Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Sebastian Eppner <eppner@uni-potsdam.de> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Postfile "forgets" the postfile |
Date | Thu, 24 Feb 2011 02:28:05 +0100 |
...the typing error (using y instead of x) is not responsible... i just saw it in the post, but in my do-file, i surely used the exact same name all the times i referred to the postfile (i also changed it several times, so its not because of some specially "bad" name). thx for helping me out! S On Thu, Feb 24, 2011 at 2:21 AM, Sebastian Eppner <eppner@uni-potsdam.de> wrote: > Hi, > > at some point in my code i define a postfile, in which i want to write > 4 variables, say one two three four. > > . postfile x one two three four using somedataset, replace > > then open a forval loop and after some calculations i want to add new > observations to the postfile, using the code: > > . post y (one) (two) (three) (four) > > unfortunately, at this point stata tells me: > post x not found > r(111); > > So apparently, Stata lost the postfile definition along the way somehow... > > I could track down the exact point at which Stata lost the > information, which is a selfwritten ado. > So the problem lies probably in the ado. I suspect the preserve / > restore / clear combination i use in there. I am not an experienced > ado writer, so please help me. the simplified structure of the ado is > like this: > > program tuples2, rclass > version 9 > syntax varlist(min=1 max=1 numeric) [if] [in] > .... > preserve > clear > > set obs2 > .....do some stuff > .....write the results i am interested in into return macros > restore > end > > > Could this "clear" be responsible for the damage? If yes, is there > another way I can "start from scratch" in the ado file (and in the end > restore the damage done) while NOT hurting the postfile definition? > > Thanks for a little help. > Sebastian > > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/statalist/faq > * http://www.ats.ucla.edu/stat/stata/ > * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/