Re: st: Programming sub-routines

[Nick Winter <nw53@cornell.edu>]

You should read up on saving results in [U] 21.10.

The short version is that for returning value(s), usually you do an rclass 
program:

*------------beginning of mainprog.ado-------------------
program mainprog

...
    subprog
    ... commands using r(whatever) , r(whichever) , etc....

...
end

program subprog, rclass
   ...

   return scalar whatever = ...
   return matrix whichever = ...
   ...
end
*------------end of mainprog.ado-------------------


For returning variables, you usually can send the name of the variable to 
be created or replaced as an argument to the subroutine:

*------------beginning of mainprog.ado-------------------
program mainprog

...
    subprog `myvar'
   ...

end

program subprog
   args variable

   ...

   generate `variable' = ...

   ...

end
*------------end of mainprog.ado-------------------


Hope this helps
--Nick Winter


At 07:46 AM 11/4/2004 +1100, you wrote:

> Dear Statalisters,
>
> Before coming to Stata my programming was honed using Pascal and I'm
> missing some of the tricks I used to access there. I would appreciate
> any advice on a particular problem I have with Stata sub-routines.
>
> I'm used to being able to modularise my programs (in Pascal) by having
> lots of sub-routines for repetitive tasks, particularly functions
> which carry out some kind of transformation. Some old database
> languages (such as DBase) used to allow "user-defined" functions
> for similar tasks.
>
> I've tried to find the equivalent in Stata, but to no avail. If I use
> a sub-routine inside a main program I can pass it arguments no
> problem. But after doing some transformations on the arguments, how can I
> send the results back to the main program? If I declare new local
> macros to store the results, they disappear when I return to the main
> program (because they're no longer in scope). If I declare global
> macros, I can access the results back in the main program, but I'm
> left with a lot of global macros. This appears to me bad practice, and
> also requires lots of dropping of global macros on exit etc.
>
> Am I missing something obvious? In Pascal terms arguments could be
> "val" or "var" (the latter allowing for a return value). But in Stata all
> arguments appear to only be "val". Is this right? Do I have any other
> options? Any advice would be most appreciated.
>
>
>
> --
> Kind regards,
>  Ian
>
> =================================
> Ian Watson
> Senior Researcher
> acirrt
> University of Sydney
> NSW 2006
> Australia
>
> Phone: 02 9351 5622
>   Fax: 02 9351 5615
>
> Email: i.watson@econ.usyd.edu.au
>   Web: www.acirrt.com
> ================================
>
> *
> *   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/
--------------------------------------------------------
Nicholas Winter                           607.255.8819 t
Assistant Professor                       607.255.4530 f
Department of Government              nw53@cornell.edu e
308 White Hall            falcon.arts.cornell.edu/nw53 w
Cornell University
Ithaca, NY 14853-4601 

*
*   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/