From Code to Community: Sponsoring The Perl and Raku Conference 2025 Learn more

package Data::Unixish;
use 5.010001;
use strict;
use Exporter qw(import);
use Package::Util::Lite qw(package_exists);
our $AUTHORITY = 'cpan:PERLANCAR'; # AUTHORITY
our $DATE = '2025-02-24'; # DATE
our $DIST = 'Data-Unixish'; # DIST
our $VERSION = '1.574'; # VERSION
our @EXPORT_OK =
qw(
aduxa cduxa fduxa lduxa
aduxc cduxc fduxc lduxc
aduxf cduxf fduxf lduxf
aduxl cduxl fduxl lduxl
siduxs
aiduxa aiduxl
liduxa liduxl
);
our %EXPORT_TAGS = (
all => [
qw/
aduxa cduxa fduxa lduxa
aduxc cduxc fduxc lduxc
aduxf cduxf fduxf lduxf
aduxl cduxl fduxl lduxl
siduxs
aiduxa aiduxl
liduxa liduxl
/],
);
sub _dux {
my $accepts = shift;
my $returns = shift;
my $func = shift;
my %args;
my ($icallback, $ocallback);
if ($accepts eq 'c') {
$icallback = shift;
}
if ($returns eq 'c') {
$ocallback = shift;
}
if ($accepts eq 'f') {
require Tie::File;
my @in;
tie @in, "Tie::File", @_;
$args{in} = \@in;
} elsif ($accepts eq 'c') {
require Tie::Simple;
my @in;
my @els;
my $elcount = 0;
tie(@in, "Tie::Simple", undef,
FETCHSIZE => sub {
my $data = shift; # from Tie::Simple
my @res = $icallback->();
$elcount += @res;
push @els, @res;
#say "D: res=[".join(",", @res), "], elcount=$elcount";
$elcount;
},
FETCH => sub {
my $data = shift; # from Tie::Simple
shift @els;
}
);
$args{in} = \@in;
} elsif ($accepts eq 'l') {
$args{in} = \@_;
} elsif ($accepts eq 'a') {
$args{in} = $_[0];
} else {
die "Invalid accepts, must be a|c|f|l";
}
if (ref($func) eq 'ARRAY') {
$args{$_} = $func->[1]{$_} for grep {/\A\w+\z/} keys %{$func->[1]};
$func = $func->[0];
}
my $pkg = "Data::Unixish::$func";
load $pkg unless package_exists($pkg);
my $funcleaf = $func; $funcleaf =~ s/.+:://;
my $funcname = "Data::Unixish::$func\::$funcleaf";
die "Subroutine &$funcname not defined" unless defined &$funcname;
my @out;
my $kidfh;
my $pid;
if ($returns eq 'c') {
require Tie::Simple;
tie @out, "Tie::Simple", undef,
PUSH => sub {
my $data = shift; # from Tie::Simple
$ocallback->($_) for @_;
};
$args{out} = \@out;
} elsif ($returns eq 'f') {
require Tie::Simple;
tie @out, "Tie::Simple", undef,
PUSH => sub {
my $data = shift; # from Tie::Simple
for my $item (@_) {
$item .= "\n" unless $item =~ /\n\z/;
print STDOUT $item;
}
};
$args{out} = \@out;
$pid = open $kidfh, "-|";
defined $pid or die "Can't fork: $!";
} else {
$args{out} = \@out;
}
unless ($pid) {
no strict 'refs'; ## no critic: TestingAndDebugging::ProhibitNoStrict
my $res = $funcname->(%args);
die "Dux function $funcname failed: $res->[0] - $res->[1]"
unless $res->[0] == 200;
}
if ($returns eq 'l') {
if (wantarray) {
return @out;
} else {
return $out[0];
}
} elsif ($returns eq 'a') {
return \@out;
} elsif ($returns eq 'c') {
return;
} elsif ($returns eq 'f') {
if ($pid) {
return $kidfh;
} else {
exit;
}
} else {
die "Invalid returns, must be a|c|f|l";
}
}
sub aduxa { _dux('a', 'a', @_) }
sub cduxa { _dux('c', 'a', @_) }
sub fduxa { _dux('f', 'a', @_) }
sub lduxa { _dux('l', 'a', @_) }
sub aduxc { _dux('a', 'c', @_) }
sub cduxc { _dux('c', 'c', @_) }
sub fduxc { _dux('f', 'c', @_) }
sub lduxc { _dux('l', 'c', @_) }
sub aduxf { _dux('a', 'f', @_) }
sub cduxf { _dux('c', 'f', @_) }
sub fduxf { _dux('f', 'f', @_) }
sub lduxf { _dux('l', 'f', @_) }
sub aduxl { _dux('a', 'l', @_) }
sub cduxl { _dux('c', 'l', @_) }
sub fduxl { _dux('f', 'l', @_) }
sub lduxl { _dux('l', 'l', @_) }
sub _idux {
my $accepts = shift;
my $returns = shift;
my $func = shift;
my %args;
my @items;
if ($accepts eq 's') {
@items = ($_[0]);
} elsif ($accepts eq 'a') {
@items = @{ $_[0] };
} elsif ($accepts eq 'l') {
@items = @_;
} else {
die "Invalid accepts, must be a|l|s";
}
if (ref($func) eq 'ARRAY') {
$args{$_} = $func->[1]{$_} for keys %{$func->[1]};
$func = $func->[0];
}
my $pkg = "Data::Unixish::$func";
load $pkg unless package_exists($pkg);
my $funcleaf = $func; $funcleaf =~ s/.+:://;
my $funcname_i = "Data::Unixish::$func\::_${funcleaf}_item";
die "Subroutine &$funcname_i not defined" unless defined &$funcname_i;
my $funcname_b = "Data::Unixish::$func\::_${funcleaf}_begin";
my $funcname_e = "Data::Unixish::$func\::_${funcleaf}_end";
my @res;
{
no strict 'refs'; ## no critic: TestingAndDebugging::ProhibitNoStrict
my @bres; @bres = $funcname_b->(\%args) if defined &$funcname_b;
for (@items) {
push @res, $funcname_i->($_, \%args);
}
$funcname_e->(\%args, @bres) if defined &$funcname_e;
}
if ($returns eq 's') {
return $res[0];
} elsif ($returns eq 'l') {
if (wantarray) {
return @res;
} else {
return $res[0];
}
} elsif ($returns eq 'a') {
return \@res;
} else {
die "Invalid returns, must be a|l|s";
}
}
sub siduxs { _idux('s', 's', @_) }
sub aiduxa { _idux('a', 'a', @_) }
sub aiduxl { _idux('a', 'l', @_) }
sub liduxa { _idux('l', 'a', @_) }
sub liduxl { _idux('l', 'l', @_) }
1;
# ABSTRACT: Implementation for Unixish, a data transformation framework
__END__
=pod
=encoding UTF-8
=head1 NAME
Data::Unixish - Implementation for Unixish, a data transformation framework
=head1 VERSION
This document describes version 1.574 of Data::Unixish (from Perl distribution Data-Unixish), released on 2025-02-24.
=head1 SYNOPSIS
# the a/f/l/c prefix determines whether function accepts
# arrayref/file(handle)/list/callback as input. the a/f/l/c suffix determines
# whether function returns an array, a list, a filehandle, or calls a callback.
# If filehandle is chosen as output, a child process is forked to process input
# as requested.
use Data::Unixish qw(
aduxa cduxa fduxa lduxa
aduxc cduxc fduxc lduxc
aduxf cduxf fduxf lduxf
aduxl cduxl fduxl lduxl
siduxs
); # or you can use :all to export all functions
# apply function, without argument
my @out = lduxl('sort', 7, 2, 4, 1); # => (1, 2, 4, 7)
my $out = lduxa('uc', "a", "b", "c"); # => ["A", "B", "C"]
my $res = fduxl('wc', "file.txt"); # => "12\n234\n2093" # like wc's output
# apply function, with some arguments
my $fh = fduxf([trunc => {width=>80, ansi=>1, mb=>1}], \*STDIN);
say while <$fh>;
# apply function to a single item, function must be itemfunc
my $res = duxitem(, $item);
# apply function to multiple items, function must be itemfunc
my @res = aduxitem(, $item1, $item2, $item3);
=head1 DESCRIPTION
This distribution implements L<Unixish>, a data transformation framework
inspired by Unix toolbox philosophy.
=head1 FUNCTIONS
The functions are not exported by default. They can be exported individually or
altogether using export tag C<:all>.
=head2 aduxa($func, \@input) => ARRAYREF
=head2 aduxc($func, $callback, \@input)
=head2 aduxf($func, \@input) => FILEHANDLE
=head2 aduxl($func, \@input) => LIST (OR SCALAR)
The C<adux*> functions accept an arrayref as input. C<$func> is a string
containing dux function name (if no arguments to the dux function is to be
supplied), or C<< [$func, \%args] >> to supply arguments to the dux function.
Dux function name corresponds to module names C<Data::Unixish::NAME> without the
prefix.
The C<*duxc> functions will call the callback repeatedly with every output item.
The C<*duxf> functions returns filehandle immediately. A child process is
forked, and dux function is run in the child process. You read output as lines
from the returned filehandle. (Currently not yet supported on Windows due to no
support for open '-|').
The C<*duxl> functions returns result as list. It can be evaluated in scalar to
return only the first element of the list. However, the whole list will be
calculated first. Use C<*duxf> for streaming interface.
=head2 cduxa($func, $icallback) => ARRAYREF
=head2 cduxc($func, $icallback, $ocallback)
=head2 cduxf($func, $icallback) => FILEHANDLE
=head2 cduxl($func, $icallback) => LIST (OR SCALAR)
The C<cdux*> functions accepts a callback (C<$icallback>) to get input elements
from. Input callback function should return a list of one or more elements, or
an empty list to signal end of stream.
An example:
cduxa($func, sub {
state $a = [1,2,3,4];
if (@$a) {
return shift(@$a);
} else {
return ();
}
});
=head2 fduxa($func, $file_or_handle, @args) => ARRAYREF
=head2 fduxc($func, $callback, $file_or_handle, @args)
=head2 fduxf($func, $file_or_handle, @args) => FILEHANDLE
=head2 fduxl($func, $file_or_handle, @args) => LIST
The C<fdux*> functions accepts filename or filehandle. C<@args> is optional and
will be passed to L<Tie::File>. Currently not yet supported on Windows.
=head2 lduxa($func, @input) => ARRAYREF
=head2 lduxc($func, $callback, @input)
=head2 lduxf($func, @input) => FILEHANDLE
=head2 lduxl($func, @input) => LIST
The C<ldux*> functions accepts list as input.
=head2 siduxs($func, $item) => $res
=head2 aiduxa($func, \@items) => ARRAYREF
=head2 aiduxl($func, \@items) => LIST
=head2 liduxa($func, @items) => ARRAYREF
=head2 liduxl($func, @items) => LIST
The C<*idux*> functions apply dux function on single item(s). Only dux functions
tagged with C<itemfunc> can be used. These functions can operate on a single
item and return a single result. Examples of itemfunc functions are C<uc>,
C<lc>, C<sprintf>. Examples of non-itemfunc functions are C<head>, C<tail>,
C<wc>.
The C<*idux*> functions can be useful if you want to call a dux function from
another dux function for each item. For example, see
C<Data::Unixish::condapply>.
=head1 FAQ
=head2 I'm getting "Use of uninitialized value in push at lib/Data/Unixish/XXX.pm line XX." messages!
This looks like a bug in perl 5.10.1 or earlier. Try upgrading to perl 5.12 or
later.
=head2 How do I use the diamond operator as input?
You can use L<Tie::Diamond>, e.g.:
use Tie::Diamond;
tie my(@in), "Tie::Diamond";
my $out = aduxa($func, \@in);
Also see the L<dux> command-line utility in the L<App::dux> distribution which
allows you to access dux function from the command-line.
=head1 HOMEPAGE
Please visit the project's homepage at L<https://metacpan.org/release/Data-Unixish>.
=head1 SOURCE
=head1 SEE ALSO
L<Unixish>
L<dux> script in L<App::dux>
=head1 AUTHOR
perlancar <perlancar@cpan.org>
=head1 CONTRIBUTORS
=for stopwords Mohammad S Anwar Steven Haryanto Toby Inkster
=over 4
=item *
Mohammad S Anwar <mohammad.anwar@yahoo.com>
=item *
Steven Haryanto <stevenharyanto@gmail.com>
=item *
Toby Inkster <mail@tobyinkster.co.uk>
=back
=head1 CONTRIBUTING
To contribute, you can send patches by email/via RT, or send pull requests on
GitHub.
Most of the time, you don't need to build the distribution yourself. You can
simply modify the code, then test via:
% prove -l
If you want to build the distribution (e.g. to try to install it locally on your
system), you can install L<Dist::Zilla>,
L<Dist::Zilla::PluginBundle::Author::PERLANCAR>,
L<Pod::Weaver::PluginBundle::Author::PERLANCAR>, and sometimes one or two other
Dist::Zilla- and/or Pod::Weaver plugins. Any additional steps required beyond
that are considered a bug and can be reported to me.
=head1 COPYRIGHT AND LICENSE
This software is copyright (c) 2025 by perlancar <perlancar@cpan.org>.
This is free software; you can redistribute it and/or modify it under
the same terms as the Perl 5 programming language system itself.
=head1 BUGS
Please report any bugs or feature requests on the bugtracker website L<https://rt.cpan.org/Public/Dist/Display.html?Name=Data-Unixish>
When submitting a bug or request, please include a test-file or a
patch to an existing test-file that illustrates the bug or desired
feature.
=cut