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.

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

Filehandle capture confuses PerlIO layers #12320

Open
p5pRT opened this issue Aug 7, 2012 · 7 comments
Open

Filehandle capture confuses PerlIO layers #12320

p5pRT opened this issue Aug 7, 2012 · 7 comments

Comments

@p5pRT
Copy link

p5pRT commented Aug 7, 2012

Migrated from rt.perl.org#114404 (status was 'open')

Searchable as RT114404$

@p5pRT
Copy link
Author

p5pRT commented Aug 7, 2012

From @xdg

Created by @xdg

When Capture​::Tiny saves, redirects and restores filehandles to capture output,
it also duplicates the original filehandle layers on the capturing filehandle.
Something in this process appears to confuse PerlIO. In the example file
below, PerlIO​::get_layers shows the same layers before and after, yet
afterwards a wide character in print warning is issued. I have confirmed this
issue in Perl 5.14.2, Perl 5.10.1 and Perl 5.8.9. (Originally reported against
Perl 5.16.0 in Capture​::Tiny bug rt.cpan.org #78819)

  use Encode;
  use Capture​::Tiny qw(capture);

  sub say { print @​_, "\n" }

  my $foo = Encode​::decode_utf8("\xC5\x82");

  binmode( STDOUT, "​:utf8" );

  out('PRE');
  capture { say "hello world" };
  out('POST');

  sub out {
  say STDOUT "$_[0]​:";
  say STDOUT "fd " . fileno(STDOUT) . " has " . join ', ', PerlIO​::get_layers( \*STDOUT );
  say STDOUT $foo;
  say STDOUT '';
  }

  ##### OUTPUT #####
  #PRE​:
  #fd 1 has unix, perlio, utf8
  #ł
  #
  #POST​:
  #fd 1 has unix, perlio, utf8
  #Wide character in print at /home/david/perl5/perlbrew/perls/perl-5.14.2/lib/5.14.2/x86_64-linux/IO/Handle.pm line 430.
  #ł

Perl Info

Flags:
    category=core
    severity=medium

Site configuration information for perl 5.14.2:

Configured by david at Thu Jan  5 16:18:53 EST 2012.

Summary of my perl5 (revision 5 version 14 subversion 2) configuration:
   
  Platform:
    osname=linux, osvers=3.0.0-14-generic, archname=x86_64-linux
    uname='linux vulcan 3.0.0-14-generic #23-ubuntu smp mon nov 21 20:28:43 utc 2011 x86_64 x86_64 x86_64 gnulinux '
    config_args='-de -Dprefix=/home/david/perl5/perlbrew/perls/perl-5.14.2'
    hint=recommended, useposix=true, d_sigaction=define
    useithreads=undef, usemultiplicity=undef
    useperlio=define, d_sfio=undef, uselargefiles=define, usesocks=undef
    use64bitint=define, use64bitall=define, uselongdouble=undef
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='cc', ccflags ='-fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
    optimize='-O2',
    cppflags='-fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include'
    ccversion='', gccversion='4.6.1', gccosandvers=''
    intsize=4, longsize=8, ptrsize=8, doublesize=8, byteorder=12345678
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
    ivtype='long', ivsize=8, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
    alignbytes=8, prototype=define
  Linker and Libraries:
    ld='cc', ldflags =' -fstack-protector -L/usr/local/lib'
    libpth=/usr/local/lib /lib/x86_64-linux-gnu /lib/../lib /usr/lib/x86_64-linux-gnu /usr/lib/../lib /lib /usr/lib
    libs=-lnsl -ldb -ldl -lm -lcrypt -lutil -lc
    perllibs=-lnsl -ldl -lm -lcrypt -lutil -lc
    libc=, so=so, useshrplib=false, libperl=libperl.a
    gnulibc_version='2.13'
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E'
    cccdlflags='-fPIC', lddlflags='-shared -O2 -L/usr/local/lib -fstack-protector'

Locally applied patches:
    


@INC for perl 5.14.2:
    /home/david/perl5/perlbrew/perls/perl-5.14.2/lib/site_perl/5.14.2/x86_64-linux
    /home/david/perl5/perlbrew/perls/perl-5.14.2/lib/site_perl/5.14.2
    /home/david/perl5/perlbrew/perls/perl-5.14.2/lib/5.14.2/x86_64-linux
    /home/david/perl5/perlbrew/perls/perl-5.14.2/lib/5.14.2
    .


Environment for perl 5.14.2:
    HOME=/home/david
    LANG=en_US.UTF-8
    LANGUAGE=en_US:en
    LC_COLLATE=C
    LC_CTYPE=en_US.UTF-8
    LC_MESSAGES=en_US.UTF-8
    LD_LIBRARY_PATH (unset)
    LOGDIR (unset)
    PATH=/home/david/perl5/perlbrew/bin:/home/david/perl5/perlbrew/perls/perl-5.14.2/bin:~/bin:~/git/utility-scripts:/opt/perl/current/bin:/home/david/bin:/usr/lib/lightdm/lightdm:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/opt/vagrant/bin:.
    PERLBREW_BASHRC_VERSION=0.39
    PERLBREW_HOME=/home/david/.perlbrew
    PERLBREW_MANPATH=/home/david/perl5/perlbrew/perls/perl-5.14.2/man
    PERLBREW_PATH=/home/david/perl5/perlbrew/bin:/home/david/perl5/perlbrew/perls/perl-5.14.2/bin
    PERLBREW_PERL=perl-5.14.2
    PERLBREW_ROOT=/home/david/perl5/perlbrew
    PERLBREW_VERSION=0.46
    PERL_BADLANG (unset)
    PERL_EXTUTILS_AUTOINSTALL=--defaultdeps
    SHELL=/bin/bash

@p5pRT
Copy link
Author

p5pRT commented Aug 7, 2012

From @xdg

Note -- triggering this behavior requires Capture​::Tiny 0.18 or earlier,
as later versions force a relayering back on top of restored filehandles
as a workaround to this bug.

Just because Capture​::Tiny has a workaround does not resolve this issue.
There is still no reason why PerlIO​::get_layers should be reporting
layers that aren't actually in effect on a handle.

-- David

@p5pRT
Copy link
Author

p5pRT commented Aug 7, 2012

@xdg - Status changed from 'new' to 'open'

@p5pRT
Copy link
Author

p5pRT commented Aug 7, 2012

From @dmcbride

On Monday August 6 2012 5​:18​:48 PM David Golden wrote​:

When Capture​::Tiny saves, redirects and restores filehandles to capture
output, it also duplicates the original filehandle layers on the capturing
filehandle. Something in this process appears to confuse PerlIO. In the
example file below, PerlIO​::get_layers shows the same layers before and
after, yet afterwards a wide character in print warning is issued. I have
confirmed this issue in Perl 5.14.2, Perl 5.10.1 and Perl 5.8.9.
(Originally reported against Perl 5.16.0 in Capture​::Tiny bug rt.cpan.org
#78819)

This is a duplicate of https://rt-archive.perl.org/perl5/Ticket/Display.html?id=113982

use Encode;
use Capture​::Tiny qw(capture);

sub say { print @​_, "\n" }

my $foo = Encode​::decode_utf8("\xC5\x82");

binmode( STDOUT, "​:utf8" );

out('PRE');
capture { say "hello world" };
out('POST');

sub out {
say STDOUT "$_[0]​:";
say STDOUT "fd " . fileno(STDOUT) . " has " . join ', ',
PerlIO​::get_layers( \*STDOUT ); say STDOUT $foo;

You need to add ", output => 1" here​:

  say "fd " . fileno(STDOUT) . " has " . join ', ', PerlIO​::get_layers(
\*STDOUT, output => 1 );

Then you see the difference. Yes, I faced the same quandary with the previous
bug report.

(This may also mean that the fix/workaround to Capture​::Tiny might be off in
that it's only checking the input layers, not the output layers, though I'm
assured that most of the time they're the same.)

say STDOUT '';

}

##### OUTPUT #####
#PRE​:
#fd 1 has unix, perlio, utf8

#
#POST​:
#fd 1 has unix, perlio, utf8
#Wide character in print at

@p5pRT
Copy link
Author

p5pRT commented Aug 7, 2012

From @xdg

On Mon, Aug 6, 2012 at 11​:11 PM, Darin McBride <dmcbride@​cpan.org> wrote​:

This is a duplicate of https://rt-archive.perl.org/perl5/Ticket/Display.html?id=113982

I don't think it's quite the same, as this bug also points out that
get_layers is reporting it incorrectly.

I've marked them as referring to each other, so whoever fixes it can
close them both.

-- David

@p5pRT
Copy link
Author

p5pRT commented Aug 7, 2012

From @Leont

On Tue, Aug 7, 2012 at 6​:11 AM, Darin McBride <dmcbride@​cpan.org> wrote​:

(This may also mean that the fix/workaround to Capture​::Tiny might be off in
that it's only checking the input layers, not the output layers, though I'm
assured that most of the time they're the same.)

The layers shouldn't be different. binmode explicitly tries to make
sure of that if the handles are different. The reason why have this
duality is that on for example sockets input and output need separate
buffers.

Leon

@p5pRT
Copy link
Author

p5pRT commented Aug 7, 2012

From @Leont

On Tue, Aug 7, 2012 at 6​:15 AM, David Golden <xdaveg@​gmail.com> wrote​:

I don't think it's quite the same, as this bug also points out that
get_layers is reporting it incorrectly.

I don't see how it's incorrect, it's just very confusing.

Leon

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants