When a csv is exported the entire range which has values is exported. When export is done in Excel, the cells of this range which are empty are marked by an additional field, separated by a comma. This comma is present in SpreadProcessing only in places where it is obligatory for example in the middle of the row. Obligatory comma in this context means that without it the file will not be read correctly by Excel or SpreadProcessing. In places where it is not obligatory, like in an empty row or in the end of the row, it is skipped. These non-obligatory commas do not influence how the file will be interpreted by Excel, however, according to the most-popular csv spec, they should be present, which means that there could be parsers which are unable to parse the resulting file. (2.4 Each line should contain the same number of fields throughout the file.) 

Example:
Exported by Excel:
---------------------------------------
Product,Unit Price,Units in Stock,Discontinued
Chai,$18.00 ,39,
Chang,,17,
Chef Anton's Cajun Seasoning,$22.00 ,53,No
,,,
Chef Anton's Gumbo Mix,$21.35 ,0,No
---------------------------------------
Exported by SpreadProcessing:
---------------------------------------
Product,Unit Price,Units in Stock,Discontinued
Chai,$18.00 ,39
Chang,,17
Chef Anton's Cajun Seasoning,$22.00 ,53,No

Chef Anton's Gumbo Mix,$21.35 ,0,No
---------------------------------------
Note the extra commas in the empty row and in the end of the first two rows in the SpreadProcessing sample.