akasolace

@govert thank you for the detailed answer. Regarding your first question. I am in the first use case you described, passing DataFrame fom/to Excel (the second use case is not...

> These wrappers can be automatically generated at runtime, using the ExcelDna.Registration extension. So if you can define the two conversions, the rest is done already. I am really not...

what about `object[,]` -> csv string using existing nugget and then use the function `DataFrame.LoadCSV(string csv)` ? That would be much simpler, right ?

@govert as you advised, in order to export DataFrame instead of returning the DataFrame `df`, I am simply returning `CustomMarshaler.dataFrameToObjects(df)` with the code below. Is it the kind of conversion...

The function `dataFrameToObjects ` is very easy, it is the one I put in my previous post. I also implemented `objectsToDataFrame` in my code. To do so I had also...

@minj no I am not for the exact same reason ... don’t want to pay to distribute free software :-) Sent with GitHawk

@keithalewis my function is returning an array double{], what you are suggesting is to return a matrix double[ , ] with only the first column being filled in, correct?

Thank you @govert and @keithalewis. If I recap I have 3 options: 1) call transpose from Excel (but not what I want to do) 2) call XlCall.Excel(XlCall.xlfTranspose, myArray) (expecting worst...

> 4. https://github.com/xlladdins/xll/blob/master/test/transpose.cpp > If this is not 10x faster than the other methods, I'll eat my hat. @keithalewis sorry for my ignorance but you are pointing to another project...

To be honest I had no idea how to approach it. I naively thought that this could be done manipulating somehow that part of the code https://github.com/Excel-DNA/ExcelDnaDoc/blob/611df5cb5f3fd88b58f4a8b032c23741374d54b2/src/ExcelDnaDoc.Html/Utility/ILInspector.cs#L83 Yesterday, meanwhile I...