Three quite different comments and an alternative.
1. If you do it this way, you don't need so many braces.
local j "0"
foreach f in `r(files)' {
if `j' == 0 use `f'
else append using `f'
local ++j
}
2. As you want the numeric interpretation, I find
local j = 0
clearer, particularly as you are testing its
numeric value later. Stata doesn't care, but
a code reader might.
3. If you are a Windows user and accustomed
to putting spaces in your filenames, this will
fall over unless the names come quoted.
local j = 0
foreach f in `r(files)' {
if `j' == 0 use `"`f'"'
else append using `"`f'"'
local ++j
}
should be safe either way.
I feel queasy about going through two loops
simultaneously unless it's unavoidable.
That's mostly just a style prejudice.
It is avoidable here, although this
alternative is less scrutable, so to
speak:
tokenize `r(files)'
use `"`1'"'
mac shift
local files `"`*'"'
foreach f of local files {
append using `"`f'"'
}
I should guess that there is not much
efficiency difference for normal problem
sizes.
Nick
[email protected]
[email protected]
> Ulrich Kohler already supplied an interesting solution. I'd
> like to show how I usually handle this sort of situation
> (which does come up often -- with either -append- or
> -merge-). This might be more of an old-fashioned programmer's method.
>
> local j "0"
> foreach f in `r(files)' {
> if `j' == 0 {
> use `f'
> }
> else {
> append using `f'
> }
> local ++j
> }
>
> One can be more clever by putting the ++j (or j++) inside the
> -if- statement, if you prefer.
> I haven't yet tried that technique; it might go
> if `j++' ==0
> (if I have the syntax right -- or is it `=j++' ... ?) or even
> if ~`j++'
*
* 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/
