RE: st: continuing a loop from after it encounters an error

[Nick Cox <n.j.cox@durham.ac.uk>]

I'd still guess that you can and should filter out variables that might exist but don't before you do anything with them. Otherwise you'd need to test on every operation whether something existed or not. 

Nick 
n.j.cox@durham.ac.uk 


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Mosely
Sent: 03 February 2011 19:53
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: continuing a loop from after it encounters an error

Thanks, -noisily capture- is what I was looking for.

It's complicated, I have an access database which has descriptions of
my variables. I'm trying to use a spreadsheet made in access to create
value labels and label variables. To do this, I am creating value
labels and variable labels in one instance of stata and then
transferring them to the master database. Some of the variables in
access are not in the master database.

On Thu, Feb 3, 2011 at 11:45 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
> My answer is the same, extended.
>
> 1. Use -capture- if you must do this.
>
> 2. I can't see why you would ever need to do anything like this. Just loop over a varlist. The globals are just middlemen here who can be cut.
>
> Nick
> n.j.cox@durham.ac.uk
>
> Nick Mosely
>
> OK, youre right, my example was weak. Here is a more accurate look at
> what I am trying to accomplish:
>
> set obs 1
> g object1 = 1
> g object3 = 3
> global p1 object1
> global p2 object2
> global p3 object3
> foreach i of numlist 1/3 {
>        foreach j of global p`i' {
>                tabulate `j'
>        }
> }
>
>    object1 |      Freq.     Percent        Cum.
> ------------+-----------------------------------
>          1 |          1      100.00      100.00
> ------------+-----------------------------------
>      Total |          1      100.00
> variable object2 not found
> r(111);
>
> I hope this makes it more clear. Thanks!
>
> On Thu, Feb 3, 2011 at 10:54 AM, Eric Booth <ebooth@ppri.tamu.edu> wrote:
>> <>
>>
>> I'm still not clear on what the "objects" are.  You mention that are "commands" and then you -display- them. Also, when I run your code, I don't get the r(199) error about it being an unrecognized command -- so there's something important missing from your example.
>>
>> Take a look at -help confirm- and decide which objects you are trying to confirm exist.
>> If these are variables you can:
>>
>> **************************!
>> clear
>>  set obs 10
>>  g object1 = 1
>>  g object3 = 3
>>
>> ****
>> foreach i in object1 object2 object3 {
>> cap confirm variable `i'
>>   if !_rc {
>>        display `i'
>>        }
>> }
>> **************************!
>>
>> If these are commands, you can also use -which- and -capture- to check to see if they are installed:
>> **************************!
>> clear
>> foreach i in  describe object1 object3 list  {
>>        cap which `i'
>>        if _rc continue
>> }
>> **************************!
>>
>> On Feb 3, 2011, at 12:41 PM, Nick Mosely wrote:
>
>>> I've built a loop that encounters an error periodically due to the
>>> fact that stata can't find one of the objects specified in the list. I
>>> don't mind the error, but I would like stata to skip objects in list
>>> that encounter the error and continue to the next iteration of the
>>> loop, rather than breaking the loop. I've tried the continue command,
>>> but am not sure that is what it's meant for.
>>>
>>> Here is a simplified version of the code. In this example, assume that
>>> objects1 and object3 are valid commands but object2 is not:
>>>
>>> foreach i in object1 object2 object3 {
>>> display `i'
>>> continue
>>> }
>>>> object1
>>>> unrecognized command:  object2
>>>> r(199);
>>>
>>> What can I do get the code to continue running through object3?

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/