Re: st: save9 and c(changed)=1 status

[Nick Cox <njcoxstata@gmail.com>]

I suspect that you can't and this is a feature. Handing users a way to
subvert c(changed) sounds like distributing guns to me.

I can't see that this is a misfeature of -saveold- either. The fact
that you have changed your dataset but only -save-d it in a old format
could be a big disadvantage, so Stata should prompt you.

The way it is usually is that you are using a newer Stata and you only
want to -save, old- to send to somebody else with an older Stata, or
for your purposes if you have an older Stata on a different machine.
You should still be prompted about saving your main dataset.

In addition, if you don't care about saving what you just did, you
just go -exit, clear-.

(Like Joerg Luedicke yesterday, I am interested in what advantage you
think -save9- has over -saveold-. I certainly don't want to use it if
you intend to implement non-standard behaviour making key decisions on
behalf of users!)

Nick

On Thu, Apr 14, 2011 at 12:41 PM, Marco Ercolani
<m.g.ercolani@bham.ac.uk> wrote:
> Dear Statalisters,
>
> How do I change Stata's creturn status from:
>  c(changed)=1
> to
>  c(changed)=0
> without actually saving the dataset currently in memory?
>
>
> (I ask because I am refining my save9 ado script and when
> Stata 10 or 11 saves the dataset in the Stata 9 format the
> status remains at c(changed)=1 and this means that attempts
> to exit after "save9" produces:
>
> . save9 MyData
> This is Stata 11.2 software:
> file MyData.dta saved
>
> . exit
> no; data in memory would be lost
>
> The same problem exists when using Stata's "saveold" command.)
>

*
*   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/