[28649] | 1 | package CPULoadTextPlugin;
|
---|
| 2 |
|
---|
| 3 | use TextPlugin;
|
---|
| 4 |
|
---|
| 5 | use strict;
|
---|
| 6 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
| 7 | no strict 'subs';
|
---|
| 8 |
|
---|
| 9 | use Crypt::Blowfish_PP;
|
---|
| 10 | use Kea;
|
---|
| 11 | use Lingua::EN::Syllable;
|
---|
| 12 |
|
---|
| 13 | sub BEGIN {
|
---|
| 14 | @CPULoadTextPlugin::ISA = ('TextPlugin');
|
---|
| 15 | }
|
---|
| 16 |
|
---|
| 17 | our $cpu_load_list = [ { 'name' => "none",
|
---|
| 18 | 'desc' => "Almost no processing - all dependent on IO" },
|
---|
| 19 | { 'name' => "low",
|
---|
| 20 | 'desc' => "A little processing - building lexicon" },
|
---|
| 21 | { 'name' => "medium",
|
---|
| 22 | 'desc' => "Some processing..." },
|
---|
| 23 | { 'name' => "high",
|
---|
| 24 | 'desc' => "Data mining and part of speech tagging" }
|
---|
| 25 | ];
|
---|
| 26 |
|
---|
| 27 | my $arguments = [ { 'name' => "process_exp",
|
---|
| 28 | 'desc' => "{BasePlugin.process_exp}",
|
---|
| 29 | 'type' => "regexp",
|
---|
| 30 | 'deft' => &get_default_process_exp(),
|
---|
| 31 | 'reqd' => "no" },
|
---|
| 32 | { 'name' => "cpu_load",
|
---|
| 33 | 'desc' => "",
|
---|
| 34 | 'type' => "enum",
|
---|
| 35 | 'deft' => "auto",
|
---|
| 36 | 'list' => $cpu_load_list,
|
---|
| 37 | 'reqd' => "no" },
|
---|
| 38 | { 'name' => 'debug',
|
---|
| 39 | 'desc' => '',
|
---|
| 40 | 'type' => 'flag',
|
---|
| 41 | 'reqd' => 'no',
|
---|
| 42 | 'deft' => '0',
|
---|
| 43 | 'hiddengli' => 'no'}
|
---|
| 44 | ];
|
---|
| 45 |
|
---|
| 46 | my $options = { 'name' => "CPULoadTextPlugin",
|
---|
| 47 | 'desc' => "TextPlugin allowing for configurable amounts of CPU load",
|
---|
| 48 | 'abstract' => "no",
|
---|
| 49 | 'inherits' => "yes",
|
---|
| 50 | 'srcreplaceable' => "yes", # Source docs in regular txt format can be replaced with GS-generated html
|
---|
| 51 | 'args' => $arguments };
|
---|
| 52 |
|
---|
| 53 |
|
---|
| 54 | sub get_default_process_exp
|
---|
| 55 | {
|
---|
| 56 | my $self = shift (@_);
|
---|
| 57 | return q^(?i)\.te?xt$^;
|
---|
| 58 | }
|
---|
| 59 |
|
---|
| 60 | sub new
|
---|
| 61 | {
|
---|
| 62 | my ($class) = shift (@_);
|
---|
| 63 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
| 64 | push(@$pluginlist, $class);
|
---|
| 65 |
|
---|
| 66 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
| 67 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
| 68 |
|
---|
| 69 | my $self = new TextPlugin($pluginlist, $inputargs, $hashArgOptLists);
|
---|
| 70 |
|
---|
| 71 | return bless $self, $class;
|
---|
| 72 | }
|
---|
| 73 |
|
---|
| 74 | # do plugin specific processing of doc_obj
|
---|
| 75 | sub process
|
---|
| 76 | {
|
---|
| 77 | my $self = shift (@_);
|
---|
| 78 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
| 79 | my $outhandle = $self->{'outhandle'};
|
---|
| 80 |
|
---|
| 81 | my $cursection = $doc_obj->get_top_section();
|
---|
| 82 |
|
---|
| 83 | # get title metadata
|
---|
| 84 | # (don't need to get title if it has been passed
|
---|
| 85 | # in from another plugin)
|
---|
| 86 | if (!defined $metadata->{'Title'})
|
---|
| 87 | {
|
---|
| 88 | my $title = $self->get_title_metadata($textref);
|
---|
| 89 | $doc_obj->add_utf8_metadata ($cursection, "Title", $title);
|
---|
| 90 | }
|
---|
| 91 | # Add FileFormat metadata
|
---|
| 92 | $doc_obj->add_metadata($cursection, "FileFormat", "Text");
|
---|
| 93 |
|
---|
| 94 | if ($self->{'cpu_load'} =~ /^(medium|high)$/ )
|
---|
| 95 | {
|
---|
| 96 | $self->generateLexicon($doc_obj, $$textref);
|
---|
| 97 | $self->generateWordLengths($doc_obj, $$textref);
|
---|
| 98 | $self->{'first'} = '80,256,1024';
|
---|
| 99 | $self->generateSummaries($doc_obj, $textref);
|
---|
| 100 | $self->generateComplexity($doc_obj, $textref);
|
---|
| 101 | $self->generateEncryptedText($doc_obj, $textref, 'thePassword');
|
---|
| 102 | }
|
---|
| 103 | if ($self->{'cpu_load'} eq 'high')
|
---|
| 104 | {
|
---|
| 105 | $self->generateKeywords($doc_obj, $$textref);
|
---|
| 106 | }
|
---|
| 107 |
|
---|
| 108 | # insert preformat tags and add text to document object
|
---|
| 109 | $self->text_to_html($textref); # modifies the text
|
---|
| 110 | $doc_obj->add_utf8_text($cursection, $$textref);
|
---|
| 111 |
|
---|
| 112 | return 1;
|
---|
| 113 | }
|
---|
| 114 |
|
---|
| 115 |
|
---|
| 116 | ## @function
|
---|
| 117 | #
|
---|
| 118 | sub _debugPrint
|
---|
| 119 | {
|
---|
| 120 | my $self = shift(@_);
|
---|
| 121 | if ($self->{'debug'})
|
---|
| 122 | {
|
---|
| 123 | my ($msg) = @_;
|
---|
| 124 | print '[DEBUG] ' . $msg . "\n";
|
---|
| 125 | }
|
---|
| 126 | }
|
---|
| 127 | ## _debugPrint() ##
|
---|
| 128 |
|
---|
| 129 |
|
---|
| 130 | ## Functions to hopefully create some CPU load ##
|
---|
| 131 |
|
---|
| 132 |
|
---|
| 133 | ## @function
|
---|
| 134 | #
|
---|
| 135 | sub generateComplexity
|
---|
| 136 | {
|
---|
| 137 | my $self = shift(@_);
|
---|
| 138 | my ($doc_obj, $textref) = @_;
|
---|
| 139 | my $text = $$textref;
|
---|
| 140 |
|
---|
| 141 | # No of words (we start with 0.1 to prevent things being divided by zero)
|
---|
| 142 | my @words = split(/[^a-zA-Z0-9]+/, $text);
|
---|
| 143 | my $number_of_words = scalar(@words);
|
---|
| 144 |
|
---|
| 145 | # No of long words, where long is 6 or more characters
|
---|
| 146 | my $number_of_long_words = 0;
|
---|
| 147 | while ($text =~ /\w{6,}/g)
|
---|
| 148 | {
|
---|
| 149 | $number_of_long_words++;
|
---|
| 150 | }
|
---|
| 151 |
|
---|
| 152 | # No of syllables
|
---|
| 153 | my $number_of_syllables = 0;
|
---|
| 154 | foreach my $the_word (@words)
|
---|
| 155 | {
|
---|
| 156 | $number_of_syllables += syllable($the_word);
|
---|
| 157 | }
|
---|
| 158 |
|
---|
| 159 | # no of sentences (looking for full stops...)
|
---|
| 160 | my $number_of_sentences = ($text =~ tr/\.//);
|
---|
| 161 |
|
---|
| 162 | $self->_debugPrint('Number of words: ' . $number_of_words);
|
---|
| 163 | $self->_debugPrint('Number of sentences: ' . $number_of_sentences);
|
---|
| 164 | $self->_debugPrint('Number of syllables: ' . $number_of_syllables);
|
---|
| 165 |
|
---|
| 166 | # Commetrics Approach
|
---|
| 167 | # A. Big Word Ratio
|
---|
| 168 | # = Total # of words / Total # of words with > 6 characters
|
---|
| 169 | # B. Word Count Score
|
---|
| 170 | # = Total # of words / Total # of sentences
|
---|
| 171 | # Score = A / B
|
---|
| 172 | my $commetrics_complexity_score = ($number_of_words / $number_of_long_words) / ($number_of_words / $number_of_sentences);
|
---|
| 173 | $self->_debugPrint('ComMetrics Complexity Score: ' . $commetrics_complexity_score);
|
---|
| 174 |
|
---|
| 175 | $doc_obj->add_metadata($doc_obj->get_top_section(), 'CommetricsScore', $commetrics_complexity_score);
|
---|
| 176 |
|
---|
| 177 | # Flesch Reading Ease:
|
---|
| 178 | # Calculate the average number of words you use per sentence.
|
---|
| 179 | # Calculate the average number of syllables per word.
|
---|
| 180 | # Multiply the average number of syllables per word multiplied by 84.6 and subtract it from the average number of words multiplied by 1.015.
|
---|
| 181 | # Subtract the result from 206.835.
|
---|
| 182 | # Algorithm: 206.835 - (1.015 * average_words_sentence) - (84.6 * average_syllables_word)
|
---|
| 183 | my $words_per_sentence = $number_of_words / $number_of_sentences;
|
---|
| 184 | my $syllables_per_word = $number_of_syllables / $number_of_words;
|
---|
| 185 | my $flesch_complexity_score = 206.835 - ($words_per_sentence * 1.015) - ($syllables_per_word * 84.6);
|
---|
| 186 | $self->_debugPrint('Flesch-Kincaid Complexity Score: ' . $flesch_complexity_score);
|
---|
| 187 | $doc_obj->add_metadata($doc_obj->get_top_section(), 'FleschKincaidScore', $flesch_complexity_score);
|
---|
| 188 | my $flesch_grade = 0.38 * $words_per_sentence + 11.8 * $syllables_per_word - 15.59;
|
---|
| 189 | $self->_debugPrint('Flesch-Kincaid Grade: ' . $flesch_grade);
|
---|
| 190 | $doc_obj->add_metadata($doc_obj->get_top_section(), 'FleschKincaidGrade', $flesch_grade);
|
---|
| 191 | }
|
---|
| 192 | ## generateComplexity() ##
|
---|
| 193 |
|
---|
| 194 |
|
---|
| 195 | ## @function
|
---|
| 196 | #
|
---|
| 197 | sub generateEncryptedText
|
---|
| 198 | {
|
---|
| 199 | my $self = shift (@_);
|
---|
| 200 | my ($doc_obj, $textref) = @_;
|
---|
| 201 |
|
---|
| 202 | my $key_length = 25;
|
---|
| 203 | my $multiplier = 40;
|
---|
| 204 |
|
---|
| 205 | # Split the string into chunks
|
---|
| 206 | my $text = $$textref;
|
---|
| 207 | # - ensure the length of the text is some multiple of chunk length
|
---|
| 208 | while ((length($text) % ($multiplier * $key_length)) > 0)
|
---|
| 209 | {
|
---|
| 210 | $text .= '#';
|
---|
| 211 | }
|
---|
| 212 | # - now split the text into $multiplier x $key_length bytes chunks
|
---|
| 213 | # the first key_length bytes is used as the key to encrypt the whole chunk
|
---|
| 214 | my $counter = 0;
|
---|
| 215 | while (length($text) > 0)
|
---|
| 216 | {
|
---|
| 217 | my $key = substr($text, 0, $key_length);
|
---|
| 218 | my $value = $key . substr($text, $key_length, $key_length * ($multiplier - 1));
|
---|
| 219 | my $blowfish = new Crypt::Blowfish_PP($key);
|
---|
| 220 | my $encrypted_text = $blowfish->encrypt($value);
|
---|
| 221 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), sprintf("Encrypted%05d", $counter), $encrypted_text);
|
---|
| 222 | # - shorten text by chunk length bytes, and repeat until text is exhausted
|
---|
| 223 | $text = substr($text, $multiplier * $key_length);
|
---|
| 224 | $counter++;
|
---|
| 225 | }
|
---|
| 226 | $self->_debugPrint('Encrypted ' . $counter . ' x ' . ($multiplier * $key_length) . ' byte chunks');
|
---|
| 227 | }
|
---|
| 228 | ## generateEncryptedText() ##
|
---|
| 229 |
|
---|
| 230 |
|
---|
| 231 | ## @function
|
---|
| 232 | #
|
---|
| 233 | sub generateKeywords
|
---|
| 234 | {
|
---|
| 235 | my $self = shift (@_);
|
---|
| 236 | my ($doc_obj, $text) = @_;
|
---|
| 237 | my $key_phrases = Kea::extract_KeyPhrases('3.0', $text, 'n10');
|
---|
| 238 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), 'Keywords', $key_phrases);
|
---|
| 239 | $self->_debugPrint('Generated keywords: "' . $key_phrases . '"');
|
---|
| 240 | }
|
---|
| 241 | ## generateKeywords() ##
|
---|
| 242 |
|
---|
| 243 |
|
---|
| 244 | ## @function
|
---|
| 245 | #
|
---|
| 246 | sub generateLexicon
|
---|
| 247 | {
|
---|
| 248 | my $self = shift (@_);
|
---|
| 249 | my ($doc_obj, $text) = @_;
|
---|
| 250 | my $raw_lexicon = {};
|
---|
| 251 | my @words = split('/[\,\.\s]+/', $text);
|
---|
| 252 | foreach my $word (@words)
|
---|
| 253 | {
|
---|
| 254 | $word = lc($word);
|
---|
| 255 | if (defined $raw_lexicon->{$word})
|
---|
| 256 | {
|
---|
| 257 | $raw_lexicon->{$word}++;
|
---|
| 258 | }
|
---|
| 259 | else
|
---|
| 260 | {
|
---|
| 261 | $raw_lexicon->{$word} = 1;
|
---|
| 262 | }
|
---|
| 263 | }
|
---|
| 264 | my @lexicon;
|
---|
| 265 | foreach my $word (sort keys %{$raw_lexicon})
|
---|
| 266 | {
|
---|
| 267 | push(@lexicon, $word . ':' . $raw_lexicon->{$word});
|
---|
| 268 | }
|
---|
| 269 | $doc_obj->add_metadata($doc_obj->get_top_section(), "Lexicon", join(', ', @lexicon));
|
---|
| 270 | $self->_debugPrint('Generated lexicon');
|
---|
| 271 | }
|
---|
| 272 | ## generateLexicon() ##
|
---|
| 273 |
|
---|
| 274 |
|
---|
| 275 | ## @function
|
---|
| 276 | # extract the first NNN characters as metadata
|
---|
| 277 | sub generateSummaries
|
---|
| 278 | {
|
---|
| 279 | my $self = shift (@_);
|
---|
| 280 | my ($doc_obj, $textref) = @_;
|
---|
| 281 |
|
---|
| 282 | foreach my $size (split /,/, $self->{'first'})
|
---|
| 283 | {
|
---|
| 284 | my $tmptext = $$textref;
|
---|
| 285 | $tmptext =~ s/^\s+//;
|
---|
| 286 | $tmptext =~ s/\s+$//;
|
---|
| 287 | $tmptext =~ s/\s+/ /gs;
|
---|
| 288 | $tmptext = substr ($tmptext, 0, $size);
|
---|
| 289 | $tmptext =~ s/\s\S*$/…/;
|
---|
| 290 | $doc_obj->add_utf8_metadata ($doc_obj->get_top_section(), 'First' . $size, $tmptext);
|
---|
| 291 | $self->_debugPrint('Generated summary of ' . $size . ' characters');
|
---|
| 292 | }
|
---|
| 293 | }
|
---|
| 294 |
|
---|
| 295 |
|
---|
| 296 | ## @function
|
---|
| 297 | #
|
---|
| 298 | sub generateWordLengths
|
---|
| 299 | {
|
---|
| 300 | my $self = shift (@_);
|
---|
| 301 | my ($doc_obj, $text) = @_;
|
---|
| 302 | my $raw_word_lengths = {};
|
---|
| 303 | my @words = split('/[\,\.\s]+/', $text);
|
---|
| 304 | foreach my $word (@words)
|
---|
| 305 | {
|
---|
| 306 | $word = lc($word);
|
---|
| 307 | my $length = length($word);
|
---|
| 308 | if (defined $raw_word_lengths->{$length})
|
---|
| 309 | {
|
---|
| 310 | $raw_word_lengths->{$length} = 1;
|
---|
| 311 | }
|
---|
| 312 | else
|
---|
| 313 | {
|
---|
| 314 | $raw_word_lengths->{$length}++;
|
---|
| 315 | }
|
---|
| 316 | }
|
---|
| 317 | my @word_lengths;
|
---|
| 318 | foreach my $word_length (sort keys %{$raw_word_lengths})
|
---|
| 319 | {
|
---|
| 320 | push(@word_lengths, $word_length . ':' . $raw_word_lengths->{$word_length});
|
---|
| 321 | }
|
---|
| 322 | $doc_obj->add_metadata($doc_obj->get_top_section(), "WordLengths", join(', ', @word_lengths));
|
---|
| 323 | $self->_debugPrint('Generated word length information');
|
---|
| 324 | }
|
---|
| 325 | ## generateWordLengths() ##
|
---|
| 326 |
|
---|
| 327 |
|
---|
| 328 | 1;
|
---|