Thanks, @prakashcv13. I will update the EDD to reflect the same, to use qstat -fw instead of qstat -f1 option to print the attributes without wrapping them.
I don’t see benefit in adding a new --wrap option. If we introduced --wrap and/or --no-wrap and mapped them to the use of -w it would confuse the meaning of the existing -w, which means “w as in wide”. Since “w as in wide” makes sense in the -fw case we are discussing, so I’d like to see us just go with that.
to me wrap/no-wrap is more intuitive as if we wrap the value or we or we dont. However, it is your call.
One thing to point out is that getopt_long() is not POSIX. It’s a GNU extension. Now that we only support linux, this might be OK. We should make this decision knowing the consequences of not being able to go back and supporting other OSes.
We should add the following line to the spec file if we rely on GNU extensions like getopt_long()…
Requires: glibc >= 2.2
Specifically, is this extension available for Windows?
That is a very good question, Ian. Everything I found online says even getopt() isn’t supported on windows. I checked our source code and indeed we have our own version of getopt() for windows. If we were to use getopt_long(), we’d need a version of that as well.
Bhroam
The EDD link has been updated: https://pbspro.atlassian.net/wiki/display/PD/PP-483%3A+PP-484%3A+PP-485%3A+Enhance+output+of+qstat+to+make+it+more+admin+and+script+friendly
Hi Nithin,
Changes to EDD looks good to me.
Thanks