API Overview

Fabric Data APIs are divided into five core components: data processing, data management, input/output, utility, and action.

Data processing: Offers a variety of transformation operations such as map(), flat_map(), filter(), join(), and groupby() to perform feature engineering and structured data manipulation on datasets.
Data management: Provides capabilities for dataset create, read, update, delete (CRUD) operations, including insert(), update(), delete(), and index creation/deletion, enabling comprehensive lifecycle management.
Input/output: Supports seamless integration with popular lakehouse formats like Parquet, Iceberg, and Data Formation for efficient data writing.
Utility: Includes functions like schema(), columns(), and explain_plan() for metadata queries and execution plan analysis, aiding in debugging and optimization.
Action: Uses methods such as execute(), limit(), and take() to initiate computations and retrieve results, supporting lazy loading and on-demand execution.

**Table 1**
Operation Type	API	Description
Dataset – Data processing	map	Applies a one-to-one function mapping to single-row input data.
	map_batchs	Applies a one-to-one function mapping to batch input data.
	flat_map	Applies a one-to-many function mapping to single-row input data.
	filter	Filters rows based on specified conditions.
	join	Joins multiple datasets.
	order_by	Sorts the dataset.
	aggregate	Aggregates the entire dataset.
	groupby	Groups the dataset.
	min	Calculates the minimum value of a specified column.
	max	Calculates the maximum value of a specified column.
	mean	Calculates the mean value of a specified column.
	unique	Retrieves a list of unique values from a specified column.
	select_columns	Selects specific columns from the dataset.
	add_column	Adds a column to the dataset.
	drop_columns	Removes specified columns from the dataset.
	rename_columns	Renames columns in the dataset.
Table – Data management	insert	Inserts data into the dataset.
	delete	Deletes data from the dataset.
	update	Updates existing data within the dataset.
Input/Output	write_parquet	Writes data to a Parquet table.
Input/Output	write_iceberg	Writes data to an Iceberg table.
Utility	schema	Displays the schema of the dataset.
	columns	Lists all column names in the dataset.
	count	Returns the total number of rows in the dataset.
	explain_plan	Prints the execution plan for a query.
	explain_performance	Executes and prints detailed performance metrics for a query.
	stats	Displays statistical information about the executed query (requires prior execution).
Action	show	Triggers execution and displays the results.
	execute	Triggers execution and returns the results.
	limit	Outputs up to a specified number of records.
	take	Triggers execution and returns a single-row iterator.
	take_batch	Triggers execution and returns a batch iterator.