[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: ado file
Nick Cox <email@example.com>
Re: st: ado file
Mon, 13 Aug 2012 17:13:52 +0100
In your case, what is important is the difference between the program
and the .ado file that defines it.
1. First time you type -dd-, Stata looks for a command -dd- that it
knows about, doesn't find one, but does find an .ado file -dd.ado-. So
it runs -dd.ado- and what happens is that
Data are -clear-ed.
Program -dd- is defined to Stata.
2. Second time you type -dd, Stata looks for a command -dd- that it
knows about, but this time it does find one. You just defined -dd- by
a series of commands starting with -program- and ending with -end-.
But this time there is no -clear-.
There are at least two takes on what you could do. One is that you
could revise your program -dd- so that -clear- is included in the
Another is that this is an example of premature programming, defined
as writing a program that isn't needed. There is almost no generality
in your program and it might as well be a do-file with arguments. But
then again there is no harm done either.
On Mon, Aug 13, 2012 at 4:48 PM, tashi lama <firstname.lastname@example.org> wrote:
> I am writing a simple ado-file where I am generating a dataset using sql query and listing the variable using syntax command. Here is my code
> program dd
> qui odbc query indigo_dev71
> odbc load, exec("select date(read_date) as date, count(*) as hits from readership where source_id=6429 and date(read_date) between '2011-01-01' and '2011-01-20' group by date(read_date)")
> syntax [varlist]
> list `varlist'
> when I execute ado file .dd date hits
> date hits |
> 1. | 01jan2011 127 |
> 2. | 02jan2011 315 |
> 3. | 03jan2011 4438 |
> 4. | 04jan2011 3978 |
> 5. | 05jan2011 3896 |
> 6. | 06jan2011 4981 |
> 7. | 07jan2011 4310 |
> 8. | 08jan2011 540 |
> 9. | 09jan2011 567 |
> 10. | 10jan2011 5572 |
> 11. | 11jan2011 4261 |
> 12. | 12jan2011 4184 |
> 13. | 13jan2011 4218 |
> 14. | 14jan2011 4894 |
> 15. | 15jan2011 1113 |
> 16. | 16jan2011 343 |
> 17. | 17jan2011 1154 |
> 18. | 18jan2011 3981 |
> 19. | 19jan2011 5669 |
> 20. | 20jan2011 3685 |
> Great. But when I run second time .dd date
> Stata returns you must start with an empty dataset. I have a "clear" command at the top of my adofile to start with a fresh memory. Could someone please explain? Also, does adofile name have to be identical to the program name? Meaning, my program name here is dd. Do I have to have a adofile dd.ado? If they could be different, how do I execute?
* For searches and help try: