Introduction
Whenever a function expects a function parameter for table columns, the following types are supported:
| Type | Description | Example |
|---|---|---|
| Numeral | Specifies a table column number, beginning with 0. Negative indexing is allowed, i.e. -1 is last column in the top row, counting leftward. | 0 (left most column) -2 (2nd. last column) |
| Quoted string | The content of the entire string is interpreted as one header name. Empty strings ("") refer to the first column with blank header name. | "Last,First Name" refers to one header name "" is a blank header name |
| Softquoted string | The content of the entire string is interpreted as one header name. Empty strings ('') refer to the first column with blank header name. shifted table column specifications are allowed here to reference neighboring columns. Some function-specific exceptions may apply and are explicitly documented in the function library documentation. | 'Last,First Name' refers to one header name '' is a blank header name |
| Parameter set | Parameter sets contain a colection of header names and/or column numbers | { 1, Last Name, first Name, "Value [EUR]" } |
Addtional function-specific rules may apply. Below are some examples:
- Header names must exist
- Header names may exist (missing ones may be created as additinal columns in the table)
- Header names must not exist (e.g. to check if specific header names are not used in a specific table
- Column numbers are allowed, or not allowed
- Column numbers may only refer to existing colums, or alternatively all columns even if not yet existing
- Header names or column numbers must be unique
- etc.
Note: If Table headers contain numbers, then you can find these headers by specifying the numbers as strings, e.g. '10', or { ..., '10'. ...}.