On 25/03/2008, Ronan Klyne <[EMAIL PROTECTED]> wrote: > sebb wrote: > > At present, if the CSV Dataset element specifies more variable names > > than there are columns, then the additional variable names are > > ignored. If there is already a variable with one of the unused names, > > it will be left unchanged. > > > > Would it might be useful to enhance CSV Dataset to process all the > > variable names? > > If so, what should be done? For example: > > - set unused variables to the empty string? > > - set unused variables to some default value? > > - remove unused variables (i.e. delete any existing definition)? > > > > [The GUI would need to be updated to add a new option, with the > > default being the existing behaviour] > > > > Anyone have any views on this? > > > My 1.00821697 pence: > > For my purposes, the current behaviour is best. > If you make a change, I'd urge against delete any existing definitions - > If the CSV file and column name set don't even match up, the least we > can do is not tread on any other code... >
Any change would definitely be optional. Also, only the variable names listed on the CSV Dataset GUI would be affected. For example, if the GUI lists 3 variables a,b,c then any rows in the file with 3 or more columns will set the variables a,b and c as before. The question was what to do about rows with fewer than 3 columns. If there are only 2 columns, what should happen to variable "c" ? At present it is ignored: if it already exists, it will be left unchanged; if it does not exist it won't be created. > # r > > -- > Ronan Klyne > Business Collaborator Developer > Tel: +44 01189 028518 > [EMAIL PROTECTED] > www.groupbc.com > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
