Stream transformation for large datasets.

Returns a stream that transforms rows in batches to avoid loading all data into memory at once.

@spec transform(list(), list(), map(), keyword()) :: {:ok, [map()]} | {:error, term()}

Transform rows, columns, and aliases to list of maps.

parameters

Parameters

• rows - List of row data (list of lists)
• columns - List of column names
• aliases - Map of column aliases
• options - Transformation options

options

Options

• :keys - Key type (:strings, :atoms, :existing_atoms). Default: :strings
• :transform - Key name transformation (:none, :camelCase, :snake_case). Default: :none
• :coerce_types - Whether to coerce database types to Elixir types. Default: false
• :null_handling - How to handle NULL values (:preserve, :remove). Default: :preserve

examples

Examples

# String keys (default)
{:ok, maps} = transform(rows, columns, aliases, [])
# => [%{"name" => "John", "age" => 25}, ...]

# Atom keys
{:ok, maps} = transform(rows, columns, aliases, keys: :atoms)
# => [%{name: "John", age: 25}, ...]

# CamelCase transformation
{:ok, maps} = transform(rows, columns, aliases, keys: :atoms, transform: :camelCase)
# => [%{firstName: "John", userAge: 25}, ...]

@spec transform_with_types(list(), list(), map(), keyword()) ::
  {:ok, [map()]} | {:error, term()}

Transform rows with type coercion based on database column types.

This is a more advanced version that can coerce database types to proper Elixir types based on PostgreSQL column type information.