On Sun, 27 Apr 2008 23:38:30 +0200, "Rafael Garcia-Suarez" <rgarciasuarez@gmail.com> wrote: > Might be correct. Do all tests pass including the new ones you'll > probably have written for this bug ? :) Writing tests is another issue... What should be tested? That length $@ is empty? (which was the problem in the bug report) Or would a proper test be to check that only the SvPV flag is set (and that there is no magic, no UTF8, ...) at the start of an eval/after a successfull eval? (If so - how does one tests the flags of the SV in the test scripts? any examples?) Kind regards, BramThread Previous | Thread Next