Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Wishlist for Stata 13 |
Date | Mon, 28 Jan 2013 21:40:01 +0000 |
David and others on versions before the present can always access basic help for the current version on-line: http://www.stata.com/help.cgi?confirm The links at the foot of every Statalist posting give another example. In this case I would try -quietly describe using <filename>- and see if it's a Stata data file that you can read. If it's not, there will be an error message. Nick On Mon, Jan 28, 2013 at 9:32 PM, David Kantor <kantor.d@att.net> wrote: > I have a wish, based on Stata 10. So I don't know if this has been taken > care of in Stata 11 or 12. > > When using -confirm new file- or -confirm file-, it would be good if the > file name could (optionally) be checked for the possibility of it being a > .dta file. > That is, suppose the filename is in a macro "filename", and my code looks > like... > > confirm new file `filename' > ... > save `filename' > > Most of the time, I should really check... > confirm new file `filename'.dta > --as Stata interprets `filename' as `filename'.dta when saving it as a > dataset. Thus, if I test only > confirm new file `filename' > and `filename' is "abc", and the file abc.dta exists, but abc (no dot) does > not, then the -confirm- will pass, but the -save- will fail. > > The solution would seem to be to test... > confirm new file `filename'.dta > But it is not always correct, either, as there is an exception: if the name > already has a "." in it, then -save- does not append ".dta". > So the robust way to code this, as I see it, is... > if strpos("`filename'", ".") { > confirm new file `filename' > } > else { > confirm new file `filename'.dta > } > > or the equivalent one-liner: > confirm new file `filename'`=cond(strpos("`filename'", "."), "", ".dta")' > > But I don't want to write all of that every time. It would be nice if that > were built-in to -confirm-. Perhaps it would be a new syntax, such as > -confirm new dta file- filename. > Has anyone else had this issue? * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/