On Tue, Mar 25, 2008 at 3:53 PM, Gergely Brautigam <[EMAIL PROTECTED]> wrote: > Well... We used Java with JUnit and .NET with NUnit... C++ for low level > testing and so fort and so on.. They all had there advantages and > disatvantages. I use perl for nearly one and a half year now. But appart > form the weird syntax and the golfs, and the perl haiqus it's a pretty > interesting language :). > > I would use JUnit and NUnit for larger project, I don't know if perl > could handle it...
For the unit tests - especially if written by the developers - you might be better off using the same language as the implementation language. Mainly because the developers are more familiar with those languages. In some case Perl might be better even there but it might not show all of its advantages.[1] In higher level tests which usually get the title "integration test" and "acceptance test" or "black box tests" you have no advantage using the same language for testing as the implementation language. In many cases the implementation is done in many languages and technologies anyway. (e.g a standard web application today uses at least HTML, CSS, JavaScript, SQL and a high level language for backend programming.) For such tests the one of the advantages of Perl is that - because of the expressiveness of the languages and the availability of CPAN modules - you need to write a lot less code than in either Java or C(#|++). regards Gabor [1] Actually I would really like to read about the experience of people here on the perl-qa list regarding the use of Perl white-box (and/or) unite testing of Java and .NET applications where Perl has a major role. I am sure Andy Lester will also like to publish such stories on http://PerlBuzz.com
