From: Vincent Pit Date: Tue, 28 Dec 2010 11:06:15 +0000 (+0100) Subject: Document differences X-Git-Tag: v0.01~9 X-Git-Url: http://git.vpit.fr/?p=perl%2Fmodules%2FTest-Leaner.git;a=commitdiff_plain;h=521c714591253428d7b6e4361d3718e6fe7fbaf2 Document differences --- diff --git a/lib/Test/Leaner.pm b/lib/Test/Leaner.pm index 92f386f..d11c4e2 100644 --- a/lib/Test/Leaner.pm +++ b/lib/Test/Leaner.pm @@ -30,6 +30,24 @@ our $VERSION = '0.01'; When profiling some L-based test script that contained about 10 000 unit tests, I realized that 60% of the time was spent in L itself, even though every single test actually involved a costly C. This module aims to be a partial replacement to L in those situations where you want to run a large number of simple tests. +Its functions behave the same as their L counterparts, except for the following differences : + +=over 4 + +=item * + +Stringification isn't forced on the test operands. +However, L honors C<'bool'> overloading, L honors C<'eq'> overloading and L honors whichever overloading category corresponds to the specified operator. + +=item * + +L, L, L, L, L, L, L and L are all guaranteed to return the truth value of the test. + +=item * + +C, C, C, C, C, C, C, C blocks and C are not implemented. + +=back =cut @@ -495,9 +513,6 @@ END { =pod -L, L, L, L, L, L, L and L are all guaranteed to return the truth value of the test. -Their L counterparts behave the same, but it is not documented anywhere. - L also provides some functions of its own, which are never exported. =head2 C