RenameFields / Struct Layer

Rename fields within struct columns. Similar to pandas rename() or R's rename() for nested structures. Maintains data structure while updating field names.

Example transformation:

Original struct:

data
{fname: 'John', lname: 'Doe'}
{fname: 'Jane', lname: 'Smith'}

After renaming (fname→first_name, lname→last_name):

data
{first_name: 'John', last_name: 'Doe'}
{first_name: 'Jane', last_name: 'Smith'}

Common applications:

Standardizing field names
Making names more descriptive
Fixing inconsistent naming
Preparing data for specific systems
Harmonizing data structures
Meeting naming conventions

Note: Unmapped field names remain unchanged. The operation preserves the original structure and data types while only updating specified field names.

Table

Select

column

The struct column to modify. Examples of input structures:

User data: {uid: 123, uname: 'John'}
Metrics: {val: 10, desc: 'count'}
Nested: {usr: {id: 1, nm: 'Alice'}}
Mixed: {dt: '2024-01-31', vals: [1,2]}

Mappings

[, ...]

List of field name mappings. Each mapping specifies an old field name and its new name. Example mappings:

{old_field: 'fname', new_field: 'first_name'}
{old_field: 'lname', new_field: 'last_name'}
{old_field: 'tel', new_field: 'phone_number'}

Must contain at least one mapping. Duplicate old_field names are not allowed.

OldField

string

Original field name in the struct. Must exactly match existing field name. Examples:

'fname' to rename first name field
'user_id' to rename identifier
'desc' to rename description

NewField

string

New name for the field. Must be a valid identifier. Examples:

'first_name' for clarity
'identifier' for standardization
'description' for completeness