Bob Proulx wrote:
> Chet Ramey wrote:
>> No, there's no way to wait for it.
>
> Hmm... That does make it very difficult to actually use correctly.
Not really, if you think about the intended use. The original intent
was to allow applications to read and write to and from processes
in cases where pipes were inconvenient or impossible. The vast
majority of the cases result in the caller reading all of the output
generated by the process, or the process consuming all of the data
sent by the caller. Most of the rest of the cases should result in
the asynchronous process receiving SIGPIPE/EPIPE, depending on process
scheduling and kernel behavior.
Yours is the first complaint about this particular implementation detail.
I suspect that most of the use cases fit the above paragraph. It doesn't
appear to be a high-priority problem.
Chet
--
``The lyf so short, the craft so long to lerne.'' - Chaucer
Live Strong. No day but today.
Chet Ramey, ITS, CWRU [EMAIL PROTECTED] http://cnswww.cns.cwru.edu/~chet/
