sv
[moreutils.git] / combine
blob99a4de6125bfdcec457529419eb700c0707e9fe6
1 #!/usr/bin/perl
3 =head1 NAME
5 combine - combine sets of lines from two files using boolean operations
7 =head1 SYNOPSIS
9 combine file1 and file2
11 combine file1 not file2
13 combine file1 or file2
15 combine file1 xor file2
17 _ file1 and file2 _
19 _ file1 not file2 _
21 _ file1 or file2 _
23 _ file1 xor file2 _
25 =head1 DESCRIPTION
27 B<combine> combines the lines in two files. Depending on the boolean
28 operation specified, the contents will be combined in different ways:
30 =over 4
32 =item and
34 Outputs lines that are in file1 if they are also present in file2.
36 =item not
38 Outputs lines that are in file1 but not in file2.
40 =item or
42 Outputs lines that are in file1 or file2.
44 =item xor
46 Outputs lines that are in either file1 or file2, but not in both files.
48 =back
50 "-" can be specified for either file to read stdin for that file.
52 The input files need not be sorted, and the lines are output in the order
53 they occur in file1 (followed by the order they occur in file2 for the two
54 "or" operations). Bear in mind that this means that the operations are not
55 commutative; "a and b" will not necessarily be the same as "b and a". To
56 obtain commutative behavior sort and uniq the result.
58 Note that this program can be installed as "_" to allow for the syntactic
59 sugar shown in the latter half of the synopsis (similar to the test/[
60 command). It is not currently installed as "_" by default, but you can
61 alias it to that if you like.
63 =head1 SEE ALSO
65 join(1)
67 =head1 AUTHOR
69 Copyright 2006 by Joey Hess <joey@kitenet.net>
71 Licensed under the GNU GPL.
73 =cut
75 use warnings;
76 use strict;
78 sub filemap {
79 my $file=shift;
80 my $sub=shift;
82 open (IN, $file) || die "$file: $!\n";
83 while (<IN>) {
84 chomp;
85 $sub->();
87 close IN;
90 sub hashify {
91 my $file=shift;
93 my %seen;
94 filemap $file, sub { $seen{$_}++ };
95 return \%seen;
98 sub compare_or {
99 my ($file1, $file2) = @_;
101 filemap $file1, sub { print "$_\n" };
102 filemap $file2, sub { print "$_\n" };
105 sub compare_xor {
106 my ($file1, $file2) = @_;
108 my (@lines2, %seen2);
109 filemap $file2,
110 sub {
111 push @lines2, $_;
112 $seen2{$_} = 1;
115 # Print all lines in file1 that are not in file2,
116 # and mark lines that are in both files by setting
117 # their value in %seen2 to 0.
118 filemap $file1,
119 sub {
120 if (exists $seen2{$_}) {
121 $seen2{$_} = 0;
123 else {
124 print "$_\n";
128 # Print all lines that are in file2 but not in file1.
129 # The value of these lines in seen2 is set to 1.
130 foreach (@lines2) {
131 print "$_\n" if $seen2{$_};
135 sub compare_not {
136 my ($file1, $file2) = @_;
138 my $seen=hashify($file2);
139 filemap $file1, sub { print "$_\n" unless $seen->{$_} };
142 sub compare_and {
143 my ($file1, $file2) = @_;
145 my $seen=hashify($file2);
146 filemap $file1, sub { print "$_\n" if $seen->{$_} };
149 if (@ARGV >= 4 && $ARGV[3] eq "_") {
150 delete $ARGV[3];
153 if (@ARGV != 3) {
154 die "Usage: combine file1 OP file2\n";
157 my $file1=shift;
158 my $op=lc shift;
159 my $file2=shift;
161 if ($::{"compare_$op"}) {
162 no strict 'refs';
163 "compare_$op"->($file1, $file2);
165 else {
166 die "unknown operation, $op\n";