From: James E. F. <jf...@ac...> - 2002-03-26 15:33:05
|
On Tue, 26 Mar 2002, angek productions wrote: > I'd like to see this process check to see if the csv > file exists and if it does, not to overwrite it - as is > the current case- but rather check the files size And > if different create another copy of it with the current > date appended to the end of the filename (or any other > way where the csv files from a particular Survey can be > distinquished from one another). > > So for example, assume that many of the surveys I > create for my company run for a year. Each quarter the > results are downloaded and analysed / compared to the > previous Quarter's results. In this instance I would > have 4 csv files on the server belonging to the same > survey named quartely_survey_310302, > quartely_survey_300602, Quartely_survey_300902 and > quartely_survey_311202. > > Guys your thoughts please. I would suggest for this case that at the end of each quarter, you end the survey, copy it, and activate the new copy. If you are doing data analysis in a separate program, then you should have no problem appending one quarters data to the rest. This way you would have multiple surveys named something like: quarterly_survey_q1_02 quarterly_survey_q2_02 quarterly_survey_q3_02 quarterly_survey_q4_02 quarterly_survey_q1_03 ... However, it may still be a good idea to not overwrite existing data files. Perhaps a confirmation screen, or a screen with a box to enter an alternate filename, would be a nice feature. -James |