Le 15/12/2016 à 09:54, Sawyer X a écrit : > > On 12/13/2016 04:15 AM, Ricardo Signes wrote: >> * Sawyer X <xsawyerx@gmail.com> [2016-12-08T06:31:31] >>> I would like to then ask the following question: How plausible is it to >>> "fix" Module::Load, considering people might expect its heuristic behavior? >> I have made a very hasty pass at this. The correct fix, I believe, is to add a >> load_module routine that does the same things as load(), but never guesses >> at whether the user means a file or a module. This needs a bit more work, >> at least, but it might already be mostly correct. >> >> It's in the branch rjbs/module-load-module and this is the commit: >> >> https://perl5.git.perl.org/perl.git/commitdiff/39b99316be07230d9e6fbc148d370772f1b873a6 > > Nice work! > > I'm not familiar enough with Module::Load's internals to review the > patch in earnest. > > Perhaps it would be easier to submit this in an RT ticket? > If I read the diff correctly, the _is_file() branch may now also call _load_mod() when $import is false. Isn't that different from the old behaviour? VincentThread Previous | Thread Next