8000 regression in list assignment with undef in list context · Issue #16685 · Perl/perl5 · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

regression in list assignment with undef in list context #16685

Closed
p5pRT opened this issue Sep 7, 2018 · 5 comments
Closed

regression in list assignment with undef in list context #16685

p5pRT opened this issue Sep 7, 2018 · 5 comments

Comments

@p5pRT
Copy link
p5pRT commented Sep 7, 2018

Migrated from rt.perl.org#133503 (status was 'open')

Searchable as RT133503$
@p5pRT
Copy link
Author
p5pRT commented Sep 7, 2018 •
edited by toddr
Loading

From wolf-dietrich_moeller@t-online.de

The documentation on assignments in perlop reads:
"a list assignment in list context produces the
list of lvalues assigned to"

The following program shows a regression,
when the "list assigned to" contains "undef".
(tested with two old versions available to me:

  • Strawberry Perl Portable 5.22.3.1-32bit
  • Strawberry Perl Portable 5.26.2.1-32bit)

The newer version 5.26 (and also 5.28) works correctly if the
third element in the list assigned to is a variable (test 3),
but ignores the third element if it is undef.
Thus @s in test 4 only has 2 elements instead of 3.

my ($y, @s);
my $x = 5;
print 'test 1: ',scalar (@s = (($y) = (8,9))),' ',$y,' ',$x//'undef',"\n";
my $x = 5;
print 'test 2: ',scalar (@s = (($y,undef) = (8,9))),' ',$y,'
',$x//'undef',"\n";
my $x = 5;
print 'test 3: ',scalar (@s = (($y,undef,$x) = (8,9))),' ',$y,'
',$x//'undef',"\n";
my $x = 5;
print 'test 4: ',scalar (@s = (($y,undef,undef) = (8,9))),' ',$y,'
',$x//'undef',"\n";
Result in 5.22 (correct, @s == 3):
test 1: 1 8 5
test 2: 2 8 5
test 3: 3 8 undef
test 4: 3 8 5

Result in 5.26 (test 4 incorrect, @s == 2):
test 1: 1 8 5
test 2: 2 8 5
test 3: 3 8 undef
test 4: 2 8 5

@p5pRT
Copy link
Author
p5pRT commented Sep 7, 2018

From @iabyn

On Fri, Sep 07, 2018 at 08​:48​:29AM -0700, Wolf-Dietrich Moeller (via RT) wrote​:

The documentation on assignments in perlop reads​:
"a list assignment in list context produces the
list of lvalues assigned to"

The following program shows a regression,
when the "list assigned to" contains "undef".
(tested with two old versions available to me​:
- Strawberry Perl Portable 5.22.3.1-32bit
- Strawberry Perl Portable 5.26.2.1-32bit)

The newer version 5.26 (and also 5.28) works correctly if the
third element in the list assigned to is a variable (test 3),
but ignores the third element if it is undef.
Thus @​s in test 4 only has 2 elements instead of 3.

Bisects to this. I haven't looked further yet.

commit b09ed99
Author​: David Mitchell <davem@​iabyn.com>
AuthorDate​: Wed Oct 19 09​:41​:53 2016 +0100
Commit​: David Mitchell <davem@​iabyn.com>
CommitDate​: Wed Oct 26 08​:37​:27 2016 +0100

  Handle list assignment in list context better
 
  In something like
 
  sub inc { $_++ for @​_ }
  inc(($a,$b,$c,$d) = (10,20))
 
  The four scalar vars will end up with the values (11,21,1,1), with
  the list assign op returning 4 lval scalars to be passed to inc.
 
  However, when the LHS includes a hash or array, any 'empty' scalars weren't
  being returned. This​:
 
  inc(($a,$b,@​array,$c) = (10))
 
  used to leave $b and $c undefined. With this commit, they are both set to
  1.
 
  This change broke some tests in hashassign.t, which were added in 2012
  by commit v5.17.6-295-gb1babc5. Half these tests were commented as
 
  # this arguable, but this is how it works
 
  and they all tested that a scalar following a hash wasn't returned in
  lvalue context; i.e. they all assumed that
 
  inc((%h, $x) = (...))
 
  *wouldn't* increment $x. This commit causes $x to be incremented, and
  I've changed the failing tests.
 
  It also adds an EXTEND(SP,1) in the scalar case; otherwise with
  scalar( () = @​a) and empty @​a, it could in theory push the '0' return
  value off the end of the stack.

--
Spock (or Data) is fired from his high-ranking position for not being able
to understand the most basic nuances of about one in three sentences that
anyone says to him.
  -- Things That Never Happen in "Star Trek" #19

@p5pRT
Copy link
Author
p5pRT commented Sep 7, 2018

The RT System itself - Status changed from 'new' to 'open'
8000

@toddr
Copy link
Member
toddr commented Jan 31, 2020

This is unchanged in 5.30

iabyn added a commit that referenced this issue Aug 11, 2020
@iabyn
list assign in list context: honour LHS undef
282d9df 
GH #16685

In

    @A = ($x, undef, undef) = (1))

@A should have 3 elements. v5.25.6-79-gb09ed995ad broke this and was
returning one element.

The fix is simple: that previous commit made it so that elements were
pushed back onto the stack only if they weren't immortal, so
&PL_sv_undef was getting skipped. Make it so they always are.
@iabyn
Copy link
Contributor
iabyn commented Aug 11, 2020

Now fixed in blead by commit v5.33.0-236-g282d9dfeb4

@iabyn iabyn closed this as completed Aug 11, 2020
steve-m-hay pushed a commit that referenced this issue Jan 6, 2021
@iabyn @steve-m-hay
list assign in list context: honour LHS undef
b802abc 
GH #16685

In

    @A = ($x, undef, undef) = (1))

@A should have 3 elements. v5.25.6-79-gb09ed995ad broke this and was
returning one element.

The fix is simple: that previous commit made it so that elements were
pushed back onto the stack only if they weren't immortal, so
&PL_sv_undef was getting skipped. Make it so they always are.

(cherry picked from commit 282d9df)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@toddr @karenetheridge @iabyn @p5pRT
0