Re: [HACKERS] pg_upgrade and system() return value

2013-01-21 Thread Bruce Momjian
On Sun, Jan 20, 2013 at 11:14:47AM -0500, Tom Lane wrote: Bruce Momjian br...@momjian.us writes: Can someone comment on the attached patch? pg_upgrade was testing if system() returned a non-zero value, while I am thinking I should be adjusting system()'s return value with WEXITSTATUS().

Re: [HACKERS] pg_upgrade and system() return value

2013-01-20 Thread Tom Lane
Bruce Momjian br...@momjian.us writes: Can someone comment on the attached patch? pg_upgrade was testing if system() returned a non-zero value, while I am thinking I should be adjusting system()'s return value with WEXITSTATUS(). AFAIK it's not very good style to test the result as an

[HACKERS] pg_upgrade and system() return value

2013-01-19 Thread Bruce Momjian
Can someone comment on the attached patch? pg_upgrade was testing if system() returned a non-zero value, while I am thinking I should be adjusting system()'s return value with WEXITSTATUS(). Is there any possible bug in back branches just compariing system()'s turn value to non-zero without