[email protected]
> I have some questions about macros.
>
> 1/ I wonder how to count the number of variables in a macro.
> I try the macro extended function (word count) but it doesn't work.
> For example, if I have:
> local myvars "var1 var2 var3 var4"
> then `myvars' contains 4 variables.
> But I do not know how to create a macro that contains the value 4.
I go
. local myvars "var1 var2 var3 var4"
. local nv : word count `myvars'
. di `nv'
4
and Stata tells me 4. (The macro contains not four
variables, but the names of four variables.)
What are you doing wrong? I have to guess.
Perhaps you are doing something like
. local nv : word count "`myvars'"
. di `nv'
1
Here you must remember Stata's definition of
a word. Stata by default parses on white space meaning
that
Amadou loves Stata macros
would be considered by Stata to be 4 words. So far,
so good. But Stata has another rule: double quotes
bind tighter than white space separates, so
"Amadou loves Stata macros"
is just _one_ word. This is just Stata's sense of
a word; forget here what you know about the grammar of
any ordinary language. Now, when you defined
. local myvars "var1 var2 var3 var4"
the " " were just ornamental; Stata took
them as delimiters, not as part of your
definition of the local myvars (and you
could have omitted them). But if you go
. local nv : word count "`myvars'"
the " " are not ornamental at all; they
are working hard to bind what follows
-: word count- into one word.
> 2/ Suppose several datasets containing the same variable floor indexed
> 1, 2, 3, etc.... Floor* number varies with datasets.
> I want to create the macro myfloor
> local myfloor "floor*"
> but then, I do not know how to create the macro containing
> the number of floor
> as in question 1.
This question was asked earlier today. As said then,
the number of variables is not often an issue. Nevertheless,
unab floor : floor*
local nfloor : word count `floor'
>
> 3/ How to work "inside" macros?
> suppose you have the following: local "var1 var2 var3 .....varN"
> suppose later in your dofile you want to do some tables on
> only - say 20-
> variables of the macro.
> Is it possible to just choose them (perhaps by specifying
> their position IN
> the macro)
> without creating other (sub)macros?
The term submacro is yours; I don't know what you mean by it.
There are several ways of doing this, but they usually
involve creating other macros.
>
> 4/ Finally my last question is how to merge/append several datasets?
> Suppose you have :
> local mydata "data1 data2 data3 data4 data5"
> foreach if i of local mydata {
> foreach j of local mydata {
> append `i' using `j'
> }
> }
>
> How to avoid data1 being appended twice?
Various ways. Here's one.
Start with the second word in the macro.
local mydata "data1 data2 data3 data4 data5"
local nsets : word count `mydata'
tokenize `mydata'
use `1'
forval i = 2/`nsets' {
append using ``i''
}
*
* 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/
