Eric Blake wrote: > On 10/01/2010 02:44 PM, Jim Meyering wrote: >> Eric Blake wrote: >>> * src/stat.c (main): Hoist default format computation out of loop. > >>> This one should be uncontroversial, but just to be safe: OK to push? >> >> Yes. >> Thank you for the factoring. >> Did you do any before/after testing, to be sure that nothing changed? > > 'make check' had the same results before and after (but given the > level of stat(1) testing in make, I wouldn't count too much on > that...). But more specifically: > > for opts in '' -t -f -ft; do > cmp <( stat-pre $opts . /dev/null) \ > <(LC_ALL=C src/stat $opts . /dev/null) > done > > showed no difference, and covers all the code paths of
Perfect. > default_format(). (The LC_ALL=C is a bit of overkill until someone > actually translates the string).
