Re: [PATCH] Avoid difference in tr semantics between System V and BSD

[Jonathan Nieder <jrnieder@gmail.com>]

Johannes Sixt wrote:

> In other tests, we check for prerequisite PERL, i.e., we are prepared
> that perl is not available. Shouldn't we do that here, too?

I think the tests assume there's a perl present even when the PERL
prereq isn't present already.  E.g.:

	nul_to_q () {
		"$PERL_PATH" -pe 'y/\000/Q/'
	}

So in practice the PERL prereq just means "NO_PERL wasn't set", or
in other words, "commands that use perl work".  Maybe something
like the following would help?

Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>

diff --git i/t/README w/t/README
index 2167125..54cd064 100644
--- i/t/README
+++ w/t/README
@@ -629,11 +629,20 @@ See the prereq argument to the test_* functions in the "Test harness
 library" section above and the "test_have_prereq" function for how to
 use these, and "test_set_prereq" for how to define your own.
 
- - PERL & PYTHON
+ - PYTHON
 
-   Git wasn't compiled with NO_PERL=YesPlease or
-   NO_PYTHON=YesPlease. Wrap any tests that need Perl or Python in
-   these.
+   Git wasn't compiled with NO_PYTHON=YesPlease. Wrap any tests that
+   need Python with this.
+
+ - PERL
+
+   Git wasn't compiled with NO_PERL=YesPlease.
+
+   Even without the PERL prerequisite, tests can assume there is a
+   usable perl interpreter at $PERL_PATH, though it need not be
+   particularly modern.
+
+   Wrap tests for commands implemented in Perl with this.
 
  - POSIXPERM