When importing data from an external data source, in order for it to be reliable queryable, LUSID needs to know something about it. A data definition tells LUSID, what a given external data item is, what type it is and whether it in some way identifies items of data. Consider presenting LUSID with a list of dictionaries where each dictionary contains the same set of keys (names). Each data item pointed to by a key would be expected to be of the same type (integer, string, decimal etc.). To identify a particular dictionary from the list, a tuple of one or more of the items in the dictionary would make it unique. If only a single item is required then the
| Name | Type | Description | Notes |
|---|---|---|---|
| address | String | The internal address (LUSID native) of the unit in the provided data itself and corresponds to the external name of the data item | |
| name | String | The name of the data item. This is the name that will appear | [optional] |
| dataType | String | A member of the set of possible data types, that all data passed under that key is expected to be of. Currently limited to one of [string, integer, decimal, result0d]. | [optional] |
| keyType | String | Is the item either a unique key for the dictionary, i.e. does it identify a unique index or conceptual 'row' within the list of dictionaries, or a partial key or is it simply a data item within that dictionary. Must be one of [Unique,PartOfUnique,Leaf, CompositeLeaf] | [optional] |
| allowNull | Boolean | The path to the field must exist (unless AllowMissing is true) but the actual value is allowed to be null. | [optional] |
| allowMissing | Boolean | The path (or column) is allowed to be missing but if it is present it is not allowed to be null unless AllowNull is true. | [optional] |