Dataflow-referentie: datatypen & conversie
Begrijp het verschil tussen veldtypen, conversie en validatie.
Op deze pagina
De navigatie en paginasamenvatting zijn Nederlands. De technische inhoud op deze pagina is voorlopig nog in het Engels en wordt stapsgewijs vertaald.
Type validation: A numeric value typed directly as a fixed text value may use either a point or a comma as decimal separator. A connection from a source field or another functoid is different: linked values are type-checked. When the linked output type and the target input type are incompatible, the designer reports a validation error instead of converting automatically.
Field types and conversion
Every field arrives with a field type. When the incoming values do not fit that type, a number field that contains n/a for example, you decide what happens. Open Field Options from the gear at the end of a field row.
- Type: String, Integer, Decimal or Boolean. Set this when a source delivers a number as text, or the other way around.
- Parser format: an optional pattern describing how incoming values are written, such as
dd-MM-yyyy. - If conversion fails:
- Abort import stops the import and reports the value. This is the default, so unusable values do not pass unnoticed.
- Skip row leaves that record out and carries on.
- Use default value substitutes the value typed in Default value.
- Use empty value leaves the field empty.
Values that arrive empty stay empty whatever the field type. They do not count as conversion failures.
Inputs that expect a number
Inputs that expect a number are checked. Hovering an input says whether it needs a whole number or accepts decimals. A value that is not a number, or a text field connected to it, is reported when you import and marked on the node.