sqrt without EXPR and Math::Complex always 0 #9620
Comments
From frank.wiegand@gmail.comThis is a bug report for perl from frank.wiegand@gmail.com, perldoc -f sqrt says: | Return the square root of EXPR. If EXPR is omitted, returns square Let's try: % perl -MMath::Complex -E '$_=+1; say sqrt' The EXPR form works fine: % perl -MMath::Complex -E 'say sqrt +1' Without -MMath::Complex everything works fine, according to the docs. Thanks, Frank Flags: Site configuration information for perl 5.10.0: Configured by Debian Project at Thu Jan 1 15:40:00 UTC 2009. Summary of my perl5 (revision 5 version 10 subversion 0) configuration: Locally applied patches: @INC for perl 5.10.0: Environment for perl 5.10.0: PATH=/sbin:/home/fw/bin:/home/fw/bin:/usr/local/bin:/usr/bin:/bin:/usr/games |
From wyp3rlx02@sneakemail.comHi Frank, thanks for the bug report, Frank Wiegand wrote:
This is because Math::Complex defines a new sqrt() function that can I did *not* check all other Math::Complex functions for the same defect. Best regards, |
From wyp3rlx02@sneakemail.com0001-Quick-fix-for-Math-Complex-sqrt-not-falling-back.patchFrom 0e83e9816665b6a5a66f31456bd554f8e0417544 Mon Sep 17 00:00:00 2001
From: Steffen Mueller <smueller@cpan.org>
Date: Fri, 16 Jan 2009 09:16:57 +0100
Subject: [PATCH] Quick fix for Math::Complex::sqrt() not falling back to $_ (RT #62412)
---
lib/Math/Complex.pm | 2 +-
lib/Math/Complex.t | 10 ++++++++++
2 files changed, 11 insertions(+), 1 deletions(-)
diff --git a/lib/Math/Complex.pm b/lib/Math/Complex.pm
index 7d05a44..6c5b3b4 100644
--- a/lib/Math/Complex.pm
+++ b/lib/Math/Complex.pm
@@ -672,7 +672,7 @@ sub arg {
# Therefore if you want the two solutions use the root().
#
sub sqrt {
- my ($z) = @_;
+ my ($z) = @_ ? @_ : $_;
my ($re, $im) = ref $z ? @{$z->_cartesian} : ($z, 0);
return $re < 0 ? cplx(0, CORE::sqrt(-$re)) : CORE::sqrt($re)
if $im == 0;
diff --git a/lib/Math/Complex.t b/lib/Math/Complex.t
index ee02341..99514fc 100755
--- a/lib/Math/Complex.t
+++ b/lib/Math/Complex.t
@@ -106,6 +106,16 @@ push(@script, <<'EOT');
(Im($z) + 1 ) < $eps;
EOT
push(@script, qq(print "ok $test\\n"}\n));
+
+ $test++;
+ push @script, <<'TEST_FOR_DEFAULT_ARG';
+{
+ local $_ = cplx( 4, 0);
+ my $res = sqrt();
+ print "# $test $_=cplx(4, 0); sqrt() == 2\n";
+ print 'not ' if abs($res - 2) > $eps;
+TEST_FOR_DEFAULT_ARG
+ push(@script, qq(print "ok $test\\n"}\n));
}
test_mutators();
--
1.5.4.3
|
|
The RT System itself - Status changed from 'new' to 'open' |
From wyp3rlx02@sneakemail.comHi Andrew, hi Jarkko, the Porting/Maintainers.pl document in core now has a field that The attached patch fixes a minor issue in Math::Complex and adds a test. Best regards, |
From wyp3rlx02@sneakemail.comHi Frank, thanks for the bug report, Frank Wiegand wrote:
This is because Math::Complex defines a new sqrt() function that can I did *not* check all other Math::Complex functions for the same defect. Best regards, |
From wyp3rlx02@sneakemail.com0001-Quick-fix-for-Math-Complex-sqrt-not-falling-back.patchFrom 0e83e9816665b6a5a66f31456bd554f8e0417544 Mon Sep 17 00:00:00 2001
From: Steffen Mueller <smueller@cpan.org>
Date: Fri, 16 Jan 2009 09:16:57 +0100
Subject: [PATCH] Quick fix for Math::Complex::sqrt() not falling back to $_ (RT #62412)
---
lib/Math/Complex.pm | 2 +-
lib/Math/Complex.t | 10 ++++++++++
2 files changed, 11 insertions(+), 1 deletions(-)
diff --git a/lib/Math/Complex.pm b/lib/Math/Complex.pm
index 7d05a44..6c5b3b4 100644
--- a/lib/Math/Complex.pm
+++ b/lib/Math/Complex.pm
@@ -672,7 +672,7 @@ sub arg {
# Therefore if you want the two solutions use the root().
#
sub sqrt {
- my ($z) = @_;
+ my ($z) = @_ ? @_ : $_;
my ($re, $im) = ref $z ? @{$z->_cartesian} : ($z, 0);
return $re < 0 ? cplx(0, CORE::sqrt(-$re)) : CORE::sqrt($re)
if $im == 0;
diff --git a/lib/Math/Complex.t b/lib/Math/Complex.t
index ee02341..99514fc 100755
--- a/lib/Math/Complex.t
+++ b/lib/Math/Complex.t
@@ -106,6 +106,16 @@ push(@script, <<'EOT');
(Im($z) + 1 ) < $eps;
EOT
push(@script, qq(print "ok $test\\n"}\n));
+
+ $test++;
+ push @script, <<'TEST_FOR_DEFAULT_ARG';
+{
+ local $_ = cplx( 4, 0);
+ my $res = sqrt();
+ print "# $test $_=cplx(4, 0); sqrt() == 2\n";
+ print 'not ' if abs($res - 2) > $eps;
+TEST_FOR_DEFAULT_ARG
+ push(@script, qq(print "ok $test\\n"}\n));
}
test_mutators();
--
1.5.4.3
|
From @AbigailOn Fri, Jan 16, 2009 at 09:41:12AM +0100, Steffen Mueller wrote:
That will not pick up any lexical $_. It would if you'd used a '_' prototype: sub sqrt (_) { Abigail |
From wyp3rlx02@sneakemail.comHi, Abigail abigail-at-abigail.be |newsgroups6| wrote:
doh, of course. Thanks for catching that. If I recall correctly, the _ In any case, please disregard my patch. Best regards, |
From @AbigailOn Fri, Jan 16, 2009 at 11:33:08AM +0100, Steffen Mueller wrote:
Indeed. Both the '_' prototype and lexical $_ are new in 5.10. Abigail |
From @jhiAbigail wrote:
Which means without BEGIN + eval games that can't be in M::C.
|
From @nwc10On Fri, Jan 16, 2009 at 07:58:00AM -0500, Jarkko Hietaniemi wrote:
I don't think that an eval is needed - simply assign one of the alternatives Or alternatively, a Corelist_pm.PL that has the two alternatives, and (Hmm. Without XS, I can't see any way to change the prototype of a function Nicholas Clark |
From @rgs2009/1/16 Nicholas Clark <nick@ccl4.org>:
Scalar::Util::set_prototype() ? |
From @jhiSteffen Mueller wrote:
I would prefer the patches being sent to us (jhi or zefram).
|
From wyp3rlx02@sneakemail.comHi, Jarkko Hietaniemi jhi-at-iki.fi |newsgroups6| wrote:
Thanks for the info, I updated Porting/Maintainers.pl accordingly. Best regards, |
From @jhiRafael Garcia-Suarez wrote:
Okay, call me dim, but how do I write an interface that (either accepts |
From @AbigailOn Sun, Jan 18, 2009 at 02:05:12PM -0500, Jarkko Hietaniemi wrote:
This works for me: #!/usr/bin/perl use strict; use Scalar::Util 'set_prototype'; $_ = "python"; sub hello { BEGIN { { __END__ With the appropriate line uncommented, it prints "Hello, world" for me Abigail |
From @jhi
No eval, true. But "no warnings 'syntax'", if that's left in I get these: Prototype mismatch: sub Math::Complex::abs (_) vs none ... Yuck. |
From @AbigailOn Sun, Jan 18, 2009 at 09:04:16PM -0500, Jarkko Hietaniemi wrote:
Ah, I didn't realize that. (I always have 'syntax' warnings disabled). But when I enable syntax warnings, I only get the warning if I do: sub hello; If I set the prototype after the sub has been defined, all seems to be well. Abigail |
From @nwc10On Mon, Jan 19, 2009 at 03:22:01AM +0100, Abigail wrote:
That's a valid warning.
Indeed. The facetious version: I have a patch. It removes code (that's always good) Inline Patchdiff --git a/cpan/Math-Complex/lib/Math/Complex.pm b/cpan/Math-Complex/lib/Math/
index 8475a2b..afec359 100644
--- a/cpan/Math-Complex/lib/Math/Complex.pm
+++ b/cpan/Math-Complex/lib/Math/Complex.pm
@@ -73,20 +73,6 @@ use Scalar::Util qw(set_prototype);
use warnings;
no warnings 'syntax'; # To avoid the (_) warnings.
-BEGIN {
- # For certain functions that we override, in 5.10 or better
- # we can set a smarter prototype that will handle the lexical $_
- # (also a 5.10+ feature).
- if ($] >= 5.010000) {
- set_prototype \&abs, '_';
- set_prototype \&cos, '_';
- set_prototype \&exp, '_';
- set_prototype \&log, '_';
- set_prototype \&sin, '_';
- set_prototype \&sqrt, '_';
- }
-}
-
my $i;
my %LOGN;
1: All those calls to set_prototype aren't achieving anything It's necessary to set the prototype *after* the subroutine is defined Otherwise, it turns out that what happens is: 1: sub is stubbed and has prototype added by the call (in that BEGIN block) And as the body has no prototype, that (lack of) prototype wins. Compare: $ ./perl -Ilib -MScalar::Util=set_prototype -lwe 'BEGIN {set_prototype \&foo, "_"}; print prototype \&foo; sub foo {}' $ ./perl -Ilib -MScalar::Util=set_prototype -lwe 'sub foo {} BEGIN {set_prototype \&foo, "_"}; print prototype \&foo' I don't have a patch to offer. Moving the BEGIN block later doesn't cause (And I'm more interested in why something else I'm trying doesn't quite work, Nicholas Clark |
From @nwc10On Wed, Nov 17, 2010 at 10:19:21PM +0000, Nicholas Clark wrote:
Specifically, by "later" I had it at the end, so that the code of the Inline Patchdiff --git a/cpan/Math-Complex/lib/Math/Complex.pm b/cpan/Math-Complex/lib/Math/
index 8475a2b..201b4fd 100644
--- a/cpan/Math-Complex/lib/Math/Complex.pm
+++ b/cpan/Math-Complex/lib/Math/Complex.pm
@@ -806,7 +806,7 @@ sub Im {
# Return or set rho(w).
#
sub rho {
- Math::Complex::abs(@_);
+ &Math::Complex::abs(@_);
}
#
@@ -874,7 +874,7 @@ sub log {
#
# Alias for log().
#
-sub ln { Math::Complex::log(@_) }
+sub ln { &Math::Complex::log(@_) }
#
# log10
Nicholas Clark |
From @dcollinsnThis was fixed in 5.10.1. -- |
|
@dcollinsn - Status changed from 'open' to 'resolved' |
Migrated from rt.perl.org#62412 (status was 'resolved')
Searchable as RT62412$
The text was updated successfully, but these errors were encountered: