This release introduces a minor change regarding the assumptions ExcelPython makes about the dimensionality of an input argument when it is not explicitly specified.

• Previous behavior: when the dimensionality of an argument is not explicitly specified using the @xlarg decorator, it is passed to Python as
  • a scalar if the input array is 1-by-1
  • a 2-dimensional tuple otherwise
• New behavior: when the dimensionality of an argument is not explicitly specified using the @xlarg decorator, it is passed to Python as
  • a scalar if the input array is 1-by-1
  • a 1-dimensional tuple if the input array is 1-by-n or n-by-1
  • a 2-dimensional tuple otherwise

If you want to force ExcelPython to behave like previous releases, decorate the function with @xlarg("argname", dims=-2).

If the dimensionality of the argument is explicitly specified by decorating the function with @xlargs with dims equal to 0, 1 or 2, the behavior is the same as in previous releases.