lingo 1.8.0 → 1.8.1
Sign up to get free protection for your applications and to get access to all the features.
- data/ChangeLog +13 -0
- data/README +49 -29
- data/Rakefile +28 -4
- data/TODO +2 -9
- data/bin/lingo +24 -0
- data/bin/lingoctl +24 -0
- data/de/lingo-dic.txt +559 -74
- data/info/gpl-hdr.txt +21 -24
- data/lib/lingo.rb +83 -112
- data/lib/lingo/agenda_item.rb +53 -0
- data/lib/lingo/attendee.rb +261 -0
- data/lib/lingo/attendee/abbreviator.rb +95 -97
- data/lib/lingo/attendee/debugger.rb +94 -93
- data/lib/lingo/attendee/decomposer.rb +76 -83
- data/lib/lingo/attendee/dehyphenizer.rb +141 -144
- data/lib/lingo/attendee/formatter.rb +65 -0
- data/lib/lingo/attendee/multi_worder.rb +302 -0
- data/lib/lingo/attendee/noneword_filter.rb +89 -84
- data/lib/lingo/attendee/object_filter.rb +91 -0
- data/lib/lingo/attendee/sequencer.rb +159 -158
- data/lib/lingo/attendee/synonymer.rb +81 -84
- data/lib/lingo/attendee/text_reader.rb +242 -0
- data/lib/lingo/attendee/text_writer.rb +169 -0
- data/lib/lingo/attendee/tokenizer.rb +192 -191
- data/lib/lingo/attendee/variator.rb +152 -156
- data/lib/lingo/attendee/vector_filter.rb +140 -135
- data/lib/lingo/attendee/word_searcher.rb +98 -0
- data/lib/lingo/buffered_attendee.rb +69 -0
- data/lib/lingo/cachable.rb +58 -0
- data/lib/lingo/call.rb +72 -0
- data/lib/lingo/cli.rb +26 -0
- data/lib/lingo/config.rb +23 -26
- data/lib/lingo/core_ext.rb +42 -0
- data/lib/lingo/ctl.rb +239 -173
- data/lib/lingo/database.rb +148 -496
- data/lib/lingo/database/crypter.rb +85 -0
- data/lib/lingo/database/gdbm_store.rb +49 -0
- data/lib/lingo/database/hash_store.rb +67 -0
- data/lib/lingo/database/libcdb_store.rb +58 -0
- data/lib/lingo/database/sdbm_store.rb +64 -0
- data/lib/lingo/database/show_progress.rb +81 -0
- data/lib/lingo/database/source.rb +134 -0
- data/lib/lingo/database/source/key_value.rb +62 -0
- data/lib/lingo/database/source/multi_key.rb +65 -0
- data/lib/lingo/database/source/multi_value.rb +65 -0
- data/lib/lingo/database/source/single_word.rb +60 -0
- data/lib/lingo/database/source/word_class.rb +64 -0
- data/lib/lingo/error.rb +122 -0
- data/lib/lingo/language.rb +78 -518
- data/lib/lingo/language/dictionary.rb +173 -0
- data/lib/lingo/language/grammar.rb +211 -0
- data/lib/lingo/language/lexical.rb +66 -0
- data/lib/lingo/language/lexical_hash.rb +88 -0
- data/lib/lingo/language/token.rb +48 -0
- data/lib/lingo/language/word.rb +130 -0
- data/lib/lingo/language/word_form.rb +83 -0
- data/lib/lingo/reportable.rb +59 -0
- data/lib/lingo/version.rb +1 -1
- data/lingo-all.cfg +14 -10
- data/lingo-call.cfg +5 -5
- data/lingo.cfg +14 -12
- data/lingo.rb +26 -0
- data/lir.cfg +13 -9
- data/spec/spec_helper.rb +1 -0
- data/test.cfg +11 -11
- data/test/attendee/ts_abbreviator.rb +0 -6
- data/test/attendee/ts_decomposer.rb +0 -6
- data/test/attendee/{ts_multiworder.rb → ts_multi_worder.rb} +1 -7
- data/test/attendee/ts_noneword_filter.rb +1 -7
- data/test/attendee/{ts_objectfilter.rb → ts_object_filter.rb} +1 -7
- data/test/attendee/ts_sequencer.rb +0 -6
- data/test/attendee/ts_synonymer.rb +0 -6
- data/test/attendee/{ts_textreader.rb → ts_text_reader.rb} +1 -7
- data/test/attendee/{ts_textwriter.rb → ts_text_writer.rb} +1 -7
- data/test/attendee/ts_tokenizer.rb +0 -6
- data/test/attendee/ts_variator.rb +0 -6
- data/test/attendee/ts_vector_filter.rb +1 -7
- data/test/attendee/{ts_wordsearcher.rb → ts_word_searcher.rb} +1 -7
- data/test/ref/artikel.non +2 -29
- data/test/ref/artikel.seq +13 -8
- data/test/ref/artikel.vec +30 -15
- data/test/ref/artikel.ven +29 -14
- data/test/ref/artikel.ver +58 -43
- data/test/ref/lir.csv +146 -145
- data/test/ref/lir.non +186 -210
- data/test/ref/lir.seq +54 -50
- data/test/test_helper.rb +41 -36
- data/test/ts_database.rb +12 -11
- data/test/ts_language.rb +118 -68
- metadata +67 -29
- data/lib/lingo/attendee/multiworder.rb +0 -301
- data/lib/lingo/attendee/objectfilter.rb +0 -86
- data/lib/lingo/attendee/textreader.rb +0 -237
- data/lib/lingo/attendee/textwriter.rb +0 -196
- data/lib/lingo/attendee/wordsearcher.rb +0 -96
- data/lib/lingo/attendees.rb +0 -289
- data/lib/lingo/const.rb +0 -131
- data/lib/lingo/modules.rb +0 -98
- data/lib/lingo/types.rb +0 -285
- data/lib/lingo/utilities.rb +0 -40
@@ -0,0 +1,62 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
#--
|
4
|
+
###############################################################################
|
5
|
+
# #
|
6
|
+
# Lingo -- A full-featured automatic indexing system #
|
7
|
+
# #
|
8
|
+
# Copyright (C) 2005-2007 John Vorhauer #
|
9
|
+
# Copyright (C) 2007-2012 John Vorhauer, Jens Wille #
|
10
|
+
# #
|
11
|
+
# Lingo is free software; you can redistribute it and/or modify it under the #
|
12
|
+
# terms of the GNU Affero General Public License as published by the Free #
|
13
|
+
# Software Foundation; either version 3 of the License, or (at your option) #
|
14
|
+
# any later version. #
|
15
|
+
# #
|
16
|
+
# Lingo is distributed in the hope that it will be useful, but WITHOUT ANY #
|
17
|
+
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS #
|
18
|
+
# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for #
|
19
|
+
# more details. #
|
20
|
+
# #
|
21
|
+
# You should have received a copy of the GNU Affero General Public License #
|
22
|
+
# along with Lingo. If not, see <http://www.gnu.org/licenses/>. #
|
23
|
+
# #
|
24
|
+
###############################################################################
|
25
|
+
#++
|
26
|
+
|
27
|
+
class Lingo
|
28
|
+
|
29
|
+
class Database
|
30
|
+
|
31
|
+
class Source
|
32
|
+
|
33
|
+
# Abgeleitet von Source behandelt die Klasse Dateien mit dem Format <tt>KeyValue</tt>.
|
34
|
+
# Eine Zeile <tt>"Fachbegriff*Fachterminus\n"</tt> wird gewandelt in <tt>[ 'fachbegriff', ['fachterminus#s'] ]</tt>.
|
35
|
+
# Die Wortklasse kann über den Parameter <tt>def-wc</tt> beeinflusst werden.
|
36
|
+
# Der Trenner zwischen Schlüssel und Projektion kann über den Parameter <tt>separator</tt> geändert werden.
|
37
|
+
|
38
|
+
class KeyValue < self
|
39
|
+
|
40
|
+
def initialize(id, lingo)
|
41
|
+
super
|
42
|
+
|
43
|
+
@separator = @config.fetch('separator', '*')
|
44
|
+
@line_pattern = Regexp.new('^(' + @legal_word + ')' + Regexp.escape(@separator) + '(' + @legal_word + ')$')
|
45
|
+
end
|
46
|
+
|
47
|
+
private
|
48
|
+
|
49
|
+
def convert_line(line, key, val)
|
50
|
+
key, val = key.strip, val.strip
|
51
|
+
val = '' if key == val
|
52
|
+
val = [val + '#' + @wordclass]
|
53
|
+
[key, val]
|
54
|
+
end
|
55
|
+
|
56
|
+
end
|
57
|
+
|
58
|
+
end
|
59
|
+
|
60
|
+
end
|
61
|
+
|
62
|
+
end
|
@@ -0,0 +1,65 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
#--
|
4
|
+
###############################################################################
|
5
|
+
# #
|
6
|
+
# Lingo -- A full-featured automatic indexing system #
|
7
|
+
# #
|
8
|
+
# Copyright (C) 2005-2007 John Vorhauer #
|
9
|
+
# Copyright (C) 2007-2012 John Vorhauer, Jens Wille #
|
10
|
+
# #
|
11
|
+
# Lingo is free software; you can redistribute it and/or modify it under the #
|
12
|
+
# terms of the GNU Affero General Public License as published by the Free #
|
13
|
+
# Software Foundation; either version 3 of the License, or (at your option) #
|
14
|
+
# any later version. #
|
15
|
+
# #
|
16
|
+
# Lingo is distributed in the hope that it will be useful, but WITHOUT ANY #
|
17
|
+
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS #
|
18
|
+
# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for #
|
19
|
+
# more details. #
|
20
|
+
# #
|
21
|
+
# You should have received a copy of the GNU Affero General Public License #
|
22
|
+
# along with Lingo. If not, see <http://www.gnu.org/licenses/>. #
|
23
|
+
# #
|
24
|
+
###############################################################################
|
25
|
+
#++
|
26
|
+
|
27
|
+
class Lingo
|
28
|
+
|
29
|
+
class Database
|
30
|
+
|
31
|
+
class Source
|
32
|
+
|
33
|
+
# Abgeleitet von Source behandelt die Klasse Dateien mit dem Format <tt>MultiKey</tt>.
|
34
|
+
# Eine Zeile <tt>"Triumph;Sieg;Erfolg\n"</tt> wird gewandelt in <tt>[ 'triumph', ['sieg', 'erfolg'] ]</tt>.
|
35
|
+
# Die Sonderbehandlung erfolgt in der Methode Database#convert, wo daraus Schlüssel-Werte-Paare in der Form
|
36
|
+
# <tt>[ 'sieg', ['triumph'] ]</tt> und <tt>[ 'erfolg', ['triumph'] ]</tt> erzeugt werden.
|
37
|
+
# Der Trenner zwischen Schlüssel und Projektion kann über den Parameter <tt>separator</tt> geändert werden.
|
38
|
+
|
39
|
+
class MultiKey < self
|
40
|
+
|
41
|
+
def initialize(id, lingo)
|
42
|
+
super
|
43
|
+
|
44
|
+
@separator = @config.fetch('separator', ';')
|
45
|
+
@line_pattern = Regexp.new('^' + @legal_word + '(?:' + Regexp.escape(@separator) + @legal_word + ')*$')
|
46
|
+
end
|
47
|
+
|
48
|
+
def set(db, key, val)
|
49
|
+
val.each { |v| db[v] = [key] }
|
50
|
+
end
|
51
|
+
|
52
|
+
private
|
53
|
+
|
54
|
+
def convert_line(line, key, val)
|
55
|
+
values = line.split(@separator).map { |value| value.strip }
|
56
|
+
[values[0], values[1..-1]]
|
57
|
+
end
|
58
|
+
|
59
|
+
end
|
60
|
+
|
61
|
+
end
|
62
|
+
|
63
|
+
end
|
64
|
+
|
65
|
+
end
|
@@ -0,0 +1,65 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
#--
|
4
|
+
###############################################################################
|
5
|
+
# #
|
6
|
+
# Lingo -- A full-featured automatic indexing system #
|
7
|
+
# #
|
8
|
+
# Copyright (C) 2005-2007 John Vorhauer #
|
9
|
+
# Copyright (C) 2007-2012 John Vorhauer, Jens Wille #
|
10
|
+
# #
|
11
|
+
# Lingo is free software; you can redistribute it and/or modify it under the #
|
12
|
+
# terms of the GNU Affero General Public License as published by the Free #
|
13
|
+
# Software Foundation; either version 3 of the License, or (at your option) #
|
14
|
+
# any later version. #
|
15
|
+
# #
|
16
|
+
# Lingo is distributed in the hope that it will be useful, but WITHOUT ANY #
|
17
|
+
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS #
|
18
|
+
# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for #
|
19
|
+
# more details. #
|
20
|
+
# #
|
21
|
+
# You should have received a copy of the GNU Affero General Public License #
|
22
|
+
# along with Lingo. If not, see <http://www.gnu.org/licenses/>. #
|
23
|
+
# #
|
24
|
+
###############################################################################
|
25
|
+
#++
|
26
|
+
|
27
|
+
class Lingo
|
28
|
+
|
29
|
+
class Database
|
30
|
+
|
31
|
+
class Source
|
32
|
+
|
33
|
+
# Abgeleitet von Source behandelt die Klasse Dateien mit dem Format <tt>MultiValue</tt>.
|
34
|
+
# Eine Zeile <tt>"Triumph;Sieg;Erfolg\n"</tt> wird gewandelt in <tt>[ nil, ['triumph', 'sieg', 'erfolg'] ]</tt>.
|
35
|
+
# Der Trenner zwischen Schlüssel und Projektion kann über den Parameter <tt>separator</tt> geändert werden.
|
36
|
+
|
37
|
+
class MultiValue < self
|
38
|
+
|
39
|
+
def initialize(id, lingo)
|
40
|
+
super
|
41
|
+
|
42
|
+
@separator = @config.fetch('separator', ';')
|
43
|
+
@line_pattern = Regexp.new('^' + @legal_word + '(?:' + Regexp.escape(@separator) + @legal_word + ')*$')
|
44
|
+
|
45
|
+
@idx = -1
|
46
|
+
end
|
47
|
+
|
48
|
+
def set(db, key, val)
|
49
|
+
db[key = "#{IDX_REF}#{@idx += 1}"] = val
|
50
|
+
val.each { |v| db[v] = [key] }
|
51
|
+
end
|
52
|
+
|
53
|
+
private
|
54
|
+
|
55
|
+
def convert_line(line, key, val)
|
56
|
+
[nil, line.split(@separator).map { |value| value.strip }]
|
57
|
+
end
|
58
|
+
|
59
|
+
end
|
60
|
+
|
61
|
+
end
|
62
|
+
|
63
|
+
end
|
64
|
+
|
65
|
+
end
|
@@ -0,0 +1,60 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
#--
|
4
|
+
###############################################################################
|
5
|
+
# #
|
6
|
+
# Lingo -- A full-featured automatic indexing system #
|
7
|
+
# #
|
8
|
+
# Copyright (C) 2005-2007 John Vorhauer #
|
9
|
+
# Copyright (C) 2007-2012 John Vorhauer, Jens Wille #
|
10
|
+
# #
|
11
|
+
# Lingo is free software; you can redistribute it and/or modify it under the #
|
12
|
+
# terms of the GNU Affero General Public License as published by the Free #
|
13
|
+
# Software Foundation; either version 3 of the License, or (at your option) #
|
14
|
+
# any later version. #
|
15
|
+
# #
|
16
|
+
# Lingo is distributed in the hope that it will be useful, but WITHOUT ANY #
|
17
|
+
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS #
|
18
|
+
# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for #
|
19
|
+
# more details. #
|
20
|
+
# #
|
21
|
+
# You should have received a copy of the GNU Affero General Public License #
|
22
|
+
# along with Lingo. If not, see <http://www.gnu.org/licenses/>. #
|
23
|
+
# #
|
24
|
+
###############################################################################
|
25
|
+
#++
|
26
|
+
|
27
|
+
class Lingo
|
28
|
+
|
29
|
+
class Database
|
30
|
+
|
31
|
+
class Source
|
32
|
+
|
33
|
+
# Abgeleitet von Source behandelt die Klasse Dateien mit dem Format <tt>SingleWord</tt>.
|
34
|
+
# Eine Zeile <tt>"Fachbegriff\n"</tt> wird gewandelt in <tt>[ 'fachbegriff', ['#s'] ]</tt>.
|
35
|
+
# Die Wortklasse kann über den Parameter <tt>def-wc</tt> beeinflusst werden.
|
36
|
+
|
37
|
+
class SingleWord < self
|
38
|
+
|
39
|
+
def initialize(id, lingo)
|
40
|
+
super
|
41
|
+
|
42
|
+
@wc = @config.fetch('def-wc', 's').downcase
|
43
|
+
@mul_wc = @config.fetch('def-mul-wc', @wc).downcase
|
44
|
+
|
45
|
+
@line_pattern = %r{^(#{@legal_word})$}
|
46
|
+
end
|
47
|
+
|
48
|
+
private
|
49
|
+
|
50
|
+
def convert_line(line, key, val)
|
51
|
+
[key = key.strip, %W[##{key =~ /\s/ ? @mul_wc : @wc}]]
|
52
|
+
end
|
53
|
+
|
54
|
+
end
|
55
|
+
|
56
|
+
end
|
57
|
+
|
58
|
+
end
|
59
|
+
|
60
|
+
end
|
@@ -0,0 +1,64 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
#--
|
4
|
+
###############################################################################
|
5
|
+
# #
|
6
|
+
# Lingo -- A full-featured automatic indexing system #
|
7
|
+
# #
|
8
|
+
# Copyright (C) 2005-2007 John Vorhauer #
|
9
|
+
# Copyright (C) 2007-2012 John Vorhauer, Jens Wille #
|
10
|
+
# #
|
11
|
+
# Lingo is free software; you can redistribute it and/or modify it under the #
|
12
|
+
# terms of the GNU Affero General Public License as published by the Free #
|
13
|
+
# Software Foundation; either version 3 of the License, or (at your option) #
|
14
|
+
# any later version. #
|
15
|
+
# #
|
16
|
+
# Lingo is distributed in the hope that it will be useful, but WITHOUT ANY #
|
17
|
+
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS #
|
18
|
+
# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for #
|
19
|
+
# more details. #
|
20
|
+
# #
|
21
|
+
# You should have received a copy of the GNU Affero General Public License #
|
22
|
+
# along with Lingo. If not, see <http://www.gnu.org/licenses/>. #
|
23
|
+
# #
|
24
|
+
###############################################################################
|
25
|
+
#++
|
26
|
+
|
27
|
+
class Lingo
|
28
|
+
|
29
|
+
class Database
|
30
|
+
|
31
|
+
class Source
|
32
|
+
|
33
|
+
# Abgeleitet von Source behandelt die Klasse Dateien mit dem Format <tt>WordClass</tt>.
|
34
|
+
# Eine Zeile <tt>"essen,essen #v essen #o esse #s\n"</tt> wird gewandelt in <tt>[ 'essen', ['esse#s', 'essen#v', 'essen#o'] ]</tt>.
|
35
|
+
# Der Trenner zwischen Schlüssel und Projektion kann über den Parameter <tt>separator</tt> geändert werden.
|
36
|
+
|
37
|
+
class WordClass < self
|
38
|
+
|
39
|
+
def initialize(id, lingo)
|
40
|
+
super
|
41
|
+
|
42
|
+
@separator = @config.fetch('separator', ',')
|
43
|
+
@line_pattern = Regexp.new('^(' + @legal_word + ')' + Regexp.escape(@separator) + '((?:' + @legal_word + '#\w)+)$')
|
44
|
+
end
|
45
|
+
|
46
|
+
private
|
47
|
+
|
48
|
+
def convert_line(line, key, val)
|
49
|
+
key, valstr = key.strip, val.strip
|
50
|
+
val = valstr.gsub(/\s+#/, '#').scan(/\S.+?\s*#\w/)
|
51
|
+
val = val.map do |str|
|
52
|
+
str =~ /^(.+)#(.)/
|
53
|
+
($1 == key ? '' : $1) + '#' + $2
|
54
|
+
end
|
55
|
+
[key, val]
|
56
|
+
end
|
57
|
+
|
58
|
+
end
|
59
|
+
|
60
|
+
end
|
61
|
+
|
62
|
+
end
|
63
|
+
|
64
|
+
end
|
data/lib/lingo/error.rb
ADDED
@@ -0,0 +1,122 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
#--
|
4
|
+
###############################################################################
|
5
|
+
# #
|
6
|
+
# Lingo -- A full-featured automatic indexing system #
|
7
|
+
# #
|
8
|
+
# Copyright (C) 2005-2007 John Vorhauer #
|
9
|
+
# Copyright (C) 2007-2012 John Vorhauer, Jens Wille #
|
10
|
+
# #
|
11
|
+
# Lingo is free software; you can redistribute it and/or modify it under the #
|
12
|
+
# terms of the GNU Affero General Public License as published by the Free #
|
13
|
+
# Software Foundation; either version 3 of the License, or (at your option) #
|
14
|
+
# any later version. #
|
15
|
+
# #
|
16
|
+
# Lingo is distributed in the hope that it will be useful, but WITHOUT ANY #
|
17
|
+
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS #
|
18
|
+
# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for #
|
19
|
+
# more details. #
|
20
|
+
# #
|
21
|
+
# You should have received a copy of the GNU Affero General Public License #
|
22
|
+
# along with Lingo. If not, see <http://www.gnu.org/licenses/>. #
|
23
|
+
# #
|
24
|
+
###############################################################################
|
25
|
+
#++
|
26
|
+
|
27
|
+
class Lingo
|
28
|
+
|
29
|
+
class LingoError < StandardError; end
|
30
|
+
|
31
|
+
class NoWritableStoreError < LingoError
|
32
|
+
|
33
|
+
attr_reader :file, :path
|
34
|
+
|
35
|
+
def initialize(file, path)
|
36
|
+
@file, @path = file, path
|
37
|
+
end
|
38
|
+
|
39
|
+
def to_s
|
40
|
+
'No writable store found in search path'
|
41
|
+
end
|
42
|
+
|
43
|
+
end
|
44
|
+
|
45
|
+
class ConfigError < LingoError
|
46
|
+
|
47
|
+
attr_reader :id
|
48
|
+
|
49
|
+
def initialize(id)
|
50
|
+
@id = id
|
51
|
+
end
|
52
|
+
|
53
|
+
end
|
54
|
+
|
55
|
+
class ConfigLoadError < ConfigError
|
56
|
+
|
57
|
+
attr_reader :err
|
58
|
+
|
59
|
+
def initialize(err)
|
60
|
+
@err = err
|
61
|
+
end
|
62
|
+
|
63
|
+
def to_s
|
64
|
+
"Error loading config: #{err}"
|
65
|
+
end
|
66
|
+
|
67
|
+
end
|
68
|
+
|
69
|
+
class NoDatabaseConfigError < ConfigError
|
70
|
+
|
71
|
+
def to_s
|
72
|
+
"No such database `#{id}' defined."
|
73
|
+
end
|
74
|
+
|
75
|
+
end
|
76
|
+
|
77
|
+
class InvalidDatabaseConfigError < ConfigError
|
78
|
+
|
79
|
+
def to_s
|
80
|
+
"Invalid database configuration `#{id}'."
|
81
|
+
end
|
82
|
+
|
83
|
+
end
|
84
|
+
|
85
|
+
class MissingConfigError < ConfigError
|
86
|
+
|
87
|
+
def to_s
|
88
|
+
"Missing configuration for `#{id}'."
|
89
|
+
end
|
90
|
+
|
91
|
+
end
|
92
|
+
|
93
|
+
class FileNotFoundError < LingoError
|
94
|
+
|
95
|
+
attr_reader :name
|
96
|
+
|
97
|
+
def initialize(name)
|
98
|
+
@name = name
|
99
|
+
end
|
100
|
+
|
101
|
+
def to_s
|
102
|
+
"No such file `#{name}'."
|
103
|
+
end
|
104
|
+
|
105
|
+
end
|
106
|
+
|
107
|
+
class SourceFileNotFoundError < FileNotFoundError
|
108
|
+
|
109
|
+
attr_reader :id
|
110
|
+
|
111
|
+
def initialize(name, id)
|
112
|
+
super(name)
|
113
|
+
@id = id
|
114
|
+
end
|
115
|
+
|
116
|
+
def to_s
|
117
|
+
"No such source file `#{name}' for `#{id}'."
|
118
|
+
end
|
119
|
+
|
120
|
+
end
|
121
|
+
|
122
|
+
end
|
data/lib/lingo/language.rb
CHANGED
@@ -1,529 +1,89 @@
|
|
1
1
|
# encoding: utf-8
|
2
2
|
|
3
3
|
#--
|
4
|
-
|
5
|
-
#
|
6
|
-
#
|
7
|
-
#
|
8
|
-
# Copyright (C) 2007
|
9
|
-
#
|
10
|
-
#
|
11
|
-
#
|
12
|
-
#
|
13
|
-
#
|
14
|
-
#
|
15
|
-
#
|
16
|
-
#
|
17
|
-
#
|
18
|
-
#
|
19
|
-
#
|
20
|
-
#
|
21
|
-
#
|
22
|
-
#
|
23
|
-
#
|
24
|
-
|
25
|
-
# welcomeATlex-lingoDOTde near 50°55'N+6°55'E.
|
26
|
-
#
|
27
|
-
# Lex Lingo rules from here on
|
4
|
+
###############################################################################
|
5
|
+
# #
|
6
|
+
# Lingo -- A full-featured automatic indexing system #
|
7
|
+
# #
|
8
|
+
# Copyright (C) 2005-2007 John Vorhauer #
|
9
|
+
# Copyright (C) 2007-2012 John Vorhauer, Jens Wille #
|
10
|
+
# #
|
11
|
+
# Lingo is free software; you can redistribute it and/or modify it under the #
|
12
|
+
# terms of the GNU Affero General Public License as published by the Free #
|
13
|
+
# Software Foundation; either version 3 of the License, or (at your option) #
|
14
|
+
# any later version. #
|
15
|
+
# #
|
16
|
+
# Lingo is distributed in the hope that it will be useful, but WITHOUT ANY #
|
17
|
+
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS #
|
18
|
+
# FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for #
|
19
|
+
# more details. #
|
20
|
+
# #
|
21
|
+
# You should have received a copy of the GNU Affero General Public License #
|
22
|
+
# along with Lingo. If not, see <http://www.gnu.org/licenses/>. #
|
23
|
+
# #
|
24
|
+
###############################################################################
|
28
25
|
#++
|
29
26
|
|
30
|
-
require_relative '
|
31
|
-
require_relative '
|
32
|
-
require_relative '
|
27
|
+
require_relative 'language/lexical_hash'
|
28
|
+
require_relative 'language/dictionary'
|
29
|
+
require_relative 'language/grammar'
|
30
|
+
require_relative 'language/word_form'
|
31
|
+
require_relative 'language/token'
|
32
|
+
require_relative 'language/lexical'
|
33
|
+
require_relative 'language/word'
|
33
34
|
|
34
35
|
class Lingo
|
35
36
|
|
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
|
-
when /^#(.)$/
|
87
|
-
# Alleinige Angabe der Wortklasse => Ergebniswort ist gleich dem Schlüssel
|
88
|
-
Lexical.new(key, $1)
|
89
|
-
when /^([^#]+?)\s*#(.)$/
|
90
|
-
# Angabe Ergebniswort und Wortklasse
|
91
|
-
Lexical.new($1, $2)
|
92
|
-
when /^([^#]+)$/
|
93
|
-
# Angabe Ergebniswort ohne Wortklasse
|
94
|
-
Lexical.new($1, @wordclass)
|
95
|
-
else
|
96
|
-
str
|
97
|
-
end
|
98
|
-
end.compact.sort.uniq unless record.nil?
|
99
|
-
|
100
|
-
# Ergebnis zurückgeben
|
101
|
-
inc('data found') unless record.nil?
|
102
|
-
store(key, record)
|
103
|
-
end
|
104
|
-
|
105
|
-
end
|
106
|
-
|
107
|
-
class Dictionary
|
108
|
-
|
109
|
-
include Cachable
|
110
|
-
include Reportable
|
111
|
-
|
112
|
-
def initialize(config, lingo)
|
113
|
-
init_reportable
|
114
|
-
init_cachable
|
115
|
-
|
116
|
-
dictionary_config = lingo.dictionary_config
|
117
|
-
|
118
|
-
# Parameter prüfen
|
119
|
-
raise "Keine Sprach-Konfiguration angegeben!" if dictionary_config.nil?
|
120
|
-
raise "Keine Parameter angegeben!" if config.nil?
|
121
|
-
raise "Keine Datenquellen angegeben!" unless config.has_key?('source')
|
122
|
-
|
123
|
-
# Parameter auslesen
|
124
|
-
@all_sources = (config['mode'].nil? || config['mode'].downcase=='all')
|
125
|
-
|
126
|
-
@sources = config['source'].map { |src| LexicalHash.new(src, lingo) }
|
127
|
-
|
128
|
-
lingo.dictionaries << self
|
129
|
-
|
130
|
-
# Parameter aus de.lang:language/dictionary auslesen
|
131
|
-
@suffixes = []
|
132
|
-
@infixes = []
|
133
|
-
|
134
|
-
dictionary_config['suffix'].each {|arr|
|
135
|
-
typ, sufli = arr
|
136
|
-
typ.downcase!
|
137
|
-
sufli.split.each {|suf|
|
138
|
-
su, ex = suf.split('/')
|
139
|
-
fix = [Regexp.new(su+'$', 'i'), ex.nil? ? '*' : ex, typ]
|
140
|
-
(typ=='f' ? @infixes : @suffixes) << fix
|
141
|
-
}
|
142
|
-
} if dictionary_config.has_key?( 'suffix' )
|
143
|
-
end
|
144
|
-
|
145
|
-
def close
|
146
|
-
@sources.each(&:close)
|
147
|
-
end
|
148
|
-
|
149
|
-
def report
|
150
|
-
super.tap { |rep| @sources.each { |src| rep.update(src.report) } }
|
151
|
-
end
|
152
|
-
|
153
|
-
# _dic_.find_word( _aString_ ) -> _aNewWord_
|
154
|
-
#
|
155
|
-
# Erstellt aus dem String ein Wort und sucht nach diesem im Wörterbuch.
|
156
|
-
def find_word(string)
|
157
|
-
# Cache abfragen
|
158
|
-
key = string.downcase
|
159
|
-
if hit?(key)
|
160
|
-
inc('cache hits')
|
161
|
-
word = retrieve(key)
|
162
|
-
word.form = string
|
163
|
-
return word
|
164
|
-
end
|
165
|
-
|
166
|
-
word = Word.new(string, WA_UNKNOWN)
|
167
|
-
lexicals = select_with_suffix(string)
|
168
|
-
unless lexicals.empty?
|
169
|
-
word.lexicals = lexicals
|
170
|
-
word.attr = WA_IDENTIFIED
|
171
|
-
end
|
172
|
-
store(key, word)
|
173
|
-
end
|
174
|
-
|
175
|
-
def find_synonyms(obj)
|
176
|
-
# alle Lexicals des Wortes
|
177
|
-
lexis = obj.lexicals
|
178
|
-
lexis = [obj] if lexis.empty? && obj.attr==WA_UNKNOWN
|
179
|
-
# alle gefundenen Synonyme
|
180
|
-
synos = []
|
181
|
-
# multiworder optimization
|
182
|
-
key_ref = %r{\A#{Regexp.escape(KEY_REF)}\d+}o
|
183
|
-
|
184
|
-
lexis.each do |lex|
|
185
|
-
# Synonyme für Teile eines Kompositum ausschließen
|
186
|
-
next if obj.attr==WA_KOMPOSITUM && lex.attr!=LA_KOMPOSITUM
|
187
|
-
# Synonyme für Synonyme ausschließen
|
188
|
-
next if lex.attr==LA_SYNONYM
|
189
|
-
|
190
|
-
select(lex.form).each do |syn|
|
191
|
-
synos << syn unless syn =~ key_ref
|
192
|
-
end
|
193
|
-
end
|
194
|
-
|
195
|
-
synos
|
196
|
-
end
|
197
|
-
|
198
|
-
# _dic_.select( _aString_ ) -> _ArrayOfLexicals_
|
199
|
-
#
|
200
|
-
# Sucht alle Wörterbücher durch und gibt den ersten Treffer zurück (+mode = first+), oder alle Treffer (+mode = all+)
|
201
|
-
def select(string)
|
202
|
-
lexicals = []
|
203
|
-
|
204
|
-
@sources.each { |src|
|
205
|
-
if lexis = src[string]
|
206
|
-
lexicals += lexis
|
207
|
-
break unless @all_sources
|
208
|
-
end
|
209
|
-
}
|
210
|
-
|
211
|
-
lexicals.sort.uniq
|
212
|
-
end
|
213
|
-
|
214
|
-
# _dic_.select_with_suffix( _aString_ ) -> _ArrayOfLexicals_
|
215
|
-
#
|
216
|
-
# Sucht alle Wörterbücher durch und gibt den ersten Treffer zurück (+mode = first+), oder alle Treffer (+mode = all+).
|
217
|
-
# Sucht dabei auch Wörter, die um wortklassenspezifische Suffixe bereinigt wurden.
|
218
|
-
def select_with_suffix(string)
|
219
|
-
lexicals = select(string)
|
220
|
-
if lexicals.empty?
|
221
|
-
suffix_lexicals(string).each { |suflex|
|
222
|
-
select(suflex.form).each { |srclex|
|
223
|
-
lexicals << srclex if suflex.attr == srclex.attr
|
224
|
-
}
|
225
|
-
}
|
226
|
-
end
|
227
|
-
lexicals
|
228
|
-
end
|
229
|
-
|
230
|
-
# _dic_.select_with_infix( _aString_ ) -> _ArrayOfLexicals_
|
231
|
-
#
|
232
|
-
# Sucht alle Wörterbücher durch und gibt den ersten Treffer zurück (+mode = first+), oder alle Treffer (+mode = all+).
|
233
|
-
# Sucht dabei auch Wörter, die eine Fugung am Ende haben.
|
234
|
-
def select_with_infix(string)
|
235
|
-
lexicals = select(string)
|
236
|
-
if lexicals.size == 0
|
237
|
-
infix_lexicals(string).each { |inlex|
|
238
|
-
select(inlex.form).each { |srclex|
|
239
|
-
lexicals << srclex
|
240
|
-
}
|
241
|
-
}
|
242
|
-
end
|
243
|
-
lexicals
|
244
|
-
end
|
245
|
-
|
246
|
-
# _dic_.suffix_lexicals( _aString_ ) -> _ArrayOfLexicals_
|
247
|
-
#
|
248
|
-
# Gibt alle möglichen Lexicals zurück, die von der Endung her auf den String anwendbar sind:
|
249
|
-
#
|
250
|
-
# dic.suffix_lexicals("Hasens") -> [(hasen/s), (hasen/e), (has/e)]
|
251
|
-
def suffix_lexicals(string)
|
252
|
-
lexicals = []
|
253
|
-
newform = regex = ext = type = nil
|
254
|
-
@suffixes.each { |suf|
|
255
|
-
regex, ext, type = suf
|
256
|
-
if string =~ regex
|
257
|
-
newform = $`+((ext=="*")?'':ext)+$'
|
258
|
-
lexicals << Lexical.new(newform, type)
|
259
|
-
end
|
260
|
-
}
|
261
|
-
lexicals
|
262
|
-
end
|
263
|
-
|
264
|
-
# _dic_.gap_lexicals( _aString_ ) -> _ArrayOfLexicals_
|
265
|
-
#
|
266
|
-
# Gibt alle möglichen Lexicals zurück, die von der Endung her auf den String anwendbar sind:
|
267
|
-
def infix_lexicals(string)
|
268
|
-
lexicals = []
|
269
|
-
newform = regex = ext = type = nil
|
270
|
-
@infixes.each { |suf|
|
271
|
-
regex, ext, type = suf
|
272
|
-
if string =~ regex
|
273
|
-
newform = $`+((ext=="*")?'':ext)+$'
|
274
|
-
lexicals << Lexical.new(newform, type)
|
275
|
-
end
|
276
|
-
}
|
277
|
-
lexicals
|
278
|
-
end
|
279
|
-
|
280
|
-
end
|
281
|
-
|
282
|
-
class Compositum
|
283
|
-
end
|
284
|
-
|
285
|
-
# Die Klasse Grammar beinhaltet grammatikalische Spezialitäten einer Sprache. Derzeit findet die
|
286
|
-
# Kompositumerkennung hier ihren Platz, die mit der Methode find_compositum aufgerufen werden kann.
|
287
|
-
# Die Klasse Grammar wird genau wie ein Dictionary initialisiert. Das bei der Initialisierung angegebene Wörterbuch ist Grundlage
|
288
|
-
# für die Erkennung der Kompositumteile.
|
289
|
-
|
290
|
-
class Grammar
|
291
|
-
|
292
|
-
# Ergebnisse der Kompositumerkennung werden gespeichert und bei erneutem Aufruf mit gleichem Suchwort genutzt
|
293
|
-
include Cachable
|
294
|
-
|
295
|
-
# Die Verarbeitung wird statistisch erfasst und mit der Option -s angezeigt
|
296
|
-
include Reportable
|
297
|
-
|
298
|
-
# initialize(config, dictionary_config) -> _Grammar_
|
299
|
-
# config = Attendee-spezifische Parameter
|
300
|
-
# dictionary_config = Datenbankkonfiguration aus de.lang
|
301
|
-
def initialize(config, lingo)
|
302
|
-
init_reportable
|
303
|
-
init_cachable
|
304
|
-
|
305
|
-
@dictionary = Dictionary.new(config, lingo)
|
306
|
-
|
307
|
-
# Sprachspezifische Einstellungen für Kompositumverarbeitung laden (die nachfolgenden Werte können in der
|
308
|
-
# Konfigurationsdatei de.lang nach belieben angepasst werden)
|
309
|
-
comp = lingo.dictionary_config['compositum']
|
310
|
-
|
311
|
-
# Ein Wort muss mindestens 8 Zeichen lang sein, damit überhaupt eine Prüfung stattfindet.
|
312
|
-
@comp_min_word_size = (comp['min-word-size'] || '8').to_i
|
313
|
-
|
314
|
-
# Die durchschnittliche Länge der Kompositum-Wortteile muss mindestens 4 Zeichen lang sein, sonst ist es kein
|
315
|
-
# gültiges Kompositum.
|
316
|
-
@comp_min_avg_part_size = (comp['min-avg-part-size'] || '4').to_i
|
317
|
-
|
318
|
-
# Der kürzeste Kompositum-Wortteil muss mindestens 1 Zeichen lang sein
|
319
|
-
@comp_min_part_size = (comp['min-part-size'] || '1').to_i
|
320
|
-
|
321
|
-
# Ein Kompositum darf aus höchstens 4 Wortteilen bestehen
|
322
|
-
@comp_max_parts = (comp['max-parts'] || '4').to_i
|
323
|
-
|
324
|
-
# Die Wortklasse eines Kompositum-Wortteils kann separat gekennzeichnet werden, um sie von Wortklassen normaler Wörter
|
325
|
-
# unterscheiden zu können z.B. Hausmeister => ['haus/s', 'meister/s'] oder Hausmeister => ['haus/s+', 'meister/s+'] mit
|
326
|
-
# append-wordclass = '+'
|
327
|
-
@append_wc = comp.fetch( 'append-wordclass', '' )
|
328
|
-
|
329
|
-
# Bestimmte Sequenzen können als ungültige Komposita erkannt werden, z.B. ist ein Kompositum aus zwei Adjetiven kein
|
330
|
-
# Kompositum, also skip-sequence = 'aa'
|
331
|
-
@sequences = comp.fetch( 'skip-sequences', [] ).collect { |sq| sq.downcase }
|
332
|
-
|
333
|
-
# Liste der Vorschläge für eine Zerlegung
|
334
|
-
@suggestions = []
|
335
|
-
end
|
336
|
-
|
337
|
-
def close
|
338
|
-
@dictionary.close
|
339
|
-
end
|
340
|
-
|
341
|
-
alias_method :report_grammar, :report
|
342
|
-
|
343
|
-
def report
|
344
|
-
rep = report_grammar
|
345
|
-
rep.update(@dictionary.report)
|
346
|
-
rep
|
347
|
-
end
|
348
|
-
|
349
|
-
# find_compositum(string) -> word wenn level=1
|
350
|
-
# find_compositum(string) -> [lexicals, stats] wenn level!=1
|
351
|
-
#
|
352
|
-
# find_compositum arbeitet in verschiedenen Leveln, da die Methode auch rekursiv aufgerufen wird. Ein Level größer 1
|
353
|
-
# entspricht daher einem rekursiven Aufruf
|
354
|
-
def find_compositum(string, level=1, has_tail=false)
|
355
|
-
# Prüfen, ob string bereits auf Kompositum getestet wurde. Wenn ja, dann Ergebnis des letztes Aufrufs zurück geben.
|
356
|
-
key = string.downcase
|
357
|
-
if level == 1 && hit?(key)
|
358
|
-
inc('cache hits')
|
359
|
-
return retrieve(key)
|
360
|
-
end
|
361
|
-
|
362
|
-
# Ergebnis vorbelegen
|
363
|
-
comp = Word.new(string, WA_UNKNOWN)
|
364
|
-
|
365
|
-
# Validitätsprüfung: nur Strings mit Mindestlänge auf Kompositum prüfen
|
366
|
-
if string.size <= @comp_min_word_size
|
367
|
-
inc('String zu kurz')
|
368
|
-
return (level==1) ? comp : [[],[],'']
|
369
|
-
end
|
370
|
-
|
371
|
-
# Kompositumerkennung initialisieren
|
372
|
-
inc('Komposita geprüft')
|
373
|
-
stats, lexis, seqs = permute_compositum(string.downcase, level, has_tail)
|
374
|
-
|
375
|
-
if level==1
|
376
|
-
# Auf Level 1 Kompositum zurück geben
|
377
|
-
if lexis.size > 0 && is_valid?( string, stats, lexis, seqs )
|
378
|
-
inc('Komposita erkannt')
|
379
|
-
comp.attr = WA_KOMPOSITUM
|
380
|
-
comp.lexicals = lexis.collect do |lex|
|
381
|
-
(lex.attr==LA_KOMPOSITUM) ? lex : Lexical.new(lex.form, lex.attr+@append_wc)
|
382
|
-
end
|
383
|
-
end
|
384
|
-
|
385
|
-
return store(key, comp)
|
386
|
-
end
|
387
|
-
|
388
|
-
# Validitätsprüfung
|
389
|
-
if lexis.size > 0 && is_valid?(string, stats, lexis, seqs)
|
390
|
-
[stats, lexis, seqs]
|
391
|
-
else
|
392
|
-
[[],[],'']
|
393
|
-
end
|
394
|
-
end
|
395
|
-
|
396
|
-
private
|
397
|
-
|
398
|
-
def is_valid?(string, stats, lexis, seqs)
|
399
|
-
is_valid = true
|
400
|
-
is_valid &&= (stats.size <= @comp_max_parts)
|
401
|
-
is_valid &&= (stats.sort[0] >= @comp_min_part_size)
|
402
|
-
is_valid &&= (string.size/stats.size) >= @comp_min_avg_part_size
|
403
|
-
is_valid &&= @sequences.index( seqs ).nil? unless @sequences.empty?
|
404
|
-
is_valid
|
405
|
-
end
|
406
|
-
|
407
|
-
# permute_string( _aString_ ) -> [lexicals, stats, seqs]
|
408
|
-
def permute_compositum(string, level, has_tail)
|
409
|
-
@suggestions[level] = [] if @suggestions[level].nil?
|
410
|
-
|
411
|
-
# Finde letzten Bindesstrich im Wort
|
412
|
-
if string =~ /^(.+)-([^-]+)$/
|
413
|
-
test_compositum($1, '-', $2, level, has_tail)
|
414
|
-
else
|
415
|
-
length = string.length
|
416
|
-
|
417
|
-
# Wortteilungen testen
|
418
|
-
1.upto(length - 1) do |p|
|
419
|
-
# String teilen und testen
|
420
|
-
fr_str, ba_str = string.slice(0...p), string.slice(p...length)
|
421
|
-
stats, lexis, seqs = test_compositum(fr_str, '', ba_str, level, has_tail)
|
422
|
-
|
423
|
-
unless lexis.empty?
|
424
|
-
if lexis[-1].attr==LA_TAKEITASIS
|
425
|
-
# => halbes Kompositum
|
426
|
-
@suggestions[level] << [stats, lexis, seqs]
|
427
|
-
else
|
428
|
-
# => ganzes Kompositum
|
429
|
-
return [stats, lexis, seqs]
|
430
|
-
end
|
431
|
-
end
|
432
|
-
end
|
433
|
-
|
434
|
-
# alle Wortteilungen durchprobiert und noch immer kein definitives Kompositum erkannt. Dann nehme besten Vorschlag.
|
435
|
-
if @suggestions[level].empty?
|
436
|
-
[[],[],'']
|
437
|
-
else
|
438
|
-
stats, lexis, seqs = @suggestions[level][0]
|
439
|
-
@suggestions[level].clear
|
440
|
-
[stats, lexis, seqs]
|
441
|
-
end
|
442
|
-
end
|
443
|
-
end
|
444
|
-
|
445
|
-
# test_compositum() -> [stats, lexicals, seq]
|
446
|
-
#
|
447
|
-
# Testet einen definiert zerlegten String auf Kompositum
|
448
|
-
def test_compositum(front_string, infix, back_string, level, has_tail)
|
449
|
-
# Statistik merken für Validitätsprüfung
|
450
|
-
stats = [front_string.size, back_string.size]
|
451
|
-
seqs = ['?', '?']
|
452
|
-
|
453
|
-
# zuerst hinteren Teil auflösen
|
454
|
-
# 1. Möglichkeit: Wort mit oder ohne Suffix
|
455
|
-
back_lexicals = @dictionary.select_with_suffix(back_string)
|
456
|
-
unless back_lexicals.empty?
|
457
|
-
back_form = has_tail ? back_string : back_lexicals.sort[0].form
|
458
|
-
seqs[1] = back_lexicals.sort[0].attr
|
459
|
-
end
|
460
|
-
|
461
|
-
# 2. Möglichkeit: Wort mit oder ohne Infix, wenn es nicht der letzte Teil des Wortes ist
|
462
|
-
if back_lexicals.empty? && has_tail
|
463
|
-
back_lexicals = @dictionary.select_with_infix(back_string)
|
464
|
-
unless back_lexicals.empty?
|
465
|
-
back_form = back_string
|
466
|
-
seqs[1] = back_lexicals.sort[0].attr
|
467
|
-
end
|
468
|
-
end
|
469
|
-
|
470
|
-
# 3. Möglichkeit: Selber ein Kompositum (nur im Bindestrich-Fall!)
|
471
|
-
if back_lexicals.empty? && infix=='-'
|
472
|
-
back_stats, back_lexicals, back_seqs = find_compositum(back_string, level+1, has_tail)
|
473
|
-
unless back_lexicals.empty?
|
474
|
-
back_form = back_lexicals.sort[0].form
|
475
|
-
seqs[1] = back_seqs
|
476
|
-
stats = stats[0..0] + back_stats
|
477
|
-
end
|
478
|
-
end
|
479
|
-
|
480
|
-
# 4. Möglichkeit: Take it as is [Nimm's, wie es ist] (nur im Bindestrich-Fall!)
|
481
|
-
if back_lexicals.empty? && infix=='-'
|
482
|
-
back_lexicals = [Lexical.new(back_string, LA_TAKEITASIS)]
|
483
|
-
back_form = back_string
|
484
|
-
seqs[1] = back_lexicals.sort[0].attr
|
485
|
-
end
|
486
|
-
|
487
|
-
# wenn immer noch nicht erkannt, dann sofort zurück
|
488
|
-
return [[],[],''] if back_lexicals.empty?
|
489
|
-
|
490
|
-
# dann vorderen Teil auflösen
|
491
|
-
#
|
492
|
-
# 1. Möglichkeit: Wort mit oder ohne Infix
|
493
|
-
front_lexicals = @dictionary.select_with_infix(front_string)
|
494
|
-
unless front_lexicals.empty?
|
495
|
-
front_form = front_string
|
496
|
-
seqs[0] = front_lexicals.sort[0].attr
|
497
|
-
end
|
498
|
-
|
499
|
-
# 2. Möglichkeit: Selber ein Kompositum
|
500
|
-
if front_lexicals.empty?
|
501
|
-
front_stats, front_lexicals, front_seqs = find_compositum(front_string, level+1, true)
|
502
|
-
unless front_lexicals.empty?
|
503
|
-
front_form = front_lexicals.sort[0].form
|
504
|
-
seqs[0] = front_seqs
|
505
|
-
stats = front_stats + stats[1..-1]
|
506
|
-
end
|
507
|
-
end
|
508
|
-
|
509
|
-
# 3. Möglichkeit: Take it as is [Nimm's, wie es ist] (nur im Bindestrich-Fall!)
|
510
|
-
if front_lexicals.empty? && infix=='-'
|
511
|
-
front_lexicals = [Lexical.new(front_string, LA_TAKEITASIS)]
|
512
|
-
seqs[0] = front_lexicals.sort[0].attr
|
513
|
-
front_form = front_string
|
514
|
-
end
|
515
|
-
|
516
|
-
# wenn immer noch nicht erkannt, dann sofort zurück
|
517
|
-
return [[],[],''] if front_lexicals.empty?
|
518
|
-
|
519
|
-
# Kompositum gefunden, Grundform bilden
|
520
|
-
lexis = (front_lexicals + back_lexicals).collect { |lex|
|
521
|
-
(lex.attr==LA_KOMPOSITUM) ? nil : lex
|
522
|
-
}.compact
|
523
|
-
lexis << Lexical.new(front_form + infix + back_form, LA_KOMPOSITUM)
|
524
|
-
|
525
|
-
[stats, lexis.sort, seqs.join ]
|
526
|
-
end
|
37
|
+
module Language
|
38
|
+
|
39
|
+
# String-Konstanten im Datenstrom
|
40
|
+
CHAR_PUNCT = '.'
|
41
|
+
|
42
|
+
TA_WORD = 'WORD'
|
43
|
+
TA_PUNCTUATION = 'PUNC'
|
44
|
+
TA_OTHER = 'OTHR'
|
45
|
+
|
46
|
+
# Standardattribut bei der Initialisierung eines Word-Objektes
|
47
|
+
WA_UNSET = '-'
|
48
|
+
# Status, nachdem das Word im Wörterbuch gefunden wurde
|
49
|
+
WA_IDENTIFIED = 'IDF'
|
50
|
+
# Status, wenn das Word nicht gefunden werden konnte
|
51
|
+
WA_UNKNOWN = '?'
|
52
|
+
# Wort ist als Kompositum erkannt worden
|
53
|
+
WA_KOMPOSITUM = 'KOM'
|
54
|
+
# Wort ist eine Mehrwortgruppe
|
55
|
+
WA_MULTIWORD = 'MUL'
|
56
|
+
# Wort ist eine Mehrwortgruppe
|
57
|
+
WA_SEQUENCE = 'SEQ'
|
58
|
+
# Word ist unbekannt, jedoch Teil einer Mehrwortgruppe
|
59
|
+
WA_UNKMULPART = 'MU?'
|
60
|
+
|
61
|
+
LA_SUBSTANTIV = 's'
|
62
|
+
LA_ADJEKTIV = 'a'
|
63
|
+
LA_VERB = 'v'
|
64
|
+
LA_EIGENNAME = 'e'
|
65
|
+
LA_KOMPOSITUM = 'k'
|
66
|
+
LA_MULTIWORD = 'm'
|
67
|
+
LA_SEQUENCE = 'q'
|
68
|
+
LA_WORTFORM = 'w'
|
69
|
+
LA_SYNONYM = 'y'
|
70
|
+
LA_STOPWORD = 't'
|
71
|
+
LA_TAKEITASIS = 'x'
|
72
|
+
LA_UNKNOWN = '?'
|
73
|
+
|
74
|
+
LA_SORTORDER = [
|
75
|
+
LA_MULTIWORD,
|
76
|
+
LA_KOMPOSITUM,
|
77
|
+
LA_SUBSTANTIV,
|
78
|
+
LA_VERB,
|
79
|
+
LA_ADJEKTIV,
|
80
|
+
LA_EIGENNAME,
|
81
|
+
LA_WORTFORM,
|
82
|
+
LA_STOPWORD,
|
83
|
+
LA_TAKEITASIS,
|
84
|
+
LA_SYNONYM,
|
85
|
+
LA_UNKNOWN
|
86
|
+
].reverse.join
|
527
87
|
|
528
88
|
end
|
529
89
|
|