File: multispell

package info (click to toggle)
hspell 1.4-3
  • links: PTS, VCS
  • area: main
  • in suites: bullseye, buster, sid
  • size: 1,556 kB
  • sloc: ansic: 2,808; perl: 1,989; makefile: 209; sh: 109; awk: 15
file content (357 lines) | stat: -rwxr-xr-x 9,577 bytes parent folder | download | duplicates (8)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
#!/usr/bin/perl -w

# Multispell is a program that lets your editor talk to two spellers
# simultaneously. It launches two speller programs, one Hebrew (hspell),
# one English (ispell), and merges their results.
#
# Public Domain
#
# Please report bugs to: Mooffie <mooffie@typo.co.il>
#
# TODO
# - UTF-8 support (+cmdline opt to select offsets type: byte-based
#   or char-based)
# - intl support. e.g.:
#	multispell -T utf8 --ispell-encoding=koi8-r
#	or:
#	multispell -T utf8 --ispell-cmd="multispell --ispell-cmd=..." ...
# - handle aspell's '$$' ?

use strict;
use Getopt::Long;
use IPC::Open2;
use IO::Handle;

# options to pass to both hspell and ispell:
my @common_opts	    = qw/ --dummy /;
# options to pass to hspell only:
my @hspell_opts	    = qw/ -n --notes --verbose /;
my $hspell_cmd	    = "hspell";
my $ispell_cmd	    = "ispell";
my $remove_niqqud   = 0;

my $debug	    = 0;
my $debug_file	    = "$ENV{HOME}/multispell-dbg.txt";
my $rc_file	    = "$ENV{HOME}/.multispellrc";
my $rc_file_global  = "/etc/multispellrc";

my ($hsplrd, $hsplwr); # pipes to hspell
my ($isplrd, $isplwr); # pipes to ispell

my $VERSION = "0.4a";

sub DBG {
    if ($debug) {
	my $msg = shift;
	print DBGFL $msg;
    }
}

sub help {
    print <<EOM;
SYNOPSIS:
    multispell [multispell-options] [hspell-options] [ispell-options]

DESCRIPTION:
    Multispell is a program that lets your editor talk to two spellers
    simultaneously. It launches two speller programs, one Hebrew (hspell),
    one English (ispell), and merges their results.

    Currently the editor can talk with multispell only in ISO-8859-8
    or CP1255.
    
VALID MULTISPELL-OPTIONS:
    --debug
	    Write debug info to ~/multispell-dbg.txt
    
    --hspell-cmd PROG
	    Use PROG instead of "$hspell_cmd"
    
    --ispell-cmd PROG
	    Use PROG instead of "$ispell_cmd"

    --remove-niqqud
	    Erase words containing CP1255's niqqud (Hebrew points),
	    and convert maqaf (0xCE) to ASCII dash.

    --drop OPTNAME
	    Don't pass option OPTNAME to ispell or to hspell. It's
	    useful in ~/.multispellrc when you want to drop an option
	    the editor is determined to pass. You can specify multiple
	    options, separated by commas. If the option requires an
	    argument, append ':' to its name.

    --help -h
	    Print this message

FILES:
    /etc/multispellrc
	    Additional command-line options to use. These _override_
	    the ones specified on the command-line.

    ~/.multispellrc
	    If found, used _instead_ of /etc/multispellrc

EOM
    exit;
}

sub load_spellers
{
    sub load_speller {
	my ($cmd) = @_;
	local (*RD, *WR);
	DBG("Loading: $cmd\n");
	open2(*RD, *WR, $cmd);
	my $signature = <RD>;
	if (!$signature || $signature !~ /^@\(#\)/) {
	    DBG("Loading failed (please check stderr)\n");
	    die "Can't load $cmd";
	} else {
	    DBG("Loaded OK: $signature");
	}
	autoflush WR 1;
	return (*RD, *WR);
    }

    ($hsplrd, $hsplwr) = load_speller($hspell_cmd);
    ($isplrd, $isplwr) = load_speller($ispell_cmd);
}

sub print_signature {
    print "@(#) International Ispell Version 3.1.20 ".
	    "(but really multispell $VERSION)\n";
}

# do_loop() is the crux of the program. it reads ispell-a requests
# from stdin, sends them to ispell and/or hspell, and writes the merged
# replies to stdout.

sub do_loop {
    my $heb_chars	 = '\xE0-\xFA';
    my $niqqud_chars	 = '\xC0-\xCD\xCF';
    my $niqqud_heb_chars = $heb_chars . $niqqud_chars;

    print_signature();

    while (my $line = <STDIN>) {
	DBG(">$line");
	my @replies = ();

	# if the last line of the input does not end in newline, add it ourselves
	# so that hspell doesn't hang waiting for it.
	# e.g.: echo -n word | multispell -a
	$line .= "\n" if $line !~ /\n/;

	# a workaround for some versions of Aspell that ignore empty input lines.
	# see http://ivrix.org.il/bugzilla/show_bug.cgi?id=3
	$line = " \n" if $line eq "\n";

	if ($remove_niqqud) {
	    # erase words containing CP1255's niqqud (hebrew points)
	    $line =~ s/([$niqqud_heb_chars]+[$niqqud_chars][$niqqud_heb_chars]*)/
			    ' ' x length($1)/oeg;
	    # convert maqaf to ASCII dash
	    $line =~ tr/\xCE/-/;
	}
	
	if ($line =~ /^[#!~@%+&*-]/) {
	    if ($line =~ /^[@&*]/) {
		# "@word" - ignore word.
		# "*word" - add word to private dict.
		# "&word" - ditto.
		# We send hebrew words to hspell, otherwise to ispell
		if ($line =~ /[$heb_chars]/o
			|| $line eq "\@ActivateExtendedProtocol\n") {
		    DBG("HSPL>$line");
		    print $hsplwr $line;
		} else {
		    DBG("ISPL>$line");
		    print $isplwr $line;
		}
	    } else {
		DBG("IHSPL>$line");
		print $hsplwr $line;
		print $isplwr $line;
	    }
	    # this is a command, so we don't read replies.
	} else {
	    # words to spell-check.
	    DBG("HSPL>$line");
	    print $hsplwr $line;
	    while (<$hsplrd>) {
		DBG("HSPL<$_");
		last if ! /\S/;
		push @replies, $_;
	    }
	   
	    # delete all heb words before sending to ispell
	    $line =~ s/[$heb_chars]/ /og;
	    DBG("ISPL>$line");
	    print $isplwr $line;
	    while (<$isplrd>) {
		DBG("ISPL<$_");
		last if ! /\S/;
		push @replies, $_;
	    }

	    # We're about to sort the replies.
	    # But first we need to take care of non-standard replies (like
	    # spelling-hints). Such replies (so we believe) add information to
	    # the last standard reply (e.g. spelling-hints explain the last
	    # reported misspelled word), so we want to concatenate them with
	    # that reply instead of sorting them as if they were independent.
	    for (my $i = 1; $i <= $#replies; $i++) {
		if ($replies[$i] =~ /^[^*+&?#-]/) { # non-standard?
		    $replies[$i-1] .= $replies[$i];
		    splice @replies, $i, 1;
		    $i--;
		}
	    }

	    # sort the replies
	    sub getidx {
		$_ = shift;
		(/^[&?] \S+ \d+ (\d+)/ || /^# \S+ (\d+)/) ? $1 : 0;
	    }
	    @replies = sort { getidx($a) <=> getidx($b) } @replies;

	    push @replies, "\n";
	    DBG("<$_") foreach (@replies);
	    print @replies;
	}
    }
}

# extract_options() takes a list of option names and extracts them and
# their values from @ARGV.
# For example, if we pass ('-n', '-d:', '-S') to this function, and @ARGV
# is ('-i', '-nx', '-denglish', '-p'), it returns "-n -d english", and
# @ARGV becomes ('-i', '-x', '-p').

sub extract_options {
    my @options = @_;
    my %getopt_hash;
    my $args = "";

    foreach my $option (@options) {
	$option =~ tr/-//d;
	my $require_arg = ($option =~ s/:$/=s/);
	$getopt_hash{$option} = sub {
	    my ($optname, $optval) = @_;
	    $optname = (length($optname) > 1 ? "--" : "-") . $optname;
	    $args .= " $optname";
	    $args .= " '$optval'" if $require_arg;
	};
    }
    GetOptions(%getopt_hash);
    return $args;
}

# tokenize_cmd_line() is used for parsing the RC files.
# it takes a command line string, e.g.:
#   -x --opt="one two" -i 'on"e" two'
# and returns an array:
#   ('-x', '--opt=one two', '-i', 'on"e" two')

sub tokenize_cmd_line {
    my $s = shift;
    return  map {
		s/['"]//;
		s/['"]$//;
		$_;
	    } grep {
		/\S/
	    } split( /(
		[^\s'"]*(?:'.*?'|".*?") |
		[^\s'"]+
	    )/x , $s);
}

main: {
    $| = 1;
    Getopt::Long::Configure("bundling_override", "pass_through");
    my $print_signature	= 0;
    my $ispella_mode	= 0;
    my $dictionary	= "";
    my @drop_opts	= ();

    # Applications use the '-d' option to select a dictionary.
    # We don't want aspell to consider the locale when '-d' is not specified,
    # because in most cases that's not what the user wants (e.g. he_IL), and,
    # anyway, we can't represent most languages other than English in
    # ISO-8859-8.
    $ENV{'LC_ALL'} = 'C';

    if (open(RC, $rc_file) || open(RC, $rc_file_global)) {
	while (<RC>) {
	    s/#.*//;
	    push @ARGV, tokenize_cmd_line($_);
	}
	close(RC);
    }

    my @saved_argv = @ARGV;
    GetOptions('debug'		=> \$debug,
	       'hspell-cmd=s'	=> \$hspell_cmd,
	       'ispell-cmd=s'	=> \$ispell_cmd,
	       'drop=s'		=> \@drop_opts,
	       'remove-niqqud'	=> \$remove_niqqud,
	       'remove-niqud'	=> \$remove_niqqud,
	       'remove-nikud'	=> \$remove_niqqud,
	       'd=s'		=> \$dictionary,
	       'v'		=> \$print_signature,
	       'a'		=> \$ispella_mode,
	       'i'		=> sub { ; }, # silently eat up hspell's '-i'
	       'help|h|?'	=> sub { help(); }
	      ) or help();

    if ($debug) {
	open(DBGFL, ">>$debug_file");
	autoflush DBGFL 1;
	DBG("\n\n" . "-" x 75 . "\n");
	DBG("I was invoked: $0");
	DBG(" '$_'") for @saved_argv;
	DBG("\n");
	DBG("On: " . localtime() . "\n");
	# According to SUSV, the following 'ps' command is portable,
	# but I don't want to take risks, so I run it only on linux.
	if ($^O =~ /linux/i) {
	    my $PPID = getppid();
	    DBG("By: " . `ps -p $PPID -o args=`);
	}
    }

    if ($print_signature) {
	print_signature();
	exit;
    }

    if (!$ispella_mode) {
	die "You can only use this speller in \"pipe-mode\" " .
	    "(using the '-a' option)\n";
    }
    
    # If $dictionary contains "hebrew", it's probably something
    # kspell or emacs passed. We ignore it since ispell knows
    # no such dictionary.
    if ($dictionary && $dictionary !~ /hebrew/i) {
	push @ARGV, '-d', $dictionary;
    }

    @drop_opts = split(/,/, join(",", @drop_opts));
    extract_options(@drop_opts) if @drop_opts;
    my $common_opts = extract_options(@common_opts);
    my $hspell_opts = extract_options(@hspell_opts) . $common_opts;
    my $ispell_opts = "@ARGV" . $common_opts;

    $hspell_cmd .= " -a $hspell_opts";
    $ispell_cmd .= " -a $ispell_opts";
    
    load_spellers();
    DBG("\n");
    do_loop();
}

# vim:ts=8:sts=4:sw=4: