On Thu, Dec 1, 2022 at 8:02 PM René Scharfe <l.s.r@xxxxxx> wrote:
> Am 02.12.2022 um 01:06 schrieb Ævar Arnfjörð Bjarmason:
> > Don't hide the exit code from the "git checkout" we run to checkout
> > our attributes file.
> > @@ -294,11 +294,17 @@ checkout_files () {
> > pfx=eol_${ceol}_crlf_${crlf}_attr_${attr}_ &&
> > for f in LF CRLF LF_mix_CR CRLF_mix_LF LF_nul
> > do
> > - rm crlf_false_attr__$f.txt &&
> > - if test -z "$ceol"; then
> > - git checkout -- crlf_false_attr__$f.txt
> > + if test -z "$ceol"
> > + then
> > + test_expect_success "setup $f checkout" '
> > + rm crlf_false_attr__$f.txt &&
> > + git checkout -- crlf_false_attr__$f.txt
> > + '
> > else
> > - git -c core.eol=$ceol checkout -- crlf_false_attr__$f.txt
> > + test_expect_success "setup $f checkout with core.eol=$ceol" '
> > + rm crlf_false_attr__$f.txt &&
> > + git -c core.eol=$ceol checkout -- crlf_false_attr__$f.txt
> > + '
>
> That adds five test_expect_success calls. Wouldn't one suffice, for the
> whole for loop, and a "|| return 1"?
Seems like a reasonable idea.
> One line above the context there's a "git config" call that should also
> be covered, right?
Also, the call to create_gitattributes() just above that line is in
the function's &&-chain, which implies that it too should be covered.
