Re: st: RE: Stata thinks files are read-only when they shouldn't be!

["Eric A. Booth" <ebooth@ppri.tamu.edu>]

Two things come to mind, either (1) your computer/OS is capriciously  
setting your file with a read-only attribute (as Nick points out, this  
can happen with Windows XP) or (2) when Stata tells the OS to write  
the file to myfile_happ.txt, the OS is moving too slowly, Stata moves  
on to the next loop and tries to write again, but the file is read- 
only because it is still being written to by the OS  (or BOTH (1) and  
(2) are happening).

The solution to (1) is to "force" xp to apply a read-write, rather  
than a read-only, attribute to your file using the -winexec attrib-  
command and the solution to (2) is to tell Stata to slow down in your  
loop so that it isn't outpacing your OS by using the -sleep- command.

So your new loop might look like:


foreach y of varlist happ job  {
	regress `y' lrwage lrwager firm_l if female==1, cluster(unm)
estimates store Self0
	regress `y' lrwage lrwager firm_l firm_iqr if female==1,
cluster(unm)	estimates store Self1
       ...
       outreg2 lrwage lrwager firm_l firm_iqr [Self0 Self1 ...]  ///
	using myfile_`y', replace bdec(2) label nocons excel
	sleep 8000
	winexec attrib -R "c:\yourpath\yourfolder\myfile_`y'.txt"    /S     /D
}


Depending on your version and SP of Windows XP you may need to modify  
the -winexec attrib- command above.  Generally, the DOS command  
follows the format:

ATTRIB [+R | -R] [+A | -A ] [+S | -S] [+H | -H] [[drive:] [path]  
filename] [/S [/D]]

but you should open DOS and try "attrib /?" or "help attrib" (? I  
can't remember what the help command in Windows XP DOS, it's been a  
while) to see what format it asks for.  The "-R" part is telling the  
computer to remove any read-only attribute of the file.

The combination of the sleep command to ask Stata to wait up a bit and  
the attrib command to force the OS to make sure the file is read-only  
before it loops again should do the trick.

Good luck.

Eric

__
Eric A. Booth
Public Policy Research Institute
Texas A&M University
ebooth@ppri.tamu.edu
Office: +979.845.6754
Fax: +979.845.0249



On May 10, 2009, at 9:44 AM, Nick Cox wrote:

> -outreg2- is a user-written program (Roy Wada, SSC).
>
> Broadly speaking, my guess is that the explanation is nothing to do  
> with
> either the official Stata commands you use or -outreg2-. It will be a
> matter of how that directory is set up within Windows. I know this
> contradicts what you say, but it's still my guess, because I can't  
> think
> of a better one.
>
> So I'd seek an explanation the operating system end.
>
> I've known Windows decide -- utterly capriciously, from my point  of
> view, although presumably it was a consequence of something I did --
> that a particular directory was read-only.
>
> The fact that you report that this "sometimes" happens seems  
> consistent
> with that guess.
>
> Conversely, if this were a bug in the commands mentioned, all heavily
> used, it would have been reported long since.
>
> Nick
> n.j.cox@durham.ac.uk
>
> kokootchke
>
> Sometimes when I run do files in which I try to save, replace or  
> modify
> a file in some way, Stata thinks that this file is read-only when it
> shouldn't be.
> The problem I'm having right now is the following:
> I am running a do file that contains a bunch of lines like:
>
>
> foreach y of varlist happ job{
> 	regress `y' lrwage lrwager firm_l if female==1, cluster(unm)
> estimates store Self0
> 	regress `y' lrwage lrwager firm_l firm_iqr if female==1,
> cluster(unm)	estimates store Self1
>        ...
>        outreg2 lrwage lrwager firm_l firm_iqr [Self0 Self1 ...] using
> myfile_`y', replace bdec(2) label nocons excel}
>
>
> When I do the file, I notice that Stata shows in the display window  
> the
> results for all the regressions but when it reaches the outreg2  
> command,
> it gives the following error:
>
> file myfile_happ.txt is read-only; cannot be modified or erasedr(608);
>
> And I get this error whether myfile_happ.txt exists in the directory  
> or
> not. Also, sometimes I only get this error for the SECOND time Stata
> goes through the loop; that is, the first time it goes through and it
> does save the file myfile_happ.txt but the second time it stops and it
> fails to save the file myfile_job.txt!!! What do you think is going  
> on??
>
> Another thing that happens often -- but I'm not having this problem
> right now -- is that I run a do file that contains a line such as:
> save, replace
> and it fails to save the file because it is apparently read-only. What
> is even stranger is that some other times, it does go through and  
> saves
> the file normally.
> I have already verified that my folder where my files are in Windows  
> is
> not read-only, but that doesn't help.
> I am running Stata 9 SE on Windows XP Pro... but I think I have also
> seen this error when I've run my code on a different computer with  
> Stata
> 10 SE.
>
>
> *
> *   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/

*
*   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/