Mark,
<warning, shooting from the hip & possibly making it more complex than
necessary>
I'm not sure if the disabling extended wait works for the writer. I
*think* it does.
If not, you could always call FCONTROL 4 with a timeout value before the
fwrite. This will wait however many seconds you specify for the fwrite the
complete. If it doesn't complete within that time, the fwrite will fail with
an error value of 22 (software time-out).
I usually test this stuff with FCOPY when playing with message files, but
it's tough to call FCONTROL...
Sorry I can't be more sure (don't have time to hack today).
RAT
ps. You might check the Interprocess Communications (IPC) Programmer's
Guide..I think it should be on the web at http://docs.hp.com.
<http://docs.hp.com>
<mailto:[log in to unmask]> Rich Trapp
Consulting for Agilent Technologies, Loveland, Colorado.
<http://www.mbsnav.com/> Managed Business Solutions
200 South College Avenue
Fort Collins, Colorado 80524-2811
970.679.2221 (voice)
970.669.3071 (fax)
-----Original Message-----
From: Mark Boyd [mailto:[log in to unmask]]
Sent: Wednesday, December 20, 2000 2:04 PM
To: 'TRAPP,RICH (Non-A-Loveland,ex1)'; [log in to unmask]
Subject: RE: fwrite to message file. Help! Cecil! Help!
the fine print reads:
if a message file is full and there a no reader, the process waits if there
is an fcontrol=45 in effect or if this is the first fwrite after an
fopen/hpfopen call. Otherwise, ccg(0) is returned.
...so if the file is full and there is a reader, the default is to wait
unless there is an fcontrol 45 in effect?
-----Original Message-----
From: TRAPP,RICH (Non-A-Loveland,ex1)
[ mailto:[log in to unmask]
<mailto:[log in to unmask]> ]
Sent: Wednesday, December 20, 2000 12:56 PM
To: 'Mark Boyd'; [log in to unmask]
Subject: RE: fwrite to message file. Help! Cecil! Help!
Mark,
I think it's FCONTROL(fnum, 45, ZERO) on the file before the fwrite.
Read the fine print in the manual...this is from memory (an the quick
reference guide!).
RAT
The default behavior for MESSAGE files is to wait.
Rich Trapp < mailto:[log in to unmask] <mailto:[log in to unmask]> >
Consulting for Agilent Technologies, Loveland, Colorado.
Managed Business Solutions < http://www.mbsnav.com/ <http://www.mbsnav.com/>
>
200 South College Avenue
Fort Collins, Colorado 80524-2811
970.679.2221 (voice)
970.669.3071 (fax)
-----Original Message-----
From: Mark Boyd [ mailto:[log in to unmask] <mailto:[log in to unmask]> ]
Sent: Wednesday, December 20, 2000 1:17 PM
To: [log in to unmask]
Subject: fwrite to message file. Help! Cecil! Help!
When writing to a message file (variable length records, one reader) that
has reached eof, the fwrite waits. Is there a way to make the fwrite fail
insted of waiting?
Thank you.
|
