Junio C Hamano <gitster@xxxxxxxxx> wrote: > Eric Wong <e@xxxxxxxxx> writes: > > >> Yes, using Perl is a good substitute for writing it in C in this > >> case. I however question the choice to use t9700/test.pl here, > >> which is clearly stated that its purpose is to "test perl interface > >> which is Git.pm", and added tests are not testing anything in Git.pm > >> at all. > >> > >> Using t9700/test.pl only because it happens to use "perl -MTest::More" > >> sounds a bit eh, suboptimal. > > > > *shrug* I figure Test::More is common enough since it's part of > > the Perl standard library; but I consider Perl a better scripting > > language than sh by far and wish our whole test suite were Perl :> > > Oh, I think we (actually the author of t9700) considers it common > enough that we have PERL_TEST_MORE prerequisite to allow us to write > tests, assuming that it is available, and let us easily skip where > it is not available. So I do not think I mind the dependency on > Test::More at all. Moving the tests to t1006 and rewriting the > tests not to use Test::More are two separate and unrelated things, > and if you are more comfortable with Test::More (and more > importantly if it is natural to write Perl based tests using > Test::More), it is not necessary to switch away from it. OK, fair enough. Given t1006 is mostly sh, I prefer keeping v2 as-is because the Test::More->builder munging of test numbers in t9700/test.pl is nasty too and I wouldn't enjoy duplicating those bits in a hypothetical t1006/test.pl, either. It would be nice to have first class support for Test::More in our suite so we could just have t/t0006-cat-file.t and t/t9700-perl-git.t implemented in Perl without sh at all, but that's a separate discussion.
