最後のは動作確認。% tar zxvf sufary-2.3.8.tar.gz % cd sufary-2.3.8 % libtoolize --force % ./configure % make % src/mkary README; src/sass a README
(ref. http://gimmix.org/max/article.php?id=311)SED=sed ;export SED
% ./configure --host=i686-apple-macosx % make % src/mkary README; src/sass a README
% cat a.txt foo フー bar バー hoge ほげ uhyo うひょ % mkary -l a.txt
#!/usr/bin/perl use strict; use warnings; use SUFARY; my $fn = shift @ARGV; my $sa = SUFARY->new($fn); # convert (demo) for (my $i = 0; $i < $sa->{arraysize}; $i++) { my $str = id2str($sa, $i); my ($key, $info) = split(' ', $str, 2); print qq(id:$i => key:"$key", info:"$info"\n); my $id = str2id($sa, $key." "); print qq(key:"$key" => id:$id\n); } # error my $str = id2str($sa, 100); print qq(error1: "$str"\n); # empty my $id = str2id($sa, "dummy"); print qq(error2: $id\n); # -1 sub id2str { my ($sa, $id) = @_; return "" unless 0 <= $id and $id < $sa->{arraysize}; my $str = $sa->get_line($sa->get_position($id)); chomp $str; return $str; } sub str2id { my ($sa, $key) = @_; my ($left, $right) = $sa->range_search($key); return -1 if not defined $left or $left != $right; return $left; }
fooとbarの順番が入れ替わっているのは、suffix array の性質上アルファベティカルにソートされてるため。% ./idstr.pl a.txt id:0 => key:"bar", info:"バー" key:"bar" => id:0 id:1 => key:"foo", info:"フー" key:"foo" => id:1 id:2 => key:"hoge", info:"ほげ" key:"hoge" => id:2 id:3 => key:"uhyo", info:"うひょ" key:"uhyo" => id:3 error1: "" error2: -1
#!/usr/bin/perl # id2str.pl use strict; use warnings; use SUFARY; my ($id, $fn) = @ARGV; my $sa = SUFARY->new($fn); print id2str($sa, $id), "\n"; sub id2str { my ($sa, $id) = @_; return "" unless 0 <= $id and $id < $sa->{arraysize}; my $str = $sa->get_line($sa->get_position($id)); chomp $str; return $str; }
#!/usr/bin/perl # str2id.pl use strict; use warnings; use SUFARY; my ($key, $fn) = @ARGV; my $sa = SUFARY->new($fn); print str2id($sa, $key." "), "\n"; sub str2id { my ($sa, $key) = @_; my ($left, $right) = $sa->range_search($key); return -1 if not defined $left or $left != $right; return $left; }
% ./id2str.pl 0 a.txt bar バー % ./id2str.pl 3 a.txt uhyo うひょ % ./str2id.pl foo a.txt 1 % ./str2id.pl uhyo a.txt 3
#!/usr/bin/perl use strict; use warnings; my $fn = shift @ARGV; my $data = load_data($fn); store_data($fn, $data); sub load_data { my ($fn) = @_; require Storable; scalar eval { Storable::retrieve($fn) } || []; } sub store_data { my($fn, $data) = @_; require Storable; Storable::nstore($data, $fn); }