Hi Junio, On Tue, 22 Mar 2016, Junio C Hamano wrote: > Johannes Schindelin <johannes.schindelin@xxxxxx> writes: > > > One of this developer's primary tools to diagnose broken regression > > tests is to run the test script using 'sh -x t... -i -v' to find out > > *which* call *actually* demonstrates the symptom. > > > > Hence it is pretty counterproductive if the test script behaves > > differently when being run via 'sh -x', in particular when using > > test_cmp or test_i18ncmp on redirected stderr. > > > > So let's use grep instead of test_cmp/test_i18ncmp to verify that stderr > > looks as expected. > > In the modern world, I would probably described the problem as > "tXXXX -i -v -x", though, not "sh -x tXXXX", but they both exhibit > the same symptom. Thanks for pointing me to -i -v -x. The introduction of the RHS -x somehow slipped by me. > I wonder if "tXXXX -i -v -x" can be made not to contaminate the > standard error stream of the test, but that would be a larger change > we probably would not have time for 2.8 final anyway. Agree, both on "we want to have this" and "after 2.8". Ciao, Dscho -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html