CastToPhysical / Manipulation Layer
Convert temporal and categorical data types to their underlying numeric representations. Similar to numpy's datetime64.astype('int64') or R's as.integer(as.factor()).
Conversions performed:
- Date → Int32 (days since epoch)
- DateTime → Int64 (timestamp units since epoch)
- Time → Int64 (time units since midnight)
- Categorical → UInt32 (category codes)
Common applications:
- Low-level data processing
- Custom temporal calculations
- Memory optimization
- Binary data storage
- System interoperability
Note: This operation is typically used for specialized processing needs or optimization purposes. Regular temporal operations should use the standard temporal types.
Transforms
[, ...]List of columns to convert to their physical representations. Multiple transforms allow parallel processing of different columns. Common scenarios:
- Bulk temporal data optimization
- Mixed type conversions
- System-level data preparations
Select
columnColumn to convert to physical representation. Supported types and results:
- Date columns → Int32 (days from epoch)
- DateTime columns → Int64 (time units from epoch)
- Time columns → Int64 (units since midnight)
- Categorical columns → UInt32 (category codes) Other data types remain unchanged.
AsColumn
nameName for the new column. If not provided, the system generates a unique name. If AsColumn matches an existing column, the existing column is replaced. The name should follow valid column naming conventions.