pugs.blogs.com | pugscode.org | pugs.kwiki.org | paste: sial.org/pbot/perl6 | <stevan> Moose... it's the new Camel ":P | .pmc == PPI source filters!
Set by Alias_ on 16 March 2006.
pfenwick waves to Alias 01:37
svnbot6 r10458 | fglock++ | PCR - new module Pugs::Compiler::RegexPerl5 - wraps a perl5 regex into a method that returns a Match object. 02:54
szbalint This may sound silly, but why does 'not' work at some places where '!' doesn't? I missed the difference between the two. 20:01
(I've bumped into a test where !-d was changed to not -d) 20:04
Steve_p szbalint: There is a difference in operator precedence between ! and "not". 20:07
PerlJam szbalint: They only differ in precedence, so maybe the place where ! doesn't work is a place where is applied to the wrong thing 20:08
(wrong relative to your expectation)
szbalint I see, thanks... 20:14
That's what I thought too after looking at S03 but I wasn't sure.
svnbot6 r10459 | szbalint++ | Fixing shebang for t/operators/filetest.t 20:19
TimToady szbalint: actually, I think that was changed because !-d showed 20:31
a bug in the parser.
so I moved the bug to one test and fixed the other tests to actually test the filetest rather than the parser.
szbalint ah okay.
TimToady (if I recall correctly)
szbalint I had a very old installation with lib.pm before 20:32
that used !-d and for some reason was still around
TimToady but either of them should work fine.
"not" will work even better after we make it not a list operator 20:33
but stacked unary operators don't really care about precedence at all.
unless you happen to stack both a prefix and a postfix, then they have to negotiate 20:34
with the default tables, however, all postfixes are tighter than all prefixes naturally
szbalint so the parser bug got fixed in the meantime or is still around?
TimToady dunno 20:35
Steve_p Oh, yeah, not(()) caused some problems in Perl 5
TimToady yes, it's a false economy to combine not's precedence into list ops. 20:36
szbalint pugs -e 'my $dir = "ext"; if (!-d $dir) { say "no such dir" }' ## this still errors away, so I guess not
TimToady The actual test for the bug is probably misplaced. It's at the end of chmod.t: 20:39
ok(try { !-e "nonesuch" }, "!-e syntax works", :todo);
Steve_p Ohhh...
TimToady afk & 20:51

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!