Provided by: librunning-commentary-perl_0.000005-1_all 
NAME
Running::Commentary - call "system" cleanly, with tracking messages
VERSION
This document describes Running::Commentary version 0.000005
SYNOPSIS
use Running::Commentary;
# Set a lexically scoped flag for all subsequent calls...
# (No announcements, if this flag set)
run_with -nomessage if !$verbose;
# Act like system(), only louder and cleaner...
run 'Resetting' => "rm -rf '$ROOT_DIR'"
or die "Couldn't reset";
# Act like system(), but croak() if the command fails...
run -critical, 'Building Makefile' => 'perl Makefile.PL';
# Calls to run() may be nested, to allow subtasks to be tracked...
run 'Running tests'
=> sub {
for my $file (@profiled_files) {
push @profiles, "$NAMING_ROOT/$file.out";
local $ENV{NYTPROF} = "file=$profiles[-1]";
run -nooutput, "Testing $file"
=> "perl -d:NYTProf $profiled_path/$file >& /dev/null";
}
};
DESCRIPTION
This module provides a single subroutine: "run()" which is designed to be a more
informative and less error-prone replacement for the built-in "system()".
It also provides a compile-time keyword: "run_with" with which you can set lexically
scoped default options for "run()".
INTERFACE
"run $MESSAGE => $SYSTEM_CMD;"
This acts like "system $SYSTEM_CMD", except that it returns true on success and false
on failure, and it announces what it's doing. For example:
run 'Resetting directories' => "rm -rf @STD_DIRS"
...would first output:
Resetting directories...
...then execute the system command, and finish the message:
Resetting directories...done
If the command failed for some reason, the completion would reflect the problem:
Resetting directories...
rm: tets: No such file or directory
Resetting directories...exited with value 1
Or:
Resetting directories...failed to execute: No such file or directory
"run $MESSAGE => sub {...};"
This form of the command expects a subroutine reference, rather than a string, as its
second argument. Once again it prints the tracking message, then executes the
subroutine, then prints the outcome.
The subroutine is run inside an "eval" block, so any exceptions it throws are
intercepted, and reported as the outcome at the end of the tracking message. To have
exceptions inside the subroutine propagate back out of the call to "run()", use the
"-critical" option (see below).
For example:
run 'Printing your data' => sub {
for my $datum (@data) {
say " $datum->{key}: $datum->{value}";
}
}
Would output:
Printing your data...
Name: Fred
Age: 28
Score: 87
Printing your data...done
You can also nest calls to "run()" using this form. For example:
run 'Running your request' => sub {
for my $cmd (split /\n/, $request) {
run "Running '$cmd'" => $cmd;
}
}
Would produce:
Running your request...
Running 'rm source'...done
Running 'rebuild_files'...done
Running 'make test'.......done
Running your request...done
"run $SYSTEM_CMD;"
"run sub {...};"
When called without a message, "run()" simply executes the system command or
subroutine without printing any kind of progress message. In other words, it merely
acts as a (quietly) better "system()".
"run_with @OPTIONS;"
The "run_with" keyword can be called with any of the options available to "run()" (see
"OPTIONS"). It takes the options given to it and makes them the default arguments to
"run()" for the remainder of the current lexical scope.
For example, to cause any subsequent failed command to throw an exception...
{
run_with -critical;
run "loading" => $LOAD_CMD;
run "checking" => $CHECK_CMD;
run "installing" => $INSTALL_CMD;
run "cleaning up" => $CLEANUP_CMD;
}
...or to silence message printing on request:
{
run_with -nomessage if $opt{-quiet};
run "loading" => $LOAD_CMD;
run "checking" => $CHECK_CMD;
run "installing" => $INSTALL_CMD;
run "cleaning up" => $CLEANUP_CMD;
}
Note that "run_with" is a compile-time keyword, not a subroutine, so it should only be
called as a statement (i.e. in void context).
OPTIONS
The following options can be included anywhere in the argument list of a call to "run()"
or "run_with".
"-nomessage"
Run the command without printing the tracking message. Normally used as a conditional
lexical option:
run_with -nomessage if $opt{quiet};
The output of the actual system command is still printed (unless "-nooutput" or
"-silent" is also specified)
"-showmessage"
Run the command, printing the tracking message. Useful to turn message printing back
on inside a scope where "-nomessage" is already in effect.
"-nooutput"
Run the command without echoing any of its output. The tracking message is still
printed (unless "-nomessage" or "-silent" is also specified)
"-showoutput"
Run the command, echoing any output. Useful to turn command echoing back on inside a
scope where "-nooutput" is already in effect.
"-silent"
Identical to: "-nomessage, -nooutput"
"-showall"
Identical to: "-showmessage, -showoutput". Useful to override "-silent" in a nested
scope.
"-critical"
Normally, if a call to "run()" fails, it simply returns "undef". However, if the
"-critical" option is specified, any call to "run" that fails will immediately throw
an exception.
"-nocritical"
Revert "run()" to returning "undef" on failure. Useful to override "-critical" in a
nested scope.
"-dry"
Instead of executing the specified system command, just print it out. Useful for dry
runs during development and testing.
"-colour => \%COLOUR_SPEC"
Specify the colours to be used for messages and output. Colours are specified as the
values of the hash, with the keys indicating what purpose each colour is to be used
for. For example:
run_with -colour => {
MESSAGE => 'white', # Colour for tracking messages
DONE => 'bold cyan', # Colour for success messages
FAILED => 'yellow on_red', # Colour for failure messages
OUTPUT => 'clear' # Colour for command output
};
The colour specifications must be single strings, which are split on whitespace and
then passed to the "Term::ANSIColor" module. If that module is not available, this
option is silently ignored.
This option may also be spelled "-color".
"-nocolour"
Print all messages and output without any special colours.
This option may also be spelled "-nocolor".
ERROR HANDLING
On failure "run()" normally either returns "undef" or throws an exception (if "-critical"
is specified).
However, "Running::Commentary" incorporates the "Lexical::Failure" module, so you can also
request other failure responses for any particular scope, by passing a named argument when
loading the module:
# Report errors by confess()-ing...
use Running::Commentary fail => 'confess';
# Report errors by returning a failure object...
use Running::Commentary fail => 'failobj';
# Report errors by setting a flag variable...
use Running::Commentary fail => \$error;
# Report errors by calling a subroutine...
use Running::Commentary fail => \&error_handler;
For details of the available options, see the documentation of "Lexical::Failure".
DIAGNOSTICS
"Bad argument to 'use Running::Commentary'"
The module accepts only one named argument:
use Running::Commentary 'fail' => $fail_mode;
(see "ERROR HANDLING").
You apparently passed it something else. Or perhaps misspelt 'fail'?
"Useless call to run() with no command"
"run()" expects at least one argument (apart from any configuration options); namely,
something to execute. That can be either a string containing a system command, or
else a subroutine reference.
You didn't give it either of those, so the call to "run()" was superfluous.
Or, possibly, you wanted "run_with" instead.
CONFIGURATION AND ENVIRONMENT
Running::Commentary requires no configuration files or environment variables.
DEPENDENCIES
This module requires Perl v5.14 or later.
It also requires the modules: "Lexical::Failure", and "Keyword::Simple".
INCOMPATIBILITIES
None reported.
BUGS AND LIMITATIONS
No bugs have been reported.
Please report any bugs or feature requests to "bug-running-commentary@rt.cpan.org", or
through the web interface at <http://rt.cpan.org>.
AUTHOR
Damian Conway "<DCONWAY@CPAN.org>"
LICENCE AND COPYRIGHT
Copyright (c) 2012, Damian Conway "<DCONWAY@CPAN.org>". All rights reserved.
This module is free software; you can redistribute it and/or modify it under the same
terms as Perl itself. See perlartistic.
DISCLAIMER OF WARRANTY
BECAUSE THIS SOFTWARE IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE SOFTWARE,
TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE
COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE SOFTWARE "AS IS" WITHOUT WARRANTY OF
ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO
THE QUALITY AND PERFORMANCE OF THE SOFTWARE IS WITH YOU. SHOULD THE SOFTWARE PROVE
DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR, OR CORRECTION.
IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT
HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE SOFTWARE AS PERMITTED BY
THE ABOVE LICENCE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL,
INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
SOFTWARE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR
LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE SOFTWARE TO OPERATE WITH ANY
OTHER SOFTWARE), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
SUCH DAMAGES.