No the file was not open when I ran the process in debug.
I also checked for any locks using;
call qp0fptos (*lstobjref '/tmp/httpapi_debugSMS.txt' *format2)
And the results of this confirmed there were no locks.
The lack of any error is the confusing issue.
Don Brown
From: "Kevin Wright" <kevin.wright@xxxxxxxxxxxx>
To: "Midrange Systems Technical Discussion"
<midrange-l@xxxxxxxxxxxxxxxxxx>
Date: 02/12/2019 08:42 AM
Subject: RE: httpapi debug file appending not replacing
Sent by: "MIDRANGE-L" <midrange-l-bounces@xxxxxxxxxxxxxxxxxx>
Schroedinger's file? Did you have it open using anything when the unlink
didn't work in your software, but didn't have it open when the other
unlink did work?
-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of Don
Brown via MIDRANGE-L
Sent: Monday, 2 December 2019 9:37 AM
To: Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxxxxxxxx>
Cc: Don Brown <DBrown@xxxxxxxxxx>
Subject: Re: httpapi debug file appending not replacing
I just ran the process in debug and verified the unlink process was being
executed and the path/filename were correct but the file still existed
after that.
Since then I have now run from IFSEBOOK CH3RENDEL which did successfully
delete the file and it uses exactly the same method.
I am really confused now.
In httpapi the file name is trimmed into the variable passed to unlink eg
eval HTTP_DEBUG_FILE = %trimr(peFilename)
callp unlink(HTTP_DEBUG_FILE)
Whereas in CH3RENDEL
unlink(%trimr(path)) < 0
But that should not make any difference ?
I am obviously missing something :-(
Thanks
Don Brown
with our affiliate link:
https://amazon.midrange.com
midrange.com
