| 1 | #!/usr/local/bin/perl | ||
| 2 | |||
| 3 | =head1 NAME | ||
| 4 | |||
| 5 | gram-ga.pl - Command-line interface to An GramadE<oacute>ir | ||
| 6 | |||
| 7 | =head1 SYNOPSIS | ||
| 8 | |||
| 9 | B<gram-ga.pl> [I<options>] [I<FILE>...] | ||
| 10 | |||
| 11 | =head1 DESCRIPTION | ||
| 12 | |||
| 13 | This script checks the grammar of Irish language input | ||
| 14 | I<FILE>s (or standard input if no files are named). The default behavior | ||
| 15 | is to write a summary of possible errors to standard output. | ||
| 16 | |||
| 17 | =head1 OPTIONS | ||
| 18 | |||
| 19 | =over 4 | ||
| 20 | |||
| 21 | =item B<-a>, B<--all>, | ||
| 22 | B<--iomlan> | ||
| 23 | |||
| 24 | report all errors (i.e. do not use ~/.neamhshuim) | ||
| 25 | |||
| 26 | =item B<-f>, B<--incode>, | ||
| 27 | B<--ionchod>=I<ENC> | ||
| 28 | |||
| 29 | specify the character encoding of the text to be checked | ||
| 30 | |||
| 31 | =item B<-t>, B<--outcode>, | ||
| 32 | B<--aschod>=I<ENC> | ||
| 33 | |||
| 34 | specify the character encoding for output | ||
| 35 | |||
| 36 | =item B<--interface>, | ||
| 37 | B<--comheadan>=xx | ||
| 38 | |||
| 39 | choose the language for error messages | ||
| 40 | |||
| 41 | =item B<--color>, B<--colour>, | ||
| 42 | B<--dath>=I<COLOR> | ||
| 43 | |||
| 44 | specify the color to use for highlighting errors | ||
| 45 | |||
| 46 | =item B<-l>, B<--list>, | ||
| 47 | B<--litriu> | ||
| 48 | |||
| 49 | write misspelled words to standard output | ||
| 50 | |||
| 51 | =item B<--moltai>, | ||
| 52 | B<--aspell> | ||
| 53 | |||
| 54 | suggest corrections for misspellings | ||
| 55 | |||
| 56 | =item B<-o>, B<--output>, | ||
| 57 | B<--aschur>=I<FILE> | ||
| 58 | |||
| 59 | write output to I<FILE> | ||
| 60 | |||
| 61 | =item B<-h>, B<--cabhair>, | ||
| 62 | B<--help> | ||
| 63 | |||
| 64 | display this help and exit | ||
| 65 | |||
| 66 | =item B<-v>, B<--leagan>, | ||
| 67 | B<--version> | ||
| 68 | |||
| 69 | output version information and exit | ||
| 70 | |||
| 71 | |||
| 72 | =item B<--api> | ||
| 73 | |||
| 74 | output a simple XML format for use with other applications | ||
| 75 | |||
| 76 | =item B<--html> | ||
| 77 | |||
| 78 | produce HTML output for viewing in a web browser | ||
| 79 | |||
| 80 | =item B<--no-unigram> | ||
| 81 | |||
| 82 | do not resolve ambiguous parts of speech by frequency | ||
| 83 | |||
| 84 | =item B<--xml> | ||
| 85 | |||
| 86 | write tagged XML stream to standard output, for debugging | ||
| 87 | |||
| 88 | |||
| 89 | =back | ||
| 90 | |||
| 91 | =head1 FILES | ||
| 92 | |||
| 93 | If there are words you wish to be ignored by the grammar checker | ||
| 94 | (proper names, etc.) you can place them in a file called I<.neamhshuim> | ||
| 95 | in your home directory, one word per line. | ||
| 96 | |||
| 97 | =head1 REQUIRES | ||
| 98 | |||
| 99 | Perl 5.8, Lingua::GA::Gramadoir | ||
| 100 | |||
| 101 | =head1 SEE ALSO | ||
| 102 | |||
| 103 | =over 4 | ||
| 104 | |||
| 105 | =item * | ||
| 106 | L<http://borel.slu.edu/gramadoir/> | ||
| 107 | |||
| 108 | =item * | ||
| 109 | L<Lingua::GA::Gramadoir> | ||
| 110 | |||
| 111 | =item * | ||
| 112 | L<perl(1)> | ||
| 113 | |||
| 114 | =back | ||
| 115 | |||
| 116 | =head1 AUTHOR | ||
| 117 | |||
| 118 | Kevin P. Scannell, E<lt>kscanne@gmail.comE<gt>. | ||
| 119 | |||
| 120 | =head1 COPYRIGHT AND LICENSE | ||
| 121 | |||
| 122 | Copyright (C) 2004, 2005, 2007 Kevin P. Scannell | ||
| 123 | |||
| 124 | This is free software; you can redistribute it and/or modify | ||
| 125 | it under the same terms as Perl itself, either Perl version 5.8.2 or, | ||
| 126 | at your option, any later version of Perl 5 you may have available. | ||
| 127 | |||
| 128 | =cut | ||
| 129 | |||
| 130 | use strict; | ||
| 131 | use warnings; | ||
| 132 | |||
| 133 | use Getopt::Long qw(:config gnu_getopt); | ||
| 134 | use Term::ANSIColor; | ||
| 135 | use Lingua::GA::Gramadoir; | ||
| 136 | use Lingua::GA::Gramadoir::Languages; | ||
| 137 | # use open OUT => ':utf8'; # bad bad bad | ||
| 138 | # gram-ga.pl -v | ||
| 139 | # != gram-ga.pl -v > file; cat file | ||
| 140 | |||
| 141 | my $lh; | ||
| 142 | my $clar; | ||
| 143 | my $VERSION = '0.70'; | ||
| 144 | |||
| 145 | sub gettext | ||
| 146 | { | ||
| 147 | my ( $string, @rest ) = @_; | ||
| 148 | |||
| 149 | $string =~ s/\\n/\n/g; | ||
| 150 | $string =~ s/\[/~[/g; | ||
| 151 | $string =~ s/\]/~]/g; | ||
| 152 | $string =~ s/\%s/[_1]/; | ||
| 153 | $string =~ s/\%s/[_2]/; | ||
| 154 | return $lh->maketext($string, @rest); | ||
| 155 | } | ||
| 156 | |||
| 157 | sub localized_die | ||
| 158 | { | ||
| 159 | my ( $signal ) = @_; | ||
| 160 | my $msg; | ||
| 161 | |||
| 162 | if ( $signal =~ m/^Unknown option: (.*)/ ) { | ||
| 163 | my $arg = $1; | ||
| 164 | chomp $arg; | ||
| 165 | $msg = gettext('unrecognized option %s', '`'.$arg.'\''); | ||
| 166 | } | ||
| 167 | elsif ( $signal =~ m/^Option (.*) requires an argument/ ) { | ||
| 168 | $msg = gettext('option %s requires an argument', '`'.$1.'\''); | ||
| 169 | } | ||
| 170 | elsif ( $signal =~ m/^Option (.*) does not take/ ) { | ||
| 171 | $msg = gettext('option %s does not allow an argument', '`'.$1.'\''); | ||
| 172 | } | ||
| 173 | elsif ( $signal =~ m/^getopt/ ) { | ||
| 174 | $msg = gettext('error parsing command-line options'); | ||
| 175 | } | ||
| 176 | elsif ( $signal =~ m/^Invalid attribute name ([^ ]*)/ ) { | ||
| 177 | $msg = gettext('Unable to set output color to %s', $1); | ||
| 178 | } | ||
| 179 | elsif ( $signal =~ m/^gram: maketext (.*)/ ) { | ||
| 180 | $msg = gettext('Language %s is not supported.', $1); | ||
| 181 | } | ||
| 182 | else { | ||
| 183 | die $signal; | ||
| 184 | } | ||
| 185 | my $dieclar = gettext('An Gramadoir'); # can't use global $clar yet | ||
| 186 | my $tryhelp = gettext('Try %s for more information.', '`gram-ga.pl --help\''); | ||
| 187 | print STDERR "$dieclar: $msg\n$tryhelp\n"; | ||
| 188 | exit 1; | ||
| 189 | } | ||
| 190 | |||
| 191 | $lh = Lingua::GA::Gramadoir::Languages->get_handle(); | ||
| 192 | $SIG{__DIE__} = 'localized_die'; | ||
| 193 | # scalars for global options | ||
| 194 | use vars qw($api $aschur $aspell $comheadan $dath $help $html $iomlan $ionchod $aschod $litriu $version $xml $nounigram); | ||
| 195 | $dath = 'bold red'; | ||
| 196 | $ionchod = 'ISO-8859-1'; | ||
| 197 | $aschod = 'utf8'; | ||
| 198 | eval { | ||
| 199 | local $SIG{__WARN__} = 'localized_die'; | ||
| 200 | GetOptions ( | ||
| 201 | 'all|iomlan|a' => \$iomlan, | ||
| 202 | 'api' => \$api, | ||
| 203 | 'aspell|moltai' => \$aspell, | ||
| 204 | 'list|litriu|l' => \$litriu, | ||
| 205 | 'color|colour|dath=s' => \$dath, | ||
| 206 | 'incode|ionchod|f=s' => \$ionchod, | ||
| 207 | 'outcode|aschod|t=s' => \$aschod, | ||
| 208 | 'no-unigram' => \$nounigram, | ||
| 209 | 'help|cabhair|h' => \$help, | ||
| 210 | 'html' => \$html, | ||
| 211 | 'interface|comheadan=s' => \$comheadan, | ||
| 212 | 'output|aschur|o=s' => \$aschur, | ||
| 213 | 'version|leagan|v' => \$version, | ||
| 214 | 'xml' => \$xml, | ||
| 215 | ) | ||
| 216 | }; | ||
| 217 | die "getopt error" if $@; | ||
| 218 | |||
| 219 | $xml = 1 if $nounigram; | ||
| 220 | unless ($dath eq "none") { | ||
| 221 | eval { color $dath }; | ||
| 222 | } | ||
| 223 | |||
| 224 | if ($aschur) { | ||
| 225 | unless ($^O eq 'MSWin32') { | ||
| 226 | $aschur =~ s#^~(\w*)#$1 ? (getpwnam($1))[7] : ( $ENV{HOME} || $ENV{LOGDIR} || (getpwuid($>))[7] )#e; | ||
| 227 | } | ||
| 228 | open(OUTSTREAM, ">:encoding($aschod)", $aschur) or | ||
| 229 | warn "Can't open $aschur: $!\n"; | ||
| 230 | } | ||
| 231 | else { | ||
| 232 | open(OUTSTREAM, ">&=STDOUT") or | ||
| 233 | warn "Couldn't alias STDOUT: $!\n"; | ||
| 234 | binmode OUTSTREAM, ":encoding($aschod)"; # must be after alias | ||
| 235 | } | ||
| 236 | |||
| 237 | binmode STDERR, ":encoding($aschod)"; | ||
| 238 | |||
| 239 | if ($comheadan) { | ||
| 240 | $lh = Lingua::GA::Gramadoir::Languages->get_handle($comheadan); | ||
| 241 | die "gram: maketext $comheadan" unless $lh; | ||
| 242 | } | ||
| 243 | # $lh->fail_with('failure_handler_auto'); | ||
| 244 | |||
| 245 | # TRANSLATORS: Please leave untranslated, but add an acute accent to the | ||
| 246 | # "o" in "Gramadoir" if available in your character set and encoding. | ||
| 247 | # If you must translate, this is the Irish for (literally) "The Grammarian" | ||
| 248 | $clar = gettext('An Gramadoir'); | ||
| 249 | |||
| 250 | if ($version) { | ||
| 251 | my $vstring = gettext('version %s', $VERSION); | ||
| 252 | my $copyright = 'Copyright (C) 2003-2007 Kevin P. Scannell'; | ||
| 253 | my $gpl = gettext('This is free software; see the source for copying conditions. There is NO\\nwarranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE,\\nto the extent permitted by law.'); | ||
| 254 | if ($html) { | ||
| 255 | $copyright =~ s#(Kevin P. Scannell)#<a href="http://borel.slu.edu/">$1</a><br><br>#; | ||
| 256 | print OUTSTREAM "<p>\n<a href=\"http://borel.slu.edu/gramadoir/\">$clar</a>, $vstring<br>\n$copyright\n<i>$gpl</i></p><hr>\n"; | ||
| 257 | } | ||
| 258 | else { | ||
| 259 | print OUTSTREAM "$clar, $vstring\n$copyright\n$gpl\n"; | ||
| 260 | } | ||
| 261 | close OUTSTREAM; | ||
| 262 | exit 0; | ||
| 263 | } | ||
| 264 | if ($help) { | ||
| 265 | my @helpmessages = ( | ||
| 266 | gettext( | ||
| 267 | 'Usage: %s [OPTIONS] [FILES]', "gram-ga.pl"), | ||
| 268 | "", | ||
| 269 | gettext( | ||
| 270 | "Options for end-users:"), | ||
| 271 | # TRANSLATORS: ~/.neamhshuim is an "ignore file" like those with spellcheckers | ||
| 272 | " -a, --all,", | ||
| 273 | gettext( | ||
| 274 | " --iomlan report all errors (i.e. do not use ~/.neamhshuim)"), | ||
| 275 | " -f, --incode,", | ||
| 276 | gettext( | ||
| 277 | " --ionchod=ENC specify the character encoding of the text to be checked"), | ||
| 278 | " -t, --outcode,", | ||
| 279 | gettext( | ||
| 280 | " --aschod=ENC specify the character encoding for output"), | ||
| 281 | " --interface,", | ||
| 282 | gettext( | ||
| 283 | " --comheadan=xx choose the language for error messages"), | ||
| 284 | " --color, --colour,", | ||
| 285 | gettext( | ||
| 286 | " --dath=COLOR specify the color to use for highlighting errors"), | ||
| 287 | " -l, --list,", | ||
| 288 | gettext( | ||
| 289 | " --litriu write misspelled words to standard output"), | ||
| 290 | " --moltai,", | ||
| 291 | gettext( | ||
| 292 | " --aspell suggest corrections for misspellings"), | ||
| 293 | " -o, --output,", | ||
| 294 | gettext( | ||
| 295 | " --aschur=FILE write output to FILE"), | ||
| 296 | " -h, --cabhair,", | ||
| 297 | gettext( | ||
| 298 | " --help display this help and exit"), | ||
| 299 | " -v, --leagan,", | ||
| 300 | gettext( | ||
| 301 | " --version output version information and exit"), | ||
| 302 | "", | ||
| 303 | gettext( | ||
| 304 | "Options for developers:"), | ||
| 305 | gettext( | ||
| 306 | " --api output a simple XML format for use with other applications"), | ||
| 307 | gettext( | ||
| 308 | " --html produce HTML output for viewing in a web browser"), | ||
| 309 | # TRANSLATORS: By default, if there is no rule in the disambiguation module | ||
| 310 | # for selecting the correct part of speech of an ambiguous word, the program | ||
| 311 | # chooses the part of speech with the highest frequency. This is sometimes | ||
| 312 | # called "unigram" tagging. The --no-unigram option turns this behavior off. | ||
| 313 | gettext( | ||
| 314 | " --no-unigram do not resolve ambiguous parts of speech by frequency"), | ||
| 315 | # TRANSLATORS: The grammar checker works by passing the input text | ||
| 316 | # through a sequence of filters which add XML markup indicating | ||
| 317 | # important grammatical information. The --xml option displays the | ||
| 318 | # marked up file as an aid in debugging. | ||
| 319 | gettext( | ||
| 320 | " --xml write tagged XML stream to standard output, for debugging"), | ||
| 321 | "", | ||
| 322 | gettext( | ||
| 323 | "If no file is given, read from standard input."), | ||
| 324 | "", | ||
| 325 | gettext( | ||
| 326 | 'Send bug reports to <%s>.','kscanne@gmail.com'), | ||
| 327 | "" | ||
| 328 | ); | ||
| 329 | |||
| 330 | print OUTSTREAM join("\n", @helpmessages); | ||
| 331 | close OUTSTREAM; | ||
| 332 | exit 0; | ||
| 333 | } | ||
| 334 | |||
| 335 | my $gr = new Lingua::GA::Gramadoir( | ||
| 336 | fix_spelling => $aspell, | ||
| 337 | use_ignore_file => ! $iomlan, | ||
| 338 | unigram_tagging => ! $nounigram, | ||
| 339 | interface_language => $comheadan, | ||
| 340 | input_encoding => $ionchod, | ||
| 341 | ); | ||
| 342 | |||
| 343 | binmode STDIN, ":bytes"; | ||
| 344 | unshift(@ARGV, '-') unless @ARGV; | ||
| 345 | while ($ARGV = shift @ARGV) { | ||
| 346 | unless ($ARGV eq "-") { | ||
| 347 | unless ( -e $ARGV ) { | ||
| 348 | warn "$clar: $ARGV: ".gettext('There is no such file.')."\n"; | ||
| 349 | next; | ||
| 350 | } | ||
| 351 | if ( -d $ARGV ) { | ||
| 352 | warn "$clar: $ARGV: ".gettext('Is a directory')."\n"; | ||
| 353 | next; | ||
| 354 | } | ||
| 355 | unless ( -f $ARGV ) { | ||
| 356 | warn "$ARGV is Not a plain file\n"; | ||
| 357 | next; | ||
| 358 | } | ||
| 359 | unless ( -r $ARGV ) { | ||
| 360 | warn "$clar: $ARGV: ".gettext('Permission denied')."\n"; | ||
| 361 | next; | ||
| 362 | } | ||
| 363 | unless (open(ARGV, "<:bytes", $ARGV)) { | ||
| 364 | warn "Can't open $ARGV: $!\n"; | ||
| 365 | next; | ||
| 366 | } | ||
| 367 | } | ||
| 368 | local $/; # slurp a file at a time | ||
| 369 | $_ = <ARGV>; | ||
| 370 | warn gettext('%s: warning: problem closing %s\n', $clar, $ARGV) unless close ARGV; | ||
| 371 | if ($litriu) { | ||
| 372 | my $missp = $gr->spell_check($_); | ||
| 373 | foreach (@$missp) { | ||
| 374 | print OUTSTREAM "$_\n"; | ||
| 375 | } | ||
| 376 | } | ||
| 377 | elsif ($xml) { | ||
| 378 | print OUTSTREAM $gr->xml_stream($_); | ||
| 379 | } | ||
| 380 | elsif ($api) { | ||
| 381 | print OUTSTREAM '<?xml version="1.0" encoding="utf-8" standalone="no"?>'."\n".'<!DOCTYPE matches SYSTEM "http://borel.slu.edu/dtds/api-output.dtd">'."\n<matches>\n"; | ||
| 382 | my $apierr = $gr->grammatical_errors($_); | ||
| 383 | foreach (@$apierr) { | ||
| 384 | print OUTSTREAM "$_\n"; | ||
| 385 | } | ||
| 386 | print OUTSTREAM "</matches>\n"; | ||
| 387 | } | ||
| 388 | else { # vanilla or html | ||
| 389 | my $status = gettext('Currently checking %s', $ARGV); | ||
| 390 | print OUTSTREAM "$status\n" unless ($ARGV eq "-"); | ||
| 391 | my $errs = $gr->grammatical_errors($_); | ||
| 392 | foreach my $error (@$errs) { | ||
| 393 | (my $ln, my $msg, my $snt, my $offset, my $len) = $error =~ m/^<error fromy="([0-9]+)".* msg="([^"]+)".* context="([^"]+)" contextoffset="([0-9]+)" errorlength="([0-9]+)"\/>$/; | ||
| 394 | my $errortext = substr($snt,$offset,$len); | ||
| 395 | $ln++; # people count lines from 1 | ||
| 396 | my $s = "<br><br>$ln: ".substr($snt,0,$offset); | ||
| 397 | if ($html) { | ||
| 398 | $s .= "<b class=gramadoir>$errortext</b>"; | ||
| 399 | } | ||
| 400 | else { | ||
| 401 | if ($dath eq "none") { | ||
| 402 | $s .= $errortext; | ||
| 403 | } | ||
| 404 | else { | ||
| 405 | $s .= colored($errortext,$dath); | ||
| 406 | } | ||
| 407 | } | ||
| 408 | $s .= substr($snt,$offset+$len)."<br>\n$msg\n\n"; # don't add punctuation | ||
| 409 | if (!$html) { | ||
| 410 | $s =~ s/<br>//g; | ||
| 411 | $s =~ s/"/"/g; | ||
| 412 | $s =~ s/</</g; | ||
| 413 | $s =~ s/>/>/g; | ||
| 414 | $s =~ s/&/&/g; | ||
| 415 | } | ||
| 416 | print OUTSTREAM $s; | ||
| 417 | } # loop over errors | ||
| 418 | } # vanilla/html block | ||
| 419 | } # loop over files | ||
| 420 | |||
| 421 | close OUTSTREAM; | ||
| 422 | exit 0; |