Re: unbuffered stdout

[Jaroslav Hajek]

On Sat, Dec 26, 2009 at 11:36 PM, John W. Eaton <address@hidden> wrote:
> On 26-Dec-2009, Jaroslav Hajek wrote:
>
> | based on several previous questions by users, as well as my own needs,
> | I implemented an easier way to ensure
> | unpaged output to stdout: there will be stdout_nobuf, which always
> | bypasses the pager.
>
> Isn't that what stderr is for?
>

Yes, I normally use it for that purpose. But there may be some
drawbacks of writing to stderr instead of stdout.

> | However, since stdout_nobuf gets the number 3, Octave's file numbers
> | will no longer be exactly equal to OS descriptors.
>
> I'm not sure and I don't have time to dig in to the details at the
> moment, but I think that could cause trouble.
>
> | OK to push?
>
> I'd prefer to not make this change.  It seems better to me to simply
> turn the pager off if you don't want it.
>

OK, I'll put this off for the time being. Turning pager seems simple
from an user's perspective, but it's a bit cumbersome to do in
functions, because it typically requires a big unwind_protect block
for the whole body, in each function that merely writes some
informational messages.

regards

-- 
RNDr. Jaroslav Hajek, PhD
computing expert & GNU Octave developer
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz