Even better, how about having -append- return the number of common
variables, and the number of unique variables from the master and using
datasets? Then one could automate the checking of whether it worked...
Nick W
> -----Original Message-----
> From: baum [mailto:[email protected]]
> Sent: Thursday, July 18, 2002 4:47 PM
> To: StataList
> Subject: st: append and strange doings
>
>
> It seems like -append- ought to be smart enough to at least
> give a warning
> if you try to append a dataset with NO variables in common
> with the master
> dataset. I just run into this helping a colleague who has a
> dozen Census
> extract files, produced elsewhere and probably used by a number of
> researchers. He was appending them in a -for- loop. _One_ of
> those files,
> it turns out, has UPPER CASE FILENAMES. The rest have all the same
> variables with lower case filenames. Yes, once you figure that out,
> "renvars _all, lower" (STB) does a very nice job of cleaning
> up this mess,
> but the outcome of appending 85 variables, none of which are
> in common with
> the 85 variables in memory, is to double the memory
> requirements---so you
> get a r(901), and eonder why that should possibly be
> happening when the
> size of the data is not increasing by that much.
>
> Might -append- reasonably give you a warning that appending
> without any
> common contents between the master and using datasets is
> surely likely to
> be a logical error?
>
> Kit
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
