The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.
NAME
    Capture::Tiny::Extended - Capture STDOUT and STDERR from from Perl, XS
    or external programs (with some extras)

VERSION
    version 0.114

SYNOPSIS
       use Capture::Tiny::Extended qw/capture tee capture_merged tee_merged/;
 
       # capture return values
 
       my ($stdout, $stderr, @return) = capture {
         # your code here
         return system( 'ls' );
       };
 
       ($merged, @return) = capture_merged {
         # your code here
         return system( 'ls' );
       };
 
       # or use explicit capture files
 
       ($stdout, $stderr, @return) = capture(
         sub { # your code here },
         { stdout => 'stdout.log' }
       );
 
       # or with sugar
 
       use Capture::Tiny::Extended qw/capture tee capture_merged tee_merged capture_files/;
 
       ($stdout, $stderr, @return) = capture {
         # your code here
       }
       capture_files (
         stdout => 'stdout.log',
         stderr => 'stderr.log',
       );

DESCRIPTION
    Capture::Tiny::Extended is a fork of Capture::Tiny. It is functionally
    identical with the parent module, except for the differences documented
    in this POD. Please see the documentation of Capture::Tiny for details
    on standard usage.

    Please note that this can be considered an experimental module in some
    respects. I am not as experienced with the subject matter (and in
    general) as David Golden and mostly implemented these features here
    because i needed them fast and did not have the time to spare to wait
    for them to get into Capture::Tiny. If you need capture functionality
    for mission-critical parts, consider whether Capture::Tiny might be
    enough for the job.

    Of course I will however make all efforts to make this as stable and
    useful as possible by keeping it up-to-date (as my time permits) with
    changes and bugfixes applied to Capture::Tiny, as well as responding and
    addressing and change requests or bug reports for this module.

DIFFERENCES
  Capturing Return Values
    When executing code within a capture you sometimes want to also keep the
    return value, for example when capturing a system() call. In
    Capture::Tiny this has to be done like this:

       use Capture::Tiny 'capture';
 
       my $res;
       my ( $out, $err ) = capture {
         $res = system( 'ls' );
       };

    Capture::Tiny::Extended automatically captures return values and returns
    them after the second return value (or first if you're using the merged
    functions).

       use Capture::Tiny::Extended 'capture';
 
       my ( $out, $err, $res ) = capture { system( 'ls' ) };

  Teeing In Realtime
    Sometimes you want to use Capture::Tiny to capture any and all output of
    an action and dump it into a log file, while also displaying it on the
    screen and then post-process the results later on (for example for
    sending status mails). The only way to do this with Capture::Tiny is
    code like this:

       use Capture::Tiny 'capture';
       use File::Slurp;
 
       my $res;
       my ( $out, $err ) = capture {
         # lockfile and other processing here along with debug output
         $res = system( 'long_running_program' );
       };
 
       file_write 'out.log', $out;
       send_mail( $err ) if $res;

    This has a very big disadvantage. If the long-running program runs too
    long, and the perl script is started by something like crontab there is
    no way for you to get at the log output. You will have to wait for it to
    complete before the captured output is written to the file.

    Capture::Tiny::Extended gives you the option to provide filenames for it
    to use as capture buffers. This means the output from the captured code
    will appear on the screen and in the file in realtime, and will
    afterwards be available to your Perl script in the variables returned by
    the capture function:

       use Capture::Tiny::Extended 'capture';
 
       my ( $out, $err, $res ) = capture(
         sub {
           # lockfile and other processing here along with debug output
           return system( 'long_running_program' );
         },
         {
           stdout => 'out.log',
           stderr => 'err.log',
         }
       );
 
       send_mail( $err ) if $res;

  capture_files
    Since using hashes in that way breaks a bit of the syntax magic of the
    capture functions (or makes them harder to read), there exists a sugar
    function to take the file arguments and pass it on to the capture
    functions:

       use Capture::Tiny::Extended qw( capture capture_files );
 
       my ( $out, $err, $res ) = capture {
         # lockfile and other processing here along with debug output
         return system( 'long_running_program' );
       }
       capture_files {
         stdout => 'out.log',
         stderr => 'err.log',
       };
 
       send_mail( $err ) if $res;

  Capture File Mode Options
    For purposes of avoiding data loss, the default behavior is to append to
    the specified files. The key 'new_files' can be set to a true value on
    the extra file hash parameter to instruct Capture::Tiny::Extended to
    attempt to make files. It will die however if the specified files
    already exist.

       use Capture::Tiny::Extended 'capture';
 
       my $out = capture_merged(
         sub { system( 'ls' ) },
         { stdout => 'out.log', new_files => 1 }
       );

    If existing files should always be overwritten, no matter what, the key
    'clobber' can be set instead:

       use Capture::Tiny::Extended 'capture';
 
       my $out = capture_merged(
         sub { system( 'ls' ) },
         { stdout => 'out.log', clobber => 1 }
       );

WHY A FORK?
    The realtime teeing feature was very important for one of my current
    projects and i needed it on CPAN to be able to easily distribute it to
    many systems. I had provided a patch for the return value capturing on
    Github to David Golden a long while ago, but due to being busy with real
    life, family and more important projects than this he was not able to
    find time to proof and integrate it and in the foreseeable future won't
    be able to either. At the same time i lack the Perl file handle,
    descriptor and layer chops to take full responsibility for Capture::Tiny
    itself. Usually i would have just written a subclass of the original,
    but since Capture::Tiny is written in functional style this was not
    possible.

    As such a fork seemed to be the best option to get these features out
    there. I'd be more than happy to see them integrated into C::T someday
    and will keep my git repository in such a state as to make this as easy
    as possible. (Lots of rebasing.)

ACKNOWLEDGEMENTS
    Capture::Tiny is an invaluable tool that uses practically indecent
    amounts of creativity to solve decidedly nontrivial problems and
    circumvents many cliffs the ordinary coder (and most certainly me) would
    inevitably crash against.

    Many thanks to David Golden for taking the time and braving all those
    traps of insanity to create Capture::Tiny.

SUPPORT
  Bugs / Feature Requests
    Please report any bugs or feature requests by email to
    "bug-capture-tiny-extended at rt.cpan.org", or through the web interface
    at
    <http://rt.cpan.org/Public/Dist/Display.html?Name=Capture-Tiny-Extended>
    . You will be automatically notified of any progress on the request by
    the system.

  Source Code
    This is open source software. The code repository is available for
    public review and contribution under the terms of the license.

    <https://github.com/wchristian/capture-tiny>

      git clone https://github.com/wchristian/capture-tiny

AUTHORS
    *   Christian Walde <mithaldu@yahoo.de>

    *   David Golden <dagolden@cpan.org>

COPYRIGHT AND LICENSE
    This software is Copyright (c) 2009 by David Golden.

    This is free software, licensed under:

      The Apache License, Version 2.0, January 2004