2 # acromang tries to solve acrostics, and is released
3 # under the terms of the GPL version 2, or any later version, at your
4 # option. See the file README and COPYING for more information.
5 # Copyright 2008 by Don Armstrong <don@donarmstrong.com>.
6 # $Id: perl_script 1153 2008-04-08 00:04:20Z don $
17 acromang - try to solve acrostics
21 acromang m??s??c??o [options]
24 --debug, -d debugging level (Default 0)
25 --help, -h display this help
26 --man, -m display manual
34 Default wordlist, should be gigantic; defaults to
37 =item B<--database-dir>
39 Directory for storing databases; defaults to ~/.acromang
43 Debug verbosity. (Default 0)
47 Display brief useage information.
66 use Params::Validate qw(:types validate_with);
69 use MLDBM qw(DB_File Storable);
70 use Fcntl qw/O_RDWR O_RDONLY O_CREAT O_TRUNC/;
71 # Use portable Storable images
72 $MLDBM::DumpMeth=q(portable);
76 use List::Util qw(min max);
78 my %options = (debug => 0,
81 wordlist => '/usr/share/dict/words',
82 database_dir => User->Home.'/.acromang/',
87 my @puzzle_types = qw(acrostic anagram substitution);
92 'wordlist=s','database_dir|database-dir=s',
94 'debug|d+','help|h|?','man|m');
96 pod2usage() if $options{help};
97 pod2usage({verbose=>2}) if $options{man};
99 $DEBUG = $options{debug};
102 if (not grep {exists $options{$_}} @puzzle_types) {
103 $options{acrostic} = 1
106 if (1 != grep {exists $options{$_}} @puzzle_types) {
107 push @USAGE_ERRORS, "You must pass exactly one of ".join(', ',map {"--$_ "} @puzzle_types);
110 pod2usage(join("\n",@USAGE_ERRORS)) if @USAGE_ERRORS;
116 my @words = map {s/[^A-Za-z\?]//g; length $_?$_:();} split /[\s\n]+/, $text;
118 print STDERR "There were no words in the input\n";
122 my $database = update_and_load_database($options{database_dir},$options{wordlist});
125 my @l_o = ('a'..'z');
127 @l_o{@l_o} = (0 .. 25);
130 # run through and use the database
132 # for now, concentrate on finding single word solutions
133 if ($options{acrostic}) {
134 foreach my $word (@words) {
135 my $length = length($word);
136 my @let = split //, $word;
138 for my $p (0..$#let) {
139 next if $let[$p] eq '?';
142 my @allowable_words = find_allowable_words(database => $database,position => \%pos,length => $length);
143 if (@allowable_words > 200 and not $options{show_all}) {
144 print STDERR "Only showing 200 of the " . @allowable_words . " possible words\n";
145 @allowable_words = @allowable_words[0..199];
147 while ((@allowable_words == 0 or $options{multiword}) and
148 (@allowable_words < 200 or $options{show_all})
150 # try for multiple word solutions, start with the longest
153 # try to split the number of known letters in half, and
154 # start increasing and decreasing in both directions
156 # don't attempt to split each part into bits unless there
159 # avoid searching for words when we only have a length and
160 # no position information
162 print map {$_,qq(\n)} sort @allowable_words;
165 elsif ($options{anagram}) {
166 foreach my $word (@words) {
167 my $length = length($word);
168 my $sorted_word = join ('', sort split //, $word);
169 my @allowable_words = find_allowable_words(database => $database,
170 anagram => $sorted_word,
172 print map {$_,qq(\n)} sort @allowable_words;
176 sub find_allowable_words {
177 my %param = validate_with(params => \@_,
178 spec => {database => {type => HASHREF,
180 position => {type => HASHREF,
183 length => {type => SCALAR,
186 substitution => {type => SCALAR,
189 rotation => {type => SCALAR,
192 anagram => {type => SCALAR,
198 my $database = $param{database};
200 if (exists $param{length} and defined $param{length} and $param{length} > 0) {
203 if (exists $param{position} and defined $param{position}) {
204 $must_match += scalar keys %{$param{position}};
206 if (exists $param{anagram} and defined $param{anagram}) {
210 if ($must_match <= 0) {
211 die "There must be something to try matching against";
214 if (exists $param{length} and defined $param{length} and $param{length} > 0) {
215 for my $word (@{$database->{length}{$param{length}}}) {
219 if (exists $param{position} and defined $param{position}) {
220 for my $position (keys %{$param{position}}) {
221 for my $word (@{$database->{position}{$position . ' ' . $param{position}{$position}}}) {
226 if (exists $param{anagram} and defined $param{anagram}) {
227 for my $word (@{$database->{anagram}{$param{anagram}}}) {
231 return grep {$words{$_} >= $must_match} keys %words;
235 sub update_and_load_database {
236 my ($dir,$wordlist) = @_;
237 # check to see if the wordlist is newer than our database
238 my $update_required = 0;
239 my $wordlist_time = 1;
240 if (! -r $wordlist) {
241 die "Word list $wordlist doesn't exist or isn't readable";
243 my $wordlist_stat = stat($wordlist);
244 $wordlist_time = max($wordlist_time,$wordlist_stat->mtime);
245 for my $db (qw(length position)) {
246 if (! -e "$dir/db_${db}") {
247 $update_required = 1;
249 elsif (stat($wordlist)->mtime < $wordlist_time) {
250 $update_required = 1;
254 if ($update_required) {
257 mkdir($dir) or die "Unable to create directory $dir: $!";
259 for my $db (qw(length position substitution rotation anagram)) {
260 $database->{$db} = {};
261 $disk_db->{$db} = {};
262 tie %{$disk_db->{$db}}, MLDBM => "$dir/db_${db}",
263 O_RDWR|O_CREAT|O_TRUNC, 0666
264 or die "Unable to open/create $dir/db_${db}: $!";
266 my $wordlist_fh = IO::File->new($wordlist,'r') or
267 die "Unable to open $wordlist for reading: $!";
269 while (<$wordlist_fh>) {
271 next unless length $_;
273 next if exists $seen_words{$word};
274 $seen_words{$word} = 1;
275 if ((keys %seen_words) % 100 == 0) {
276 print STDERR "Handled ".(keys %seen_words) . "words, on $word\n";
278 my @l = split //, $word;
279 my $l = length($word);
281 # position from the beginning
282 my $temp = $database->{position}{"$p $l[$p]"};
283 $temp = [] if not defined $temp;
284 push @{$temp}, $word;
285 $database->{position}{"$p $l[$p]"} = $temp;
286 # this is the position from the end
287 $temp = $database->{position}{$p-$l . " " . $l[$p-$l]};
288 $temp = [] if not defined $temp;
289 push @{$temp}, $word;
290 $database->{position}{$p-$l . " " . $l[$p-$l]} = $temp;
292 my $temp = $database->{length}{$l};
293 $temp = [] if not defined $temp;
294 push @{$temp}, $word;
295 $database->{length}{$l} = $temp;
296 # this is the substitution database
297 # in it, hello and giddy both become abccd
300 if (exists $uc{$_}) {
308 @s_map{@uc_order} = ('a'..'z')[0..$#uc_order] if @uc_order;
309 my $mapped_word = join('',map {$s_map{$_}} @l);
310 $temp = $database->{substitution}{$mapped_word};
311 $temp = [] if not defined $temp;
312 push @{$temp}, $word;
313 $database->{substitution}{$mapped_word} = $temp;
315 # this is the rotation database (for ceaser cyphers)
316 # in it, hello becomes axeeh, giddy acxxs
318 my $index = $l_o{$fl};
319 $mapped_word = join('',map {$l_o[($l_o{$_} - $index + 26) % 26]} @l);
320 $temp = $database->{rotation}{$mapped_word};
321 $temp = [] if not defined $temp;
322 push @{$temp}, $word;
323 $database->{rotation}{$mapped_word} = $temp;
325 # this is the anagram database
326 # in it, hello becomes ehllo, giddy ddgiy
327 $mapped_word = join('',sort @l);
328 $temp = $database->{anagram}{$mapped_word};
329 $temp = [] if not defined $temp;
330 push @{$temp}, $word;
331 $database->{anagram}{$mapped_word} = $temp;
333 for my $key1 (keys %{$database}) {
334 for my $key2 (keys %{$database->{$key1}}) {
335 $disk_db->{$key1}{$key2} = $database->{$key1}{$key2};
340 for my $db (qw(length position)) {
341 $database->{$db} = {};
342 tie %{$database->{$db}}, MLDBM => "$dir/db_${db}",
344 or die "Unable to open $dir/db_${db}: $!";