On Wed, Jun 27, 2012 at 04:35:00PM -0000, Father Chrysostomos wrote: > Nicholas Clark asked: > > Is there a way? > > As of, um, c9d5e35e, that appears to be the only way. Hmm. Perhaps it *should* be more reachable. Currently, if one does this: $ ./perl -Ilib -e 'require "less.pm"; warn $INC{"less.pm"}' lib/less.pm at -e line 1. whereas this fails with this message: $ ./perl -Ilib -e 'require "./less.pm"; warn $INC{"less.pm"}' Can't locate ./less.pm in @INC (@INC contains: lib /usr/local/blead/lib/perl5/site_perl/5.17.2/darwin-2level /usr/local/blead/lib/perl5/site_perl/5.17.2 /usr/local/blead/vendor/lib/perl5/vendor_perl/5.17.2/darwin-2level /usr/local/blead/vendor/lib/perl5/vendor_perl/5.17.2 /usr/local/blead/lib/perl5/5.17.2/darwin-2level /usr/local/blead/lib/perl5/5.17.2 .) at -e line 1. but that's not totally true, as the @INC search wasn't attempted (at least, not for directories. coderefs are checked) The message changed in commit cf42f822f042b255 (in Jan 2002). It also gives: $ ./perl -Ilib -e 'require "less.pm"; warn $INC{"less.pm"}' lib/less.pm at -e line 1. $ ./perl -Ilib -e 'require "./less.pm"; warn $INC{"less.pm"}' Can't locate ./less.pm in @INC (@INC contains: lib /usr/local/lib/perl5/5.7.2/x86_64-linux /usr/local/lib/perl5/5.7.2 /usr/local/lib/perl5/site_perl/5.7.2/x86_64-linux /usr/local/lib/perl5/site_perl/5.7.2 /usr/local/lib/perl5/site_perl .) at -e line 1. whereas at commit 3d1a39c8be0d615d, 2 earlier it was this: $ ./perl -Ilib -e 'require "less.pm"; warn $INC{"less.pm"}' lib/less.pm at -e line 1. $ ./perl -Ilib -e 'require "./less.pm"; warn $INC{"less.pm"}' Can't locate ./less.pm at -e line 1. It's a side effect of an intended behaviour change - that coderefs in @INC should be tried for all pathnames. Thread starts here: http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2002-01/msg00875.html But I'm not exactly sure. If there are no coderefs in @INC, it seems obvious that the message should be the old one, not mentioning @INC. But if there are coderefs in @INC, what should be displayed in the message? All of @INC? Or just the bits that were tested? Nicholas ClarkThread Previous | Thread Next