readline: sets $! even if no error occured #8289

p5pRT · 2006-01-20T10:32:35Z

Migrated from rt.perl.org#38285 (status was 'rejected')

Searchable as RT38285$

p5pRT · 2006-01-20T10:32:39Z

From Nikolaus@rath.org

Created by Nikolaus@rath.org

This is a bug report for perl from Nikolaus@rath.org,
generated with the help of perlbug 1.35 running under perl v5.8.7.

-----------------------------------------------------------------

Apparently, readline sets $! to "invalid filedescriptor" even if no
error occured:

[0] nokile:~/Work$ cat perlbug2.pl
#!/usr/bin/perl

# Overload IO::File to die on errors
package IO::File::Fatal;
use strict;
use warnings;
use IO::File;
use base 'IO::File';

sub getline {
my $this = shift;
undef $!;
my $ret = readline($this);
warn "got line '$ret'\n"; # for debugging
die $! if $!;
return $ret;
}

######## MAIN PROGRAM ##########
package main;
use strict;
use warnings;

my $fh = new IO::File::Fatal(shift(), "<");
print $fh->getline();

[0] nokile:~/Work$ echo "Just one line" > data

[0] nokile:~/Work$ ./perlbug2.pl data
got line 'Just one line
'
Bad file descriptor at ./perlbug2.pl line 15, <GEN0> line 1.

[9] nokile:~/Work$

Perl Info


Flags:
    category=core
    severity=medium

Site configuration information for perl v5.8.7:

Configured by Debian Project at Mon Dec 12 13:59:01 UTC 2005.

Summary of my perl5 (revision 5 version 8 subversion 7) configuration:
  Platform:
    osname=linux, osvers=2.6.10, archname=i486-linux-gnu-thread-multi
    uname='linux rothera 2.6.10 #1 smp fri may 13 09:24:22 utc 2005 i686 gnulinux '
    config_args='-Dusethreads -Duselargefiles -Dccflags=-DDEBIAN -Dcccdlflags=-fPIC -Darchname=i486-linux-gnu -Dprefix=/usr -Dprivlib=/usr/share/perl/5.8 -Darchlib=/usr/lib/perl/5.8 -Dvendorprefix=/usr -Dvendorlib=/usr/share/perl5 -Dvendorarch=/usr/lib/perl5 -Dsiteprefix=/usr/local -Dsitelib=/usr/local/share/perl/5.8.7 -Dsitearch=/usr/local/lib/perl/5.8.7 -Dman1dir=/usr/share/man/man1 -Dman3dir=/usr/share/man/man3 -Dsiteman1dir=/usr/local/man/man1 -Dsiteman3dir=/usr/local/man/man3 -Dman1ext=1 -Dman3ext=3perl -Dpager=/usr/bin/sensible-pager -Uafs -Ud_csh -Uusesfio -Uusenm -Duseshrplib -Dlibperl=libperl.so.5.8.7 -Dd_dosuid -des'
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=define use5005threads=undef useithreads=define usemultiplicity=define
    useperlio=define d_sfio=undef uselargefiles=define usesocks=undef
    use64bitint=undef use64bitall=undef uselongdouble=undef
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='cc', ccflags ='-D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -DDEBIAN -fno-strict-aliasing -pipe -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
    optimize='-O2',
    cppflags='-D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -DDEBIAN -fno-strict-aliasing -pipe -I/usr/local/include'
    ccversion='', gccversion='4.0.2 20050808 (prerelease) (Ubuntu 4.0.1-4ubuntu9)', gccosandvers=''
    intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12
    ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
    alignbytes=4, prototype=define
  Linker and Libraries:
    ld='cc', ldflags =' -L/usr/local/lib'
    libpth=/usr/local/lib /lib /usr/lib
    libs=-lgdbm -lgdbm_compat -ldb -ldl -lm -lpthread -lc -lcrypt
    perllibs=-ldl -lm -lpthread -lc -lcrypt
    libc=/lib/libc-2.3.5.so, so=so, useshrplib=true, libperl=libperl.so.5.8.7
    gnulibc_version='2.3.5'
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E'
    cccdlflags='-fPIC', lddlflags='-shared -L/usr/local/lib'

Locally applied patches:
    SPRINTF0 - fixes for sprintf formatting issues - CVE-2005-3962


@INC for perl v5.8.7:
    /etc/perl
    /usr/local/lib/perl/5.8.7
    /usr/local/share/perl/5.8.7
    /usr/lib/perl5
    /usr/share/perl5
    /usr/lib/perl/5.8
    /usr/share/perl/5.8
    /usr/local/lib/site_perl/i486-linux-gnu-thread-multi
    /usr/local/lib/site_perl
    .


Environment for perl v5.8.7:
    HOME=/home/nikratio
    LANG=en_US.UTF-8
    LANGUAGE (unset)
    LC_ADDRESS=de_DE.UTF-8
    LC_COLLATE=de_DE.UTF-8
    LC_CTYPE=de_DE.UTF-8
    LC_IDENTIFICATION=en_US.UTF-8
    LC_MEASUREMENT=de_DE.UTF-8
    LC_MESSAGES=en_US.UTF-8
    LC_MONETARY=de_DE.UTF-8
    LC_NAME=de_DE.UTF-8
    LC_NUMERIC=de_DE.UTF-8
    LC_PAPER=de_DE.UTF-8
    LC_TELEPHONE=de_DE.UTF-8
    LC_TIME=de_DE.UTF-8
    LD_LIBRARY_PATH (unset)
    LOGDIR (unset)
    PATH=~/bin:/usr/local/bin:/usr/bin:/bin:/usr/bin/X11
    PERL_BADLANG (unset)
    PERL_UNICODE=SDAL
    SHELL=/bin/bash

p5pRT · 2006-01-20T20:46:34Z

From eric-amick@comcast.net

[nikratio - Fri Jan 20 02:32:39 2006]:

This is a bug report for perl from Nikolaus@rath.org,
generated with the help of perlbug 1.35 running under perl v5.8.7.

-----------------------------------------------------------------

Apparently, readline sets $! to "invalid filedescriptor" even if no
error occured:

[0] nokile:~/Work$ cat perlbug2.pl
#!/usr/bin/perl

# Overload IO::File to die on errors
package IO::File::Fatal;
use strict;
use warnings;
use IO::File;
use base 'IO::File';

sub getline {
my $this = shift;
undef $!;
my $ret = readline($this);
warn "got line '$ret'\n"; # for debugging
die $! if $!;
return $ret;
}

######## MAIN PROGRAM ##########
package main;
use strict;
use warnings;

my $fh = new IO::File::Fatal(shift(), "<");
print $fh->getline();

[0] nokile:~/Work$ echo "Just one line" > data

[0] nokile:~/Work$ ./perlbug2.pl data
got line 'Just one line
'
Bad file descriptor at ./perlbug2.pl line 15, <GEN0> line 1.

[9] nokile:~/Work$

From the entry for $! in perlvar:

This means that the value of $! is meaningful only *immediately* after a
*failure* [emphasis in original]

The readline was successful, so you shouldn't be checking $!. Not a bug.

p5pRT · 2006-01-21T03:45:59Z

From @tamias

On Fri, Jan 20, 2006 at 12:46:34PM -0800, Eric Amick via RT wrote:

[nikratio - Fri Jan 20 02:32:39 2006]:

sub getline {
my $this = shift;
undef $!;
my $ret = readline($this);
warn "got line '$ret'\n"; # for debugging
die $! if $!;
return $ret;
}

From the entry for $! in perlvar:

This means that the value of $! is meaningful only *immediately* after a
*failure* [emphasis in original]

The readline was successful, so you shouldn't be checking $!. Not a bug.

However, from the entry for readline:

If readline encounters an operating system error, $! will be
set with the corresponding error message. It can be helpful
to check $! when you are reading from filehandles you don't
trust, such as a tty or a socket. The following example
uses the operator form of "readline", and takes the
necessary steps to ensure that "readline" was successful.

for (;;) {
undef $!;
unless (defined( $line = <> )) {
die $! if $!;
last; # reached EOF
}
# ...
}

I think this example is in error.

Ronald

p5pRT · 2006-01-21T03:46:01Z

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

p5pRT · 2006-01-21T19:18:29Z

From @demerphq

On 1/21/06, Ronald J Kimball <rjk-perl-p5p@tamias.net> wrote:

On Fri, Jan 20, 2006 at 12:46:34PM -0800, Eric Amick via RT wrote:

[nikratio - Fri Jan 20 02:32:39 2006]:

sub getline {
my $this = shift;
undef $!;
my $ret = readline($this);
warn "got line '$ret'\n"; # for debugging
die $! if $!;
return $ret;
}

From the entry for $! in perlvar:

This means that the value of $! is meaningful only *immediately* after a
*failure* [emphasis in original]

The readline was successful, so you shouldn't be checking $!. Not a bug.

However, from the entry for readline:
If readline encounters an operating system error\, $\! will be
set with the corresponding error message\.  It can be helpful
to check $\! when you are reading from filehandles you don't
trust\, such as a tty or a socket\.  The following example
uses the operator form of "readline"\, and takes the
necessary steps to ensure that "readline" was successful\.

    for $;;$ \{
        undef $\!;
        unless $defined\( $line = \<> $\) \{
            die $\! if $\!;
            last; \# reached EOF
        \}
        \# \.\.\.
    \}
I think this example is in error.

Why?

Note the difference between the example and the op. One undef's $!
before reading and then then immediately after checks $! but only if
the result is undefined. The other checks $! every time but after a
warn, which itself could set $!.

IMO the op should have said:

sub getline {
my $this = shift;
undef $!;
my $ret = readline($this);
die $!
if ! defined( $ret ) && $!; # must happen before anything else
return if ! defined( $ret );
warn "got line '$ret'\n"; # for debugging
return $ret;
}

Cheers,
yves

--
perl -Mre=debug -e "/just|another|perl|hacker/"

p5pRT · 2006-01-27T19:25:32Z

From Nikolaus@rath.org

Apparently, readline sets $! to "invalid filedescriptor" even if no
error occured:

sub getline {
my $this = shift;
undef $!;
my $ret = readline($this);
warn "got line '$ret'\n"; # for debugging
die $! if $!;
return $ret;
}

The readline was successful, so you shouldn't be checking $!. Not a bug.

So how do I know whether readline was successful if I can't check $!? An
undef might just be eof and no error.

Apart from that, perldoc -f readline recommends checking $!.

p5pRT · 2006-01-27T19:32:20Z

From Nikolaus@rath.org

Note the difference between the example and the op. One undef's $!
before reading and then then immediately after checks $! but only if
the result is undefined. The other checks $! every time but after a
warn, which itself could set $!.

Well, I inserted warn() only after discovering the problem. Checking $!
only if the readline return value is undef actually circumvents the problem.

However, I still consider the current behaviour a bug. Is there any
reason for setting $! if no error occured?

p5pRT · 2006-05-03T15:15:10Z

From Mark.Martinec@ijs.si

however, I still consider the current behaviour a bug.
Is there any reason for setting $! if no error occured?

It is not a bug, this follows the same principle as all system
calls in Unix follow: if a status returned from a system
routine indicates and error, then errno contains an error
description. If status does not indicate an error, then
errno is undefined - it may contain any value. There is
no guarantee that it is 0 or that its old value is retained.

IMO, this ticket should be closed as 'not a bug'.

Btw, there is a similar bug report #39060, but that one
is genuine, and is not to be confused with this one.

Mark

p5pRT · 2009-05-26T10:55:15Z

From skasal@redhat.com

On Wed May 03 08:15:10 2006, mmartinec wrote:

IMO, this ticket should be closed as 'not a bug'.

I reviewed this and I agree with the reasoning. So I'm trying to close
this.

p5pRT · 2009-05-26T10:58:35Z

@nwc10 - Status changed from 'open' to 'resolved'

p5pRT · 2009-06-02T15:04:29Z

skasal@redhat.com - Status changed from 'resolved' to 'rejected'

p5pRT closed this as completed Jun 2, 2009

p5pRT added Severity Medium distro-Linux type-core labels Oct 18, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

readline: sets $! even if no error occured #8289

readline: sets $! even if no error occured #8289

p5pRT commented Jan 20, 2006

p5pRT commented Jan 20, 2006

p5pRT commented Jan 20, 2006

p5pRT commented Jan 21, 2006

p5pRT commented Jan 21, 2006

p5pRT commented Jan 21, 2006

p5pRT commented Jan 27, 2006

p5pRT commented Jan 27, 2006

p5pRT commented May 3, 2006

p5pRT commented May 26, 2009

p5pRT commented May 26, 2009

p5pRT commented Jun 2, 2009

readline: sets $! even if no error occured #8289

readline: sets $! even if no error occured #8289

Comments

p5pRT commented Jan 20, 2006

p5pRT commented Jan 20, 2006

From Nikolaus@rath.org

Created by Nikolaus@rath.org

p5pRT commented Jan 20, 2006

From eric-amick@comcast.net

p5pRT commented Jan 21, 2006

From @tamias

p5pRT commented Jan 21, 2006

p5pRT commented Jan 21, 2006

From @demerphq

p5pRT commented Jan 27, 2006

From Nikolaus@rath.org

p5pRT commented Jan 27, 2006

From Nikolaus@rath.org

p5pRT commented May 3, 2006

From Mark.Martinec@ijs.si

p5pRT commented May 26, 2009

From skasal@redhat.com

p5pRT commented May 26, 2009

p5pRT commented Jun 2, 2009