, detail = TRUE, ...)

fabric_mirrored_databases(workspace, detail = TRUE, ...)

fabric_sql_databases(workspace, detail = TRUE, ...)

fabric_semantic_models(workspace, detail = FALSE, ...)

fabric_eventhouses(workspace, detail = TRUE, ...)

fabric_kql_databases(workspace, detail = TRUE, ...)

fabric_notebooks(workspace, detail = TRUE, ...)

fabric_data_pipelines(workspace, detail = TRUE, ...)

fabric_spark_job_definitions(workspace, detail = TRUE, ...)

fabric_environments(workspace, detail = TRUE, ...)

fabric_graphql_apis(workspace, detail = FALSE, ...)

Arguments

workspace: Workspace name, ID, or object returned by fabric_workspaces(). A name is convenient for interactive use; an object avoids an extra lookup
detail: Whether to retrieve connection details as well as names and IDs. This takes more requests and may require additional permissions. For fabric_item(), NULL enriches every supported type except User Data Functions, whose detail endpoint does not support application identities. The typed Semantic Model, GraphQL, and User Data Function helpers also default to lightweight records
...: Authentication and API arguments forwarded to fabric_items() Do not supply type; each helper sets that value

Value

A list with one FabricItem object or type-specific R6 subclass per matching item. Each object contains common item metadata, applicable connection fields, and workload methods. See fabric_items() for details

Typed support matrix

Default detail is the value used when detail is omitted. FabricItem in the final column means that the typed helper and workload Get route are supported but no workload-specific R6 subclass is currently provided.

Helper	Fabric type	Default detail	R6 class
`fabric_lakehouses()`	`Lakehouse`	`TRUE`	`FabricLakehouse`
`fabric_warehouses()`	`Warehouse`	`TRUE`	`FabricWarehouse`
`fabric_warehouse_snapshots()`	`WarehouseSnapshot`	`TRUE`	`FabricWarehouseSnapshot`
`fabric_mirrored_databases()`	`MirroredDatabase`	`TRUE`	`FabricMirroredDatabase`
`fabric_sql_databases()`	`SQLDatabase`	`TRUE`	`FabricSqlDatabase`
`fabric_semantic_models()`	`SemanticModel`	`FALSE`	`FabricSemanticModel`
`fabric_eventhouses()`	`Eventhouse`	`TRUE`	`FabricEventhouse`
`fabric_kql_databases()`	`KQLDatabase`	`TRUE`	`FabricKqlDatabase`
`fabric_notebooks()`	`Notebook`	`TRUE`	`FabricJobItem`
`fabric_data_pipelines()`	`DataPipeline`	`TRUE`	`FabricJobItem`
`fabric_spark_job_definitions()`	`SparkJobDefinition`	`TRUE`	`FabricJobItem`
`fabric_environments()`	`Environment`	`TRUE`	`FabricItem`
`fabric_user_data_functions()`	`UserDataFunction`	`FALSE`	`FabricItem`
`fabric_graphql_apis()`	`GraphQLApi`	`FALSE`	`FabricGraphQLApi`

Choosing a helper

fabric_lakehouses(), fabric_warehouses(), fabric_warehouse_snapshots(), and fabric_mirrored_databases() find data stores with $sql_query() (fabric_sql_query()) and other SQL methods; Lakehouses and mirrored databases can also be accessed through OneLake
fabric_sql_databases() finds transactional Fabric SQL databases
fabric_semantic_models() finds business models with $dax_query() (fabric_pbi_dax_query()) and refresh lifecycle methods
fabric_eventhouses() and fabric_kql_databases() find real-time data stores with $query() (fabric_kql_query()) and $read_table() (fabric_kql_read_table()), plus ingestion and export methods
fabric_notebooks() finds notebooks with job lifecycle and schedule methods
fabric_data_pipelines() and fabric_spark_job_definitions() find the other executable items with the same job methods
fabric_environments() finds reusable Spark runtime configurations
fabric_user_data_functions() finds serverless Python function items
fabric_graphql_apis() finds APIs configured in Fabric with $query() (fabric_graphql_query()), $schema() (fabric_graphql_schema()), and $paginate() (fabric_graphql_paginate())

Filtering and returned fields

Each helper requests its exact Fabric item type and verifies that every returned object has that type. The objects otherwise keep all fields returned by Fabric, including fields added by the service in the future

Folder recursion, workspace-specific private-link routing, authentication, and detail_errors have the same behavior as in fabric_items(). With detail = TRUE, each helper calls its documented workload-specific Get API and preserves fields such as Spark job and Environment properties. The User Data Function detail endpoint supports delegated users but not service principals or managed identities; those callers can use detail = FALSE

References

List items REST API

List data pipelines

List Spark job definitions

List environments

List User Data Functions

Get data pipeline

Get Spark job definition

Get environment

Get User Data Function

Examples

if (FALSE) { # \dontrun{
# Discover a workspace once, then reuse its object for typed discovery
workspace <- fabric_workspaces()[[1]]

# Discover data items that feed the package's query and storage helpers
lakehouses <- fabric_lakehouses(workspace)
warehouses <- fabric_warehouses(workspace)
snapshots <- fabric_warehouse_snapshots(workspace)
mirrored_databases <- fabric_mirrored_databases(workspace)
sql_databases <- fabric_sql_databases(workspace)
semantic_models <- fabric_semantic_models(workspace)
eventhouses <- fabric_eventhouses(workspace)
kql_databases <- fabric_kql_databases(workspace)
graphql_apis <- fabric_graphql_apis(workspace)

# Each method calls the corresponding exported function
# fabric_lakehouse_tables()
lakehouses[[1L]]$tables()
# fabric_sql_connection_info()
warehouses[[1L]]$sql_connection_info()
# fabric_pbi_dax_query()
semantic_models[[1L]]$dax_query(
  dax = Sys.getenv("FABRIC_DAX_QUERY")
)

# Runnable methods call fabric_job_run() and fabric_job_wait()
notebook <- fabric_notebooks(workspace)[[1]]
pipeline <- fabric_data_pipelines(workspace)[[1]]
spark_job <- fabric_spark_job_definitions(workspace)[[1]]

notebook$wait(notebook$run(), timeout = 900)
pipeline$wait(pipeline$run(), timeout = 900)
spark_job$wait(spark_job$run(), timeout = 900)

# Discover supporting Spark and serverless-function items as well
environments <- fabric_environments(workspace)
functions <- fabric_user_data_functions(workspace)
} # }