Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx> writes: > Change the GIT_TEST_GETTEXT_POISON variable from being "non-empty?" to > being a more standard boolean variable. > > Since it needed to be checked in both C code and shellscript (via test > -n) it was one of the remaining shellscript-like variables. Now that > we have "git env--helper" we can change that. > > Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx> > --- Hmph. Even though I earlier said "we do not terribly mind breaking developers and that is why we allow these patches", I have second thoughts. Turning "If it is empty, it is false" to "if you want to say false, say it in one of those approved ways" is one thing. Forcing SWITCH=YesPlease to be rewritten to SWITCH=yes is quite different---we are breaking everybody who would have to read and follow po/README and t/README. We _might_ have to grandfarther YesPlease as a special value that is understood by "git env" (but not "git config") to ease the transition, as that token has been used as a sample true value in many places. But let's read on. Assuming that breaking those who hardcoded YesPlease in their scripts is OK, this patch looked sensible. Thanks.
