Hi Eike,
One important matter is still open: When opening calc-documents that use the 'old' functions, they are not always computed correctly. The 'new' functions differ in paramter count. I do not know where to add code for backward compatibility when opening these documents, but with some help/hints I might be able to fix that.There are three cases to be considered: 1. reading ODF 1.2 (ODFF) 2. writing ODF 1.0/1.1 (PODF) 3. reading ODF 1.0/1.1 (PODF) [...]
I'm afraid I'm lost here. I can't find my way in/around FormulaCompiler::Factor(). To put it simple: with respect to calc-formulas I understand what the user sees and I (often) understand what's happening in ScInterpreter. But the bit inbetween as rather foggy for me. And to be honest, I doubt if a thorough study of that area will help me much. So I've got another three cases to consider :-/ 1. Give up fdo50950 - I don't like this option, not vanity but I would the functions itself to be available in LibO 3. Get some help from co-developers who are familiar with this kind of poking - that would be nice, but is out of my control 2. Struggle on in the hope of magically getting there - this will be the default as I don't like case 1 I will see that the submodules are included in the patch sofar that I attached to the bug itself and concentrate on other bugs (like IFERROR/IFNA/NUMBERVALUE). Winfried