[tapx-dev] [185] ** maybe a little silly **

Eric Wilhelm scratchcomputing at gmail.com
Mon Jul 16 08:06:48 BST 2007


Ok, it is definitely not working.  The problem is I haven't been 
tracking the changes very closely so I haven't got a clue where it 
broke.

# on Sunday 15 July 2007 11:51 pm:

>Added Paths:
>-----------
>    trunk/t/execls/
>    trunk/t/execls/1..1
>    trunk/t/execls/ok 1

This is cute, but I think ExtUtils::Manifest will croak on the "ok 1" 
file.  Maybe we should just exec cat?  data/catme.1, data/catme.2, 
etc ?

Help!  It can't really be this hard to run a test.  I don't know what 
I'm doing here or why the innards have to be manipulated in such a sick 
way.  They certainly shouldn't be copied and pasted.  Do we just need 
an exec.t?

>+    my @output;
>+    local $^W;
>+    local *TAP::Harness::_should_show_count = sub {0};
>+    local *TAP::Harness::output = sub {
>+        my $self = shift;
>+        push @output => grep { $_ ne '' }
>+          map {
>+            local $_ = $_;
>+            chomp;
>+            trim($_)
>+          } @_;
>+    };
>+    my $harness = TAP::Harness->new(
>+        {   verbose => 1,
>+            exec    => [$ls],
>+        }
>+    );

I did drop this into r77 and verify that it would have passed then, so I 
know something is broken and the test should pass on linux.  Andy, 
where did the relevant code go?

I'm having a similar problem in Module::Build.  The testing tools are 
not accessible enough or I can't find any sort of foothold that won't 
cost me an afternoon to figure out the "right way" to test a seemingly 
simple thing.  Perhaps a trail hasn't been blazed here yet and I just 
need to add some helpers to lessen the pain, but I really thing a 
testing howto with general guidelines for the project would help.  I 
don't have problems with testing in general, just this "testing a test 
harness" and "testing a build system" sort of stuff which quickly goes 
supermeta.  I think it requires a certain amount of trickery and some 
notes would really help alleviate the hall-of-mirrors effect.

Sorry to dump code and run, but that's all I can manage at the moment.  
Please fix my crap code kthxbye :-D

--Eric
-- 
"Matter will be damaged in direct proportion to its value."
--Murphy's Constant
---------------------------------------------------
    http://scratchcomputing.com
---------------------------------------------------


More information about the tapx-dev mailing list