ruby-sfst 0.4.3 → 0.4.4
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +1 -0
- data/COPYING +280 -0
- data/Gemfile +3 -0
- data/Gemfile.lock +54 -0
- data/README.md +1 -1
- data/Rakefile +9 -18
- data/bin/console +7 -0
- data/bin/setup +6 -0
- data/ext/sfst/alphabet.cc +879 -0
- data/ext/sfst/alphabet.h +302 -0
- data/ext/sfst/basic.cc +85 -0
- data/ext/{sfst_machine → sfst}/basic.h +7 -4
- data/ext/sfst/compact.cc +629 -0
- data/ext/sfst/compact.h +100 -0
- data/ext/sfst/determinise.cc +279 -0
- data/ext/{sfst_machine → sfst}/extconf.rb +2 -1
- data/ext/sfst/fst.cc +1150 -0
- data/ext/sfst/fst.h +374 -0
- data/ext/sfst/hopcroft.cc +681 -0
- data/ext/sfst/interface.cc +1921 -0
- data/ext/sfst/interface.h +171 -0
- data/ext/sfst/make-compact.cc +323 -0
- data/ext/{sfst_machine → sfst}/make-compact.h +15 -13
- data/ext/sfst/mem.h +80 -0
- data/ext/sfst/operators.cc +1273 -0
- data/ext/{sfst_machine → sfst}/sfst_machine.cc +89 -78
- data/ext/sfst/sgi.h +72 -0
- data/ext/sfst/utf8.cc +149 -0
- data/ext/{sfst_machine → sfst}/utf8.h +7 -4
- data/lib/sfst.rb +2 -1
- data/lib/sfst/version.rb +1 -1
- data/ruby-sfst.gemspec +23 -23
- metadata +107 -35
- data/ext/sfst_machine/alphabet.cc +0 -812
- data/ext/sfst_machine/alphabet.h +0 -273
- data/ext/sfst_machine/basic.cc +0 -84
- data/ext/sfst_machine/compact.cc +0 -616
- data/ext/sfst_machine/compact.h +0 -98
- data/ext/sfst_machine/determinise.cc +0 -303
- data/ext/sfst_machine/fst.cc +0 -1000
- data/ext/sfst_machine/fst.h +0 -369
- data/ext/sfst_machine/interface.cc +0 -1842
- data/ext/sfst_machine/interface.h +0 -93
- data/ext/sfst_machine/make-compact.cc +0 -327
- data/ext/sfst_machine/mem.h +0 -74
- data/ext/sfst_machine/operators.cc +0 -1131
- data/ext/sfst_machine/sgi.h +0 -44
- data/ext/sfst_machine/utf8.cc +0 -146
- data/test/test_sfst.fst +0 -3
- data/test/test_sfst.rb +0 -114
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 4892aa62071212d5f65b05afb0476d59852c75c2
|
4
|
+
data.tar.gz: a00ef4a9b22a7b1fda9c0c6e8f89df2a1fe24205
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 194dbde4538a06b21367d1083c7f32d75558539b31e4df0bdcb6a53793ec619e3ace0243e41253117a95a0ec36b26dbe4c81f65ad23e33fe366e4c0d7758a82e
|
7
|
+
data.tar.gz: 2b6074dc37444cd3bdce4c75db8b3dfb4574841c16cc67d82bc39ecc5539f2c1af36b2c0f456e0bf021ac89898efdb8898a4ae4b82dd5221ec6e9724b7ec665e
|
data/CHANGELOG.md
CHANGED
data/COPYING
ADDED
@@ -0,0 +1,280 @@
|
|
1
|
+
GNU GENERAL PUBLIC LICENSE
|
2
|
+
Version 2, June 1991
|
3
|
+
|
4
|
+
Copyright (C) 1989, 1991 Free Software Foundation, Inc.
|
5
|
+
59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
6
|
+
Everyone is permitted to copy and distribute verbatim copies
|
7
|
+
of this license document, but changing it is not allowed.
|
8
|
+
|
9
|
+
Preamble
|
10
|
+
|
11
|
+
The licenses for most software are designed to take away your
|
12
|
+
freedom to share and change it. By contrast, the GNU General Public
|
13
|
+
License is intended to guarantee your freedom to share and change free
|
14
|
+
software--to make sure the software is free for all its users. This
|
15
|
+
General Public License applies to most of the Free Software
|
16
|
+
Foundation's software and to any other program whose authors commit to
|
17
|
+
using it. (Some other Free Software Foundation software is covered by
|
18
|
+
the GNU Library General Public License instead.) You can apply it to
|
19
|
+
your programs, too.
|
20
|
+
|
21
|
+
When we speak of free software, we are referring to freedom, not
|
22
|
+
price. Our General Public Licenses are designed to make sure that you
|
23
|
+
have the freedom to distribute copies of free software (and charge for
|
24
|
+
this service if you wish), that you receive source code or can get it
|
25
|
+
if you want it, that you can change the software or use pieces of it
|
26
|
+
in new free programs; and that you know you can do these things.
|
27
|
+
|
28
|
+
To protect your rights, we need to make restrictions that forbid
|
29
|
+
anyone to deny you these rights or to ask you to surrender the rights.
|
30
|
+
These restrictions translate to certain responsibilities for you if you
|
31
|
+
distribute copies of the software, or if you modify it.
|
32
|
+
|
33
|
+
For example, if you distribute copies of such a program, whether
|
34
|
+
gratis or for a fee, you must give the recipients all the rights that
|
35
|
+
you have. You must make sure that they, too, receive or can get the
|
36
|
+
source code. And you must show them these terms so they know their
|
37
|
+
rights.
|
38
|
+
|
39
|
+
We protect your rights with two steps: (1) copyright the software, and
|
40
|
+
(2) offer you this license which gives you legal permission to copy,
|
41
|
+
distribute and/or modify the software.
|
42
|
+
|
43
|
+
Also, for each author's protection and ours, we want to make certain
|
44
|
+
that everyone understands that there is no warranty for this free
|
45
|
+
software. If the software is modified by someone else and passed on, we
|
46
|
+
want its recipients to know that what they have is not the original, so
|
47
|
+
that any problems introduced by others will not reflect on the original
|
48
|
+
authors' reputations.
|
49
|
+
|
50
|
+
Finally, any free program is threatened constantly by software
|
51
|
+
patents. We wish to avoid the danger that redistributors of a free
|
52
|
+
program will individually obtain patent licenses, in effect making the
|
53
|
+
program proprietary. To prevent this, we have made it clear that any
|
54
|
+
patent must be licensed for everyone's free use or not licensed at all.
|
55
|
+
|
56
|
+
The precise terms and conditions for copying, distribution and
|
57
|
+
modification follow.
|
58
|
+
|
59
|
+
GNU GENERAL PUBLIC LICENSE
|
60
|
+
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
|
61
|
+
|
62
|
+
0. This License applies to any program or other work which contains
|
63
|
+
a notice placed by the copyright holder saying it may be distributed
|
64
|
+
under the terms of this General Public License. The "Program", below,
|
65
|
+
refers to any such program or work, and a "work based on the Program"
|
66
|
+
means either the Program or any derivative work under copyright law:
|
67
|
+
that is to say, a work containing the Program or a portion of it,
|
68
|
+
either verbatim or with modifications and/or translated into another
|
69
|
+
language. (Hereinafter, translation is included without limitation in
|
70
|
+
the term "modification".) Each licensee is addressed as "you".
|
71
|
+
|
72
|
+
Activities other than copying, distribution and modification are not
|
73
|
+
covered by this License; they are outside its scope. The act of
|
74
|
+
running the Program is not restricted, and the output from the Program
|
75
|
+
is covered only if its contents constitute a work based on the
|
76
|
+
Program (independent of having been made by running the Program).
|
77
|
+
Whether that is true depends on what the Program does.
|
78
|
+
|
79
|
+
1. You may copy and distribute verbatim copies of the Program's
|
80
|
+
source code as you receive it, in any medium, provided that you
|
81
|
+
conspicuously and appropriately publish on each copy an appropriate
|
82
|
+
copyright notice and disclaimer of warranty; keep intact all the
|
83
|
+
notices that refer to this License and to the absence of any warranty;
|
84
|
+
and give any other recipients of the Program a copy of this License
|
85
|
+
along with the Program.
|
86
|
+
|
87
|
+
You may charge a fee for the physical act of transferring a copy, and
|
88
|
+
you may at your option offer warranty protection in exchange for a fee.
|
89
|
+
|
90
|
+
2. You may modify your copy or copies of the Program or any portion
|
91
|
+
of it, thus forming a work based on the Program, and copy and
|
92
|
+
distribute such modifications or work under the terms of Section 1
|
93
|
+
above, provided that you also meet all of these conditions:
|
94
|
+
|
95
|
+
a) You must cause the modified files to carry prominent notices
|
96
|
+
stating that you changed the files and the date of any change.
|
97
|
+
|
98
|
+
b) You must cause any work that you distribute or publish, that in
|
99
|
+
whole or in part contains or is derived from the Program or any
|
100
|
+
part thereof, to be licensed as a whole at no charge to all third
|
101
|
+
parties under the terms of this License.
|
102
|
+
|
103
|
+
c) If the modified program normally reads commands interactively
|
104
|
+
when run, you must cause it, when started running for such
|
105
|
+
interactive use in the most ordinary way, to print or display an
|
106
|
+
announcement including an appropriate copyright notice and a
|
107
|
+
notice that there is no warranty (or else, saying that you provide
|
108
|
+
a warranty) and that users may redistribute the program under
|
109
|
+
these conditions, and telling the user how to view a copy of this
|
110
|
+
License. (Exception: if the Program itself is interactive but
|
111
|
+
does not normally print such an announcement, your work based on
|
112
|
+
the Program is not required to print an announcement.)
|
113
|
+
|
114
|
+
These requirements apply to the modified work as a whole. If
|
115
|
+
identifiable sections of that work are not derived from the Program,
|
116
|
+
and can be reasonably considered independent and separate works in
|
117
|
+
themselves, then this License, and its terms, do not apply to those
|
118
|
+
sections when you distribute them as separate works. But when you
|
119
|
+
distribute the same sections as part of a whole which is a work based
|
120
|
+
on the Program, the distribution of the whole must be on the terms of
|
121
|
+
this License, whose permissions for other licensees extend to the
|
122
|
+
entire whole, and thus to each and every part regardless of who wrote it.
|
123
|
+
|
124
|
+
Thus, it is not the intent of this section to claim rights or contest
|
125
|
+
your rights to work written entirely by you; rather, the intent is to
|
126
|
+
exercise the right to control the distribution of derivative or
|
127
|
+
collective works based on the Program.
|
128
|
+
|
129
|
+
In addition, mere aggregation of another work not based on the Program
|
130
|
+
with the Program (or with a work based on the Program) on a volume of
|
131
|
+
a storage or distribution medium does not bring the other work under
|
132
|
+
the scope of this License.
|
133
|
+
|
134
|
+
3. You may copy and distribute the Program (or a work based on it,
|
135
|
+
under Section 2) in object code or executable form under the terms of
|
136
|
+
Sections 1 and 2 above provided that you also do one of the following:
|
137
|
+
|
138
|
+
a) Accompany it with the complete corresponding machine-readable
|
139
|
+
source code, which must be distributed under the terms of Sections
|
140
|
+
1 and 2 above on a medium customarily used for software interchange; or,
|
141
|
+
|
142
|
+
b) Accompany it with a written offer, valid for at least three
|
143
|
+
years, to give any third party, for a charge no more than your
|
144
|
+
cost of physically performing source distribution, a complete
|
145
|
+
machine-readable copy of the corresponding source code, to be
|
146
|
+
distributed under the terms of Sections 1 and 2 above on a medium
|
147
|
+
customarily used for software interchange; or,
|
148
|
+
|
149
|
+
c) Accompany it with the information you received as to the offer
|
150
|
+
to distribute corresponding source code. (This alternative is
|
151
|
+
allowed only for noncommercial distribution and only if you
|
152
|
+
received the program in object code or executable form with such
|
153
|
+
an offer, in accord with Subsection b above.)
|
154
|
+
|
155
|
+
The source code for a work means the preferred form of the work for
|
156
|
+
making modifications to it. For an executable work, complete source
|
157
|
+
code means all the source code for all modules it contains, plus any
|
158
|
+
associated interface definition files, plus the scripts used to
|
159
|
+
control compilation and installation of the executable. However, as a
|
160
|
+
special exception, the source code distributed need not include
|
161
|
+
anything that is normally distributed (in either source or binary
|
162
|
+
form) with the major components (compiler, kernel, and so on) of the
|
163
|
+
operating system on which the executable runs, unless that component
|
164
|
+
itself accompanies the executable.
|
165
|
+
|
166
|
+
If distribution of executable or object code is made by offering
|
167
|
+
access to copy from a designated place, then offering equivalent
|
168
|
+
access to copy the source code from the same place counts as
|
169
|
+
distribution of the source code, even though third parties are not
|
170
|
+
compelled to copy the source along with the object code.
|
171
|
+
|
172
|
+
4. You may not copy, modify, sublicense, or distribute the Program
|
173
|
+
except as expressly provided under this License. Any attempt
|
174
|
+
otherwise to copy, modify, sublicense or distribute the Program is
|
175
|
+
void, and will automatically terminate your rights under this License.
|
176
|
+
However, parties who have received copies, or rights, from you under
|
177
|
+
this License will not have their licenses terminated so long as such
|
178
|
+
parties remain in full compliance.
|
179
|
+
|
180
|
+
5. You are not required to accept this License, since you have not
|
181
|
+
signed it. However, nothing else grants you permission to modify or
|
182
|
+
distribute the Program or its derivative works. These actions are
|
183
|
+
prohibited by law if you do not accept this License. Therefore, by
|
184
|
+
modifying or distributing the Program (or any work based on the
|
185
|
+
Program), you indicate your acceptance of this License to do so, and
|
186
|
+
all its terms and conditions for copying, distributing or modifying
|
187
|
+
the Program or works based on it.
|
188
|
+
|
189
|
+
6. Each time you redistribute the Program (or any work based on the
|
190
|
+
Program), the recipient automatically receives a license from the
|
191
|
+
original licensor to copy, distribute or modify the Program subject to
|
192
|
+
these terms and conditions. You may not impose any further
|
193
|
+
restrictions on the recipients' exercise of the rights granted herein.
|
194
|
+
You are not responsible for enforcing compliance by third parties to
|
195
|
+
this License.
|
196
|
+
|
197
|
+
7. If, as a consequence of a court judgment or allegation of patent
|
198
|
+
infringement or for any other reason (not limited to patent issues),
|
199
|
+
conditions are imposed on you (whether by court order, agreement or
|
200
|
+
otherwise) that contradict the conditions of this License, they do not
|
201
|
+
excuse you from the conditions of this License. If you cannot
|
202
|
+
distribute so as to satisfy simultaneously your obligations under this
|
203
|
+
License and any other pertinent obligations, then as a consequence you
|
204
|
+
may not distribute the Program at all. For example, if a patent
|
205
|
+
license would not permit royalty-free redistribution of the Program by
|
206
|
+
all those who receive copies directly or indirectly through you, then
|
207
|
+
the only way you could satisfy both it and this License would be to
|
208
|
+
refrain entirely from distribution of the Program.
|
209
|
+
|
210
|
+
If any portion of this section is held invalid or unenforceable under
|
211
|
+
any particular circumstance, the balance of the section is intended to
|
212
|
+
apply and the section as a whole is intended to apply in other
|
213
|
+
circumstances.
|
214
|
+
|
215
|
+
It is not the purpose of this section to induce you to infringe any
|
216
|
+
patents or other property right claims or to contest validity of any
|
217
|
+
such claims; this section has the sole purpose of protecting the
|
218
|
+
integrity of the free software distribution system, which is
|
219
|
+
implemented by public license practices. Many people have made
|
220
|
+
generous contributions to the wide range of software distributed
|
221
|
+
through that system in reliance on consistent application of that
|
222
|
+
system; it is up to the author/donor to decide if he or she is willing
|
223
|
+
to distribute software through any other system and a licensee cannot
|
224
|
+
impose that choice.
|
225
|
+
|
226
|
+
This section is intended to make thoroughly clear what is believed to
|
227
|
+
be a consequence of the rest of this License.
|
228
|
+
|
229
|
+
8. If the distribution and/or use of the Program is restricted in
|
230
|
+
certain countries either by patents or by copyrighted interfaces, the
|
231
|
+
original copyright holder who places the Program under this License
|
232
|
+
may add an explicit geographical distribution limitation excluding
|
233
|
+
those countries, so that distribution is permitted only in or among
|
234
|
+
countries not thus excluded. In such case, this License incorporates
|
235
|
+
the limitation as if written in the body of this License.
|
236
|
+
|
237
|
+
9. The Free Software Foundation may publish revised and/or new versions
|
238
|
+
of the General Public License from time to time. Such new versions will
|
239
|
+
be similar in spirit to the present version, but may differ in detail to
|
240
|
+
address new problems or concerns.
|
241
|
+
|
242
|
+
Each version is given a distinguishing version number. If the Program
|
243
|
+
specifies a version number of this License which applies to it and "any
|
244
|
+
later version", you have the option of following the terms and conditions
|
245
|
+
either of that version or of any later version published by the Free
|
246
|
+
Software Foundation. If the Program does not specify a version number of
|
247
|
+
this License, you may choose any version ever published by the Free Software
|
248
|
+
Foundation.
|
249
|
+
|
250
|
+
10. If you wish to incorporate parts of the Program into other free
|
251
|
+
programs whose distribution conditions are different, write to the author
|
252
|
+
to ask for permission. For software which is copyrighted by the Free
|
253
|
+
Software Foundation, write to the Free Software Foundation; we sometimes
|
254
|
+
make exceptions for this. Our decision will be guided by the two goals
|
255
|
+
of preserving the free status of all derivatives of our free software and
|
256
|
+
of promoting the sharing and reuse of software generally.
|
257
|
+
|
258
|
+
NO WARRANTY
|
259
|
+
|
260
|
+
11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
|
261
|
+
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
|
262
|
+
OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
|
263
|
+
PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
|
264
|
+
OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
|
265
|
+
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
|
266
|
+
TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
|
267
|
+
PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
|
268
|
+
REPAIR OR CORRECTION.
|
269
|
+
|
270
|
+
12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
|
271
|
+
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
|
272
|
+
REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
|
273
|
+
INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
|
274
|
+
OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
|
275
|
+
TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
|
276
|
+
YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
|
277
|
+
PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
|
278
|
+
POSSIBILITY OF SUCH DAMAGES.
|
279
|
+
|
280
|
+
END OF TERMS AND CONDITIONS
|
data/Gemfile
ADDED
data/Gemfile.lock
ADDED
@@ -0,0 +1,54 @@
|
|
1
|
+
PATH
|
2
|
+
remote: .
|
3
|
+
specs:
|
4
|
+
ruby-sfst (0.4.4)
|
5
|
+
|
6
|
+
GEM
|
7
|
+
remote: https://rubygems.org/
|
8
|
+
specs:
|
9
|
+
coderay (1.1.2)
|
10
|
+
diff-lcs (1.3)
|
11
|
+
docile (1.1.5)
|
12
|
+
json (2.1.0)
|
13
|
+
method_source (0.9.0)
|
14
|
+
pry (0.11.3)
|
15
|
+
coderay (~> 1.1.0)
|
16
|
+
method_source (~> 0.9.0)
|
17
|
+
rake (12.3.0)
|
18
|
+
rake-compiler (1.0.4)
|
19
|
+
rake
|
20
|
+
rspec (3.7.0)
|
21
|
+
rspec-core (~> 3.7.0)
|
22
|
+
rspec-expectations (~> 3.7.0)
|
23
|
+
rspec-mocks (~> 3.7.0)
|
24
|
+
rspec-core (3.7.0)
|
25
|
+
rspec-support (~> 3.7.0)
|
26
|
+
rspec-expectations (3.7.0)
|
27
|
+
diff-lcs (>= 1.2.0, < 2.0)
|
28
|
+
rspec-support (~> 3.7.0)
|
29
|
+
rspec-mocks (3.7.0)
|
30
|
+
diff-lcs (>= 1.2.0, < 2.0)
|
31
|
+
rspec-support (~> 3.7.0)
|
32
|
+
rspec-support (3.7.0)
|
33
|
+
simplecov (0.15.1)
|
34
|
+
docile (~> 1.1.0)
|
35
|
+
json (>= 1.8, < 3)
|
36
|
+
simplecov-html (~> 0.10.0)
|
37
|
+
simplecov-html (0.10.2)
|
38
|
+
yard (0.9.12)
|
39
|
+
|
40
|
+
PLATFORMS
|
41
|
+
ruby
|
42
|
+
|
43
|
+
DEPENDENCIES
|
44
|
+
bundler (~> 1.16)
|
45
|
+
pry (~> 0.11)
|
46
|
+
rake (~> 12.3)
|
47
|
+
rake-compiler (~> 1.0)
|
48
|
+
rspec (~> 3.7)
|
49
|
+
ruby-sfst!
|
50
|
+
simplecov (~> 0.15)
|
51
|
+
yard (~> 0.9)
|
52
|
+
|
53
|
+
BUNDLED WITH
|
54
|
+
1.16.1
|
data/README.md
CHANGED
@@ -3,7 +3,7 @@
|
|
3
3
|
`ruby-sfst` is a wrapper for the Stuttgart Finite State Transducer Tools
|
4
4
|
(SFST).
|
5
5
|
|
6
|
-
The wrapper is based on SFST 1.
|
6
|
+
The wrapper is based on SFST 1.4.7c. See
|
7
7
|
http://www.cis.uni-muenchen.de/~schmid/tools/SFST/ for details on how to obtain
|
8
8
|
SFST and how to write SFST transducers.
|
9
9
|
|
data/Rakefile
CHANGED
@@ -1,23 +1,14 @@
|
|
1
|
-
|
2
|
-
require
|
3
|
-
Bundler::GemHelper.install_tasks
|
1
|
+
require "bundler/gem_tasks"
|
2
|
+
require "rspec/core/rake_task"
|
4
3
|
|
5
|
-
|
6
|
-
|
7
|
-
|
8
|
-
task.files = ['README.md', 'lib/**/*.rb']
|
9
|
-
task.options = [
|
10
|
-
'--output-dir', 'doc/yard',
|
11
|
-
'--markup', 'markdown',
|
12
|
-
]
|
13
|
-
end
|
14
|
-
end
|
4
|
+
RSpec::Core::RakeTask.new(:spec)
|
5
|
+
|
6
|
+
require "rake/extensiontask"
|
15
7
|
|
16
|
-
|
8
|
+
task :build => :compile
|
17
9
|
|
18
|
-
Rake::
|
19
|
-
|
10
|
+
Rake::ExtensionTask.new("sfst") do |ext|
|
11
|
+
ext.lib_dir = "lib/sfst"
|
20
12
|
end
|
21
13
|
|
22
|
-
|
23
|
-
task :default => :test
|
14
|
+
task :default => [:clobber, :compile, :spec]
|
data/bin/console
ADDED
data/bin/setup
ADDED
@@ -0,0 +1,879 @@
|
|
1
|
+
|
2
|
+
/*******************************************************************/
|
3
|
+
/* */
|
4
|
+
/* FILE alphabet.C */
|
5
|
+
/* MODULE alphabet */
|
6
|
+
/* PROGRAM SFST */
|
7
|
+
/* AUTHOR Helmut Schmid, IMS, University of Stuttgart */
|
8
|
+
/* */
|
9
|
+
/* PURPOSE basic FST functions */
|
10
|
+
/* */
|
11
|
+
/*******************************************************************/
|
12
|
+
|
13
|
+
#include <climits>
|
14
|
+
#include <cstring>
|
15
|
+
|
16
|
+
#include "utf8.h"
|
17
|
+
#include "alphabet.h"
|
18
|
+
|
19
|
+
#include <map>
|
20
|
+
using std::map;
|
21
|
+
|
22
|
+
namespace SFST {
|
23
|
+
|
24
|
+
using std::vector;
|
25
|
+
using std::ostream;
|
26
|
+
|
27
|
+
const int BUFFER_SIZE=100000;
|
28
|
+
|
29
|
+
char EpsilonString[]="<>";
|
30
|
+
|
31
|
+
|
32
|
+
|
33
|
+
/*******************************************************************/
|
34
|
+
/* */
|
35
|
+
/* Alphabet::add */
|
36
|
+
/* */
|
37
|
+
/*******************************************************************/
|
38
|
+
|
39
|
+
void Alphabet::add( const char *symbol, Character c )
|
40
|
+
|
41
|
+
{
|
42
|
+
char *s = fst_strdup(symbol);
|
43
|
+
cm[c] = s;
|
44
|
+
sm[s] = c;
|
45
|
+
}
|
46
|
+
|
47
|
+
|
48
|
+
/*******************************************************************/
|
49
|
+
/* */
|
50
|
+
/* Alphabet::Alphabet */
|
51
|
+
/* */
|
52
|
+
/*******************************************************************/
|
53
|
+
|
54
|
+
Alphabet::Alphabet()
|
55
|
+
|
56
|
+
{
|
57
|
+
utf8 = false;
|
58
|
+
add(EpsilonString, Label::epsilon);
|
59
|
+
}
|
60
|
+
|
61
|
+
|
62
|
+
/*******************************************************************/
|
63
|
+
/* */
|
64
|
+
/* Alphabet::clear */
|
65
|
+
/* */
|
66
|
+
/*******************************************************************/
|
67
|
+
|
68
|
+
void Alphabet::clear()
|
69
|
+
|
70
|
+
{
|
71
|
+
char **s=new char*[cm.size()];
|
72
|
+
ls.clear();
|
73
|
+
sm.clear();
|
74
|
+
|
75
|
+
size_t i, n=0;
|
76
|
+
for( CharMap::iterator it=cm.begin(); it!=cm.end(); it++ )
|
77
|
+
s[n++] = it->second;
|
78
|
+
cm.clear();
|
79
|
+
|
80
|
+
for( i=0; i<n; i++ )
|
81
|
+
free(s[i]);
|
82
|
+
delete[] s;
|
83
|
+
}
|
84
|
+
|
85
|
+
|
86
|
+
/*******************************************************************/
|
87
|
+
/* */
|
88
|
+
/* Alphabet::print */
|
89
|
+
/* */
|
90
|
+
/*******************************************************************/
|
91
|
+
|
92
|
+
void Alphabet::print(void)
|
93
|
+
|
94
|
+
{
|
95
|
+
for( LabelSet::const_iterator it=begin(); it!=end(); it++ ) {
|
96
|
+
Label l = *it;
|
97
|
+
fprintf(stderr, "%s\n", write_label(l));
|
98
|
+
}
|
99
|
+
}
|
100
|
+
|
101
|
+
|
102
|
+
/*******************************************************************/
|
103
|
+
/* */
|
104
|
+
/* Alphabet::new_marker */
|
105
|
+
/* */
|
106
|
+
/*******************************************************************/
|
107
|
+
|
108
|
+
Character Alphabet::new_marker()
|
109
|
+
|
110
|
+
{
|
111
|
+
// find some unused character code
|
112
|
+
for(Character i=1; i!=0; i++)
|
113
|
+
if (cm.find(i) == cm.end()) {
|
114
|
+
// create a unique identifier string
|
115
|
+
char symbol[100];
|
116
|
+
sprintf(symbol,">%ld<",(long)i);
|
117
|
+
add(symbol, i);
|
118
|
+
return i;
|
119
|
+
}
|
120
|
+
|
121
|
+
throw "Error: too many symbols in transducer definition";
|
122
|
+
}
|
123
|
+
|
124
|
+
|
125
|
+
/*******************************************************************/
|
126
|
+
/* */
|
127
|
+
/* is_marker_symbol */
|
128
|
+
/* */
|
129
|
+
/*******************************************************************/
|
130
|
+
|
131
|
+
static bool is_marker_symbol( const char *s )
|
132
|
+
|
133
|
+
{
|
134
|
+
// recogize strings matching the expression ">[0-9]+<"
|
135
|
+
if (s != NULL && *s == '>') {
|
136
|
+
do { s++; } while (*s >= '0' && *s <= '9');
|
137
|
+
if (*s=='<' && *(s+1) == 0 && *(s-1) != '>')
|
138
|
+
return true;
|
139
|
+
}
|
140
|
+
return false;
|
141
|
+
}
|
142
|
+
|
143
|
+
|
144
|
+
/*******************************************************************/
|
145
|
+
/* */
|
146
|
+
/* Alphabet::delete_markers */
|
147
|
+
/* */
|
148
|
+
/*******************************************************************/
|
149
|
+
|
150
|
+
void Alphabet::delete_markers()
|
151
|
+
|
152
|
+
{
|
153
|
+
vector<char*> sym;
|
154
|
+
vector<Character> code;
|
155
|
+
vector<Label> label;
|
156
|
+
|
157
|
+
for( CharMap::const_iterator it=cm.begin(); it!=cm.end(); it++ ) {
|
158
|
+
Character c=it->first;
|
159
|
+
char *s=it->second;
|
160
|
+
if (!is_marker_symbol(s)) {
|
161
|
+
sym.push_back(fst_strdup(s));
|
162
|
+
code.push_back(c);
|
163
|
+
}
|
164
|
+
}
|
165
|
+
|
166
|
+
for( LabelSet::const_iterator it=begin(); it!=end(); it++ ) {
|
167
|
+
Label l=*it;
|
168
|
+
if (!is_marker_symbol(code2symbol(l.upper_char())) &&
|
169
|
+
!is_marker_symbol(code2symbol(l.lower_char())))
|
170
|
+
label.push_back(l);
|
171
|
+
}
|
172
|
+
|
173
|
+
clear();
|
174
|
+
|
175
|
+
for( size_t i=0; i<sym.size(); i++ ) {
|
176
|
+
add_symbol(sym[i], code[i]);
|
177
|
+
free(sym[i]);
|
178
|
+
}
|
179
|
+
for( size_t i=0; i<label.size(); i++ )
|
180
|
+
insert( label[i] );
|
181
|
+
}
|
182
|
+
|
183
|
+
|
184
|
+
/*******************************************************************/
|
185
|
+
/* */
|
186
|
+
/* Alphabet::add_symbol */
|
187
|
+
/* */
|
188
|
+
/*******************************************************************/
|
189
|
+
|
190
|
+
Character Alphabet::add_symbol(const char *symbol)
|
191
|
+
|
192
|
+
{
|
193
|
+
if (sm.find(symbol) != sm.end())
|
194
|
+
return sm[symbol];
|
195
|
+
|
196
|
+
// assign the symbol to some unused character
|
197
|
+
for(Character i=1; i!=0; i++)
|
198
|
+
if (cm.find(i) == cm.end()) {
|
199
|
+
add(symbol, i);
|
200
|
+
return i;
|
201
|
+
}
|
202
|
+
|
203
|
+
throw "Error: too many symbols in transducer definition";
|
204
|
+
}
|
205
|
+
|
206
|
+
|
207
|
+
/*******************************************************************/
|
208
|
+
/* */
|
209
|
+
/* Alphabet::add_symbol */
|
210
|
+
/* */
|
211
|
+
/*******************************************************************/
|
212
|
+
|
213
|
+
void Alphabet::add_symbol( const char *symbol, Character c )
|
214
|
+
|
215
|
+
{
|
216
|
+
// check whether the symbol was previously defined
|
217
|
+
int sc=symbol2code(symbol);
|
218
|
+
if (sc != EOF) {
|
219
|
+
if ((Character)sc == c)
|
220
|
+
return;
|
221
|
+
|
222
|
+
if (strlen(symbol) < 60) {
|
223
|
+
static char message[100];
|
224
|
+
sprintf(message, "Error: reinserting symbol '%s' in alphabet with incompatible character value %u %u", symbol, (unsigned)sc, (unsigned)c);
|
225
|
+
throw message;
|
226
|
+
}
|
227
|
+
else
|
228
|
+
throw "reinserting symbol in alphabet with incompatible character value";
|
229
|
+
}
|
230
|
+
|
231
|
+
// check whether the character is already in use
|
232
|
+
const char *s=code2symbol(c);
|
233
|
+
if (s == NULL)
|
234
|
+
add(symbol, c);
|
235
|
+
else {
|
236
|
+
if (strcmp(s, symbol) != 0) {
|
237
|
+
static char message[100];
|
238
|
+
if (strlen(symbol) < 70)
|
239
|
+
sprintf(message,"Error: defining symbol %s as character %d (previously defined as %s)", symbol, (unsigned)c, s);
|
240
|
+
else
|
241
|
+
sprintf(message,"Error: defining a (very long) symbol with previously used character");
|
242
|
+
throw message;
|
243
|
+
}
|
244
|
+
}
|
245
|
+
}
|
246
|
+
|
247
|
+
|
248
|
+
/*******************************************************************/
|
249
|
+
/* */
|
250
|
+
/* Alphabet::write_char */
|
251
|
+
/* */
|
252
|
+
/*******************************************************************/
|
253
|
+
|
254
|
+
void Alphabet::write_char( Character c, char *buffer, int *pos,
|
255
|
+
bool with_brackets) const
|
256
|
+
{
|
257
|
+
const char *s = code2symbol(c);
|
258
|
+
|
259
|
+
// quote colons
|
260
|
+
if (strcmp(s,":") == 0 || strcmp(s,"\\") == 0) {
|
261
|
+
buffer[(*pos)++] = '\\';
|
262
|
+
buffer[(*pos)++] = s[0];
|
263
|
+
}
|
264
|
+
else if (s) {
|
265
|
+
int i = 0;
|
266
|
+
int l=(int)strlen(s)-1;
|
267
|
+
if (!with_brackets && s[i] == '<' && s[l] == '>') { i++; l--; }
|
268
|
+
while (i <= l)
|
269
|
+
buffer[(*pos)++] = s[i++];
|
270
|
+
}
|
271
|
+
else {
|
272
|
+
unsigned int uc = c;
|
273
|
+
if (uc>=32 && uc<256)
|
274
|
+
buffer[(*pos)++] = (char)c;
|
275
|
+
else {
|
276
|
+
sprintf(buffer+(*pos),"\\%u", uc);
|
277
|
+
*pos += (int)strlen(buffer+(*pos));
|
278
|
+
}
|
279
|
+
}
|
280
|
+
buffer[*pos] = '\0';
|
281
|
+
}
|
282
|
+
|
283
|
+
|
284
|
+
/*******************************************************************/
|
285
|
+
/* */
|
286
|
+
/* Alphabet::write_char */
|
287
|
+
/* */
|
288
|
+
/*******************************************************************/
|
289
|
+
|
290
|
+
const char *Alphabet::write_char( Character c, bool with_brackets ) const
|
291
|
+
|
292
|
+
{
|
293
|
+
static char buffer[1000];
|
294
|
+
int n=0;
|
295
|
+
|
296
|
+
write_char( c, buffer, &n, with_brackets );
|
297
|
+
return buffer;
|
298
|
+
}
|
299
|
+
|
300
|
+
|
301
|
+
/*******************************************************************/
|
302
|
+
/* */
|
303
|
+
/* Alphabet::write_label */
|
304
|
+
/* */
|
305
|
+
/*******************************************************************/
|
306
|
+
|
307
|
+
void Alphabet::write_label( Label l, char *buffer, int *pos,
|
308
|
+
bool with_brackets ) const
|
309
|
+
{
|
310
|
+
Character lc=l.lower_char();
|
311
|
+
Character uc=l.upper_char();
|
312
|
+
write_char( lc, buffer, pos, with_brackets );
|
313
|
+
if (lc != uc) {
|
314
|
+
buffer[(*pos)++] = ':';
|
315
|
+
write_char( uc, buffer, pos, with_brackets );
|
316
|
+
}
|
317
|
+
}
|
318
|
+
|
319
|
+
|
320
|
+
/*******************************************************************/
|
321
|
+
/* */
|
322
|
+
/* Alphabet::write_label */
|
323
|
+
/* */
|
324
|
+
/*******************************************************************/
|
325
|
+
|
326
|
+
const char *Alphabet::write_label( Label l, bool with_brackets ) const
|
327
|
+
|
328
|
+
{
|
329
|
+
static char buffer[1000];
|
330
|
+
int n=0;
|
331
|
+
write_label( l, buffer, &n, with_brackets );
|
332
|
+
return buffer;
|
333
|
+
}
|
334
|
+
|
335
|
+
|
336
|
+
/*******************************************************************/
|
337
|
+
/* */
|
338
|
+
/* Alphabet::insert_symbols */
|
339
|
+
/* */
|
340
|
+
/*******************************************************************/
|
341
|
+
|
342
|
+
void Alphabet::insert_symbols( const Alphabet &a )
|
343
|
+
|
344
|
+
{
|
345
|
+
for( CharMap::const_iterator it=a.cm.begin(); it!=a.cm.end(); it++ )
|
346
|
+
add_symbol(it->second, it->first);
|
347
|
+
}
|
348
|
+
|
349
|
+
|
350
|
+
/*******************************************************************/
|
351
|
+
/* */
|
352
|
+
/* Alphabet::complement */
|
353
|
+
/* */
|
354
|
+
/*******************************************************************/
|
355
|
+
|
356
|
+
void Alphabet::complement( vector<Character> &sym )
|
357
|
+
|
358
|
+
{
|
359
|
+
vector<Character> result;
|
360
|
+
for( CharMap::const_iterator it=cm.begin(); it!=cm.end(); it++ ) {
|
361
|
+
Character c = it->first;
|
362
|
+
if (c != Label::epsilon) {
|
363
|
+
size_t i;
|
364
|
+
for( i=0; i<sym.size(); i++ )
|
365
|
+
if (sym[i] == c)
|
366
|
+
break;
|
367
|
+
if (i == sym.size())
|
368
|
+
result.push_back(c);
|
369
|
+
}
|
370
|
+
}
|
371
|
+
sym.swap(result);
|
372
|
+
}
|
373
|
+
|
374
|
+
|
375
|
+
/*******************************************************************/
|
376
|
+
/* */
|
377
|
+
/* Alphabet::copy */
|
378
|
+
/* */
|
379
|
+
/*******************************************************************/
|
380
|
+
|
381
|
+
void Alphabet::copy( const Alphabet &a, Level level )
|
382
|
+
|
383
|
+
{
|
384
|
+
utf8 = a.utf8;
|
385
|
+
sm.resize(a.sm.size());
|
386
|
+
cm.resize(a.sm.size());
|
387
|
+
insert_symbols( a );
|
388
|
+
for( LabelSet::const_iterator it=a.begin(); it!=a.end(); it++ ) {
|
389
|
+
Label l = *it;
|
390
|
+
if (level == lower)
|
391
|
+
ls.insert( Label(l.lower_char()) );
|
392
|
+
else if (level == upper)
|
393
|
+
ls.insert( Label(l.upper_char()) );
|
394
|
+
else
|
395
|
+
ls.insert( l );
|
396
|
+
}
|
397
|
+
}
|
398
|
+
|
399
|
+
|
400
|
+
/*******************************************************************/
|
401
|
+
/* */
|
402
|
+
/* Alphabet::compose */
|
403
|
+
/* */
|
404
|
+
/*******************************************************************/
|
405
|
+
|
406
|
+
void Alphabet::compose( const Alphabet &la, const Alphabet &ua )
|
407
|
+
|
408
|
+
{
|
409
|
+
// insert the symbols
|
410
|
+
insert_symbols(la);
|
411
|
+
insert_symbols(ua);
|
412
|
+
utf8 = la.utf8;
|
413
|
+
|
414
|
+
map<Character, set<Character> > cs;
|
415
|
+
|
416
|
+
// create a table for a quick lookup of the target characters
|
417
|
+
for( iterator it=ua.begin(); it!=ua.end(); it++ ) {
|
418
|
+
Character lc=it->lower_char();
|
419
|
+
if (lc == Label::epsilon)
|
420
|
+
insert(*it);
|
421
|
+
else
|
422
|
+
cs[lc].insert(it->upper_char());
|
423
|
+
}
|
424
|
+
|
425
|
+
for( iterator it=la.begin(); it!=la.end(); it++ ) {
|
426
|
+
Character uc=it->upper_char();
|
427
|
+
if (uc == Label::epsilon)
|
428
|
+
insert(*it);
|
429
|
+
else {
|
430
|
+
if (cs.find(uc) != cs.end()) {
|
431
|
+
set<Character> s=cs[uc];
|
432
|
+
Character lc=it->lower_char();
|
433
|
+
for( set<Character>::iterator it=s.begin(); it!=s.end(); it++)
|
434
|
+
insert(Label(lc, *it));
|
435
|
+
}
|
436
|
+
}
|
437
|
+
}
|
438
|
+
}
|
439
|
+
|
440
|
+
|
441
|
+
/*******************************************************************/
|
442
|
+
/* */
|
443
|
+
/* operator<<(Alphabet) */
|
444
|
+
/* */
|
445
|
+
/*******************************************************************/
|
446
|
+
|
447
|
+
ostream &operator<<( ostream &s, const Alphabet &a )
|
448
|
+
|
449
|
+
{
|
450
|
+
for( Alphabet::CharMap::const_iterator it=a.cm.begin(); it!=a.cm.end(); it++ )
|
451
|
+
s << it->first << " -> " << it->second << "\n";
|
452
|
+
for( Alphabet::iterator it=a.begin(); it!=a.end(); it++ )
|
453
|
+
s << a.write_label(*it) << " ";
|
454
|
+
s << "\n";
|
455
|
+
return s;
|
456
|
+
}
|
457
|
+
|
458
|
+
|
459
|
+
/*******************************************************************/
|
460
|
+
/* */
|
461
|
+
/* Alphabet::next_mcsym */
|
462
|
+
/* */
|
463
|
+
/* recognizes multi-character symbols which are enclosed with */
|
464
|
+
/* angle brackets <...>. If the argument flag insert is false, */
|
465
|
+
/* the multi-character symbol must already be in the lexicon in */
|
466
|
+
/* order to be recognized. */
|
467
|
+
/* */
|
468
|
+
/*******************************************************************/
|
469
|
+
|
470
|
+
int Alphabet::next_mcsym( char* &string, bool insert )
|
471
|
+
|
472
|
+
{
|
473
|
+
char *start=string;
|
474
|
+
|
475
|
+
if (*start == '<')
|
476
|
+
// symbol might start here
|
477
|
+
for( char *end=start+1; *end; end++ )
|
478
|
+
if (*end == '>') {
|
479
|
+
// matching pair of angle brackets found
|
480
|
+
// mark the end of the substring with \0
|
481
|
+
char lastc = *(++end);
|
482
|
+
*end = 0;
|
483
|
+
|
484
|
+
int c;
|
485
|
+
if (insert)
|
486
|
+
c = add_symbol( start );
|
487
|
+
else
|
488
|
+
c = symbol2code(start);
|
489
|
+
// restore the original string
|
490
|
+
*end = lastc;
|
491
|
+
|
492
|
+
if (c != EOF) {
|
493
|
+
// symbol found
|
494
|
+
// return its code
|
495
|
+
string = end;
|
496
|
+
return (Character)c;
|
497
|
+
}
|
498
|
+
else
|
499
|
+
// not a complex character
|
500
|
+
break;
|
501
|
+
}
|
502
|
+
return EOF;
|
503
|
+
}
|
504
|
+
|
505
|
+
|
506
|
+
/*******************************************************************/
|
507
|
+
/* */
|
508
|
+
/* Alphabet::next_code */
|
509
|
+
/* */
|
510
|
+
/*******************************************************************/
|
511
|
+
|
512
|
+
int Alphabet::next_code( char* &string, bool extended, bool insert )
|
513
|
+
|
514
|
+
{
|
515
|
+
if (*string == 0)
|
516
|
+
return EOF; // finished
|
517
|
+
|
518
|
+
int c = next_mcsym(string, insert);
|
519
|
+
if (c != EOF)
|
520
|
+
return c;
|
521
|
+
|
522
|
+
if (extended && *string == '\\')
|
523
|
+
string++; // remove quotation
|
524
|
+
|
525
|
+
if (utf8) {
|
526
|
+
unsigned int c = utf8toint( &string );
|
527
|
+
if (c == 0) {
|
528
|
+
fprintf(stderr, "Error in UTF-8 encoding at: <%s>\n", string);
|
529
|
+
return EOF; // error encountered in utf8 character
|
530
|
+
}
|
531
|
+
return (int)add_symbol(int2utf8(c));
|
532
|
+
}
|
533
|
+
else {
|
534
|
+
char buffer[2];
|
535
|
+
buffer[0] = *string;
|
536
|
+
buffer[1] = 0;
|
537
|
+
string++;
|
538
|
+
return (int)add_symbol(buffer);
|
539
|
+
}
|
540
|
+
}
|
541
|
+
|
542
|
+
|
543
|
+
/*******************************************************************/
|
544
|
+
/* */
|
545
|
+
/* Alphabet::next_label */
|
546
|
+
/* */
|
547
|
+
/*******************************************************************/
|
548
|
+
|
549
|
+
Label Alphabet::next_label( char* &string, bool extended )
|
550
|
+
|
551
|
+
{
|
552
|
+
// read first character
|
553
|
+
int c = next_code( string, extended );
|
554
|
+
if (c == EOF)
|
555
|
+
return Label(); // end of string reached
|
556
|
+
|
557
|
+
Character lc=(Character)c;
|
558
|
+
if (!extended || *string != ':') { // single character?
|
559
|
+
if (lc == Label::epsilon)
|
560
|
+
return next_label(string, extended); // ignore epsilon
|
561
|
+
return Label(lc);
|
562
|
+
}
|
563
|
+
|
564
|
+
// read second character
|
565
|
+
string++; // jump over ':'
|
566
|
+
c = next_code( string, extended );
|
567
|
+
if (c == EOF) {
|
568
|
+
static char buffer[1000];
|
569
|
+
sprintf(buffer,"Error: incomplete symbol in input file: %s", string);
|
570
|
+
throw buffer;
|
571
|
+
}
|
572
|
+
|
573
|
+
Label l(lc, (Character)c);
|
574
|
+
if (l.is_epsilon())
|
575
|
+
return next_label(string, extended); // ignore epsilon transitions
|
576
|
+
return l;
|
577
|
+
}
|
578
|
+
|
579
|
+
|
580
|
+
/*******************************************************************/
|
581
|
+
/* */
|
582
|
+
/* Alphabet::string2symseq */
|
583
|
+
/* */
|
584
|
+
/*******************************************************************/
|
585
|
+
|
586
|
+
void Alphabet::string2symseq( char *s, vector<Character> &ch )
|
587
|
+
|
588
|
+
{
|
589
|
+
int c;
|
590
|
+
while ((c = next_code(s, false, false)) != EOF)
|
591
|
+
ch.push_back((Character)c);
|
592
|
+
}
|
593
|
+
|
594
|
+
|
595
|
+
/*******************************************************************/
|
596
|
+
/* */
|
597
|
+
/* Alphabet::string2labelseq */
|
598
|
+
/* */
|
599
|
+
/*******************************************************************/
|
600
|
+
|
601
|
+
void Alphabet::string2labelseq( char *s, vector<Label> &labels )
|
602
|
+
|
603
|
+
{
|
604
|
+
Label l;
|
605
|
+
while ((l = next_label(s)) != Label::epsilon)
|
606
|
+
labels.push_back(l);
|
607
|
+
}
|
608
|
+
|
609
|
+
|
610
|
+
/*******************************************************************/
|
611
|
+
/* */
|
612
|
+
/* Alphabet::store */
|
613
|
+
/* */
|
614
|
+
/*******************************************************************/
|
615
|
+
|
616
|
+
void Alphabet::store( FILE *file ) const
|
617
|
+
|
618
|
+
{
|
619
|
+
char c=(utf8)? (char)1: (char)0;
|
620
|
+
fputc(c, file);
|
621
|
+
|
622
|
+
// write the symbol mapping
|
623
|
+
Character n=(Character)cm.size();
|
624
|
+
fwrite(&n, sizeof(n), 1, file);
|
625
|
+
for( CharMap::const_iterator it=cm.begin(); it!=cm.end(); it++ ) {
|
626
|
+
Character c=it->first;
|
627
|
+
char *s=it->second;
|
628
|
+
fwrite(&c, sizeof(c), 1, file);
|
629
|
+
fwrite(s, sizeof(char), strlen(s)+1, file);
|
630
|
+
}
|
631
|
+
|
632
|
+
// write the character pairs
|
633
|
+
n = (Character)size();
|
634
|
+
fwrite(&n, sizeof(n), 1, file);
|
635
|
+
for( LabelSet::const_iterator p=ls.begin(); p!=ls.end(); p++ ) {
|
636
|
+
Character c=p->lower_char();
|
637
|
+
fwrite(&c, sizeof(c), 1, file);
|
638
|
+
c = p->upper_char();
|
639
|
+
fwrite(&c, sizeof(c), 1, file);
|
640
|
+
}
|
641
|
+
|
642
|
+
if (ferror(file))
|
643
|
+
throw "Error encountered while writing alphabet to file\n";
|
644
|
+
}
|
645
|
+
|
646
|
+
|
647
|
+
/*******************************************************************/
|
648
|
+
/* */
|
649
|
+
/* Alphabet::read */
|
650
|
+
/* */
|
651
|
+
/*******************************************************************/
|
652
|
+
|
653
|
+
void Alphabet::read( FILE *file )
|
654
|
+
|
655
|
+
{
|
656
|
+
utf8 = (fgetc(file) != 0);
|
657
|
+
|
658
|
+
// read the symbol mapping
|
659
|
+
Character n=0;
|
660
|
+
read_num(&n, sizeof(n), file);
|
661
|
+
for( unsigned i=0; i<n; i++) {
|
662
|
+
char buffer[BUFFER_SIZE];
|
663
|
+
Character c;
|
664
|
+
read_num(&c, sizeof(c), file);
|
665
|
+
if (!read_string(buffer, BUFFER_SIZE, file) ||
|
666
|
+
feof(file) || ferror(file))
|
667
|
+
throw "Error1 occurred while reading alphabet!\n";
|
668
|
+
add_symbol(buffer, c);
|
669
|
+
}
|
670
|
+
|
671
|
+
// read the character pairs
|
672
|
+
read_num(&n, sizeof(n), file);
|
673
|
+
if (ferror(file))
|
674
|
+
throw "Error2 occurred while reading alphabet!\n";
|
675
|
+
for( unsigned i=0; i<n; i++) {
|
676
|
+
Character lc, uc;
|
677
|
+
read_num(&lc, sizeof(lc), file);
|
678
|
+
read_num(&uc, sizeof(uc), file);
|
679
|
+
insert(Label(lc, uc));
|
680
|
+
}
|
681
|
+
if (ferror(file))
|
682
|
+
throw "Error3 occurred while reading alphabet!\n";
|
683
|
+
}
|
684
|
+
|
685
|
+
|
686
|
+
/*******************************************************************/
|
687
|
+
/* */
|
688
|
+
/* Alphabet::compute_score */
|
689
|
+
/* */
|
690
|
+
/*******************************************************************/
|
691
|
+
|
692
|
+
int Alphabet::compute_score( Analysis &ana )
|
693
|
+
|
694
|
+
{
|
695
|
+
// check whether the morpheme boundaries are explicitly marked
|
696
|
+
// with <X> tags
|
697
|
+
int score=0;
|
698
|
+
for( size_t i=0; i<ana.size(); i++ ) {
|
699
|
+
|
700
|
+
// get next symbol
|
701
|
+
const char *sym=write_char(ana[i].lower_char());
|
702
|
+
|
703
|
+
if (strcmp(sym,"<X>") == 0)
|
704
|
+
score--;
|
705
|
+
}
|
706
|
+
if (score < 0)
|
707
|
+
return score;
|
708
|
+
|
709
|
+
// No explicit morpheme boundary markers have been found.
|
710
|
+
// Count the number of part-of-speech and PREF tags.
|
711
|
+
for( size_t i=0; i<ana.size(); i++ ) {
|
712
|
+
|
713
|
+
// get next symbol
|
714
|
+
const char *sym=write_char(ana[i].lower_char());
|
715
|
+
|
716
|
+
// Is it not a multi-character symbol
|
717
|
+
if (sym[0] != '<' || sym[1] == 0)
|
718
|
+
continue;
|
719
|
+
|
720
|
+
// Is it a POS tag starting with "+" like <+NN>?
|
721
|
+
if (sym[1] == '+') {
|
722
|
+
const char *t=sym+2;
|
723
|
+
for( ; *t >= 'A' && *t <= 'Z'; t++) ;
|
724
|
+
if (t > sym+2 && *t == '>')
|
725
|
+
return score;
|
726
|
+
}
|
727
|
+
|
728
|
+
// Is it a potential POS tag (i.e. all uppercase)?
|
729
|
+
const char *t = sym+1;
|
730
|
+
for( ; *t >= 'A' && *t <= 'Z'; t++) ;
|
731
|
+
if (t == sym+1 || *t != '>')
|
732
|
+
continue;
|
733
|
+
|
734
|
+
// uppercase symbol found
|
735
|
+
if (strcmp(sym,"<SUFF>") == 0 ||
|
736
|
+
strcmp(sym,"<OLDORTH>") == 0 ||
|
737
|
+
strcmp(sym,"<NEWORTH>") == 0)
|
738
|
+
continue; // not what we are looking for
|
739
|
+
|
740
|
+
// disprefer nouns with prefixes
|
741
|
+
if (strcmp(sym,"<PREF>") == 0)
|
742
|
+
score-=2;
|
743
|
+
|
744
|
+
if (strcmp(sym,"<V>") == 0 || strcmp(sym,"<ADJ>") == 0) {
|
745
|
+
bool is_verb=(strcmp(sym,"<V>")==0);
|
746
|
+
// get the next non-empty symbol
|
747
|
+
Character c=Label::epsilon;
|
748
|
+
size_t k;
|
749
|
+
for( k=i+1; k<ana.size(); k++ )
|
750
|
+
if ((c = ana[k].lower_char()) != Label::epsilon)
|
751
|
+
break;
|
752
|
+
// Is it a participle
|
753
|
+
if (c != Label::epsilon) {
|
754
|
+
sym = write_char(c);
|
755
|
+
if (strcmp(sym,"<OLDORTH>") == 0 ||
|
756
|
+
strcmp(sym,"<NEWORTH>") == 0 ||
|
757
|
+
strcmp(sym,"<SUFF>") == 0) {
|
758
|
+
for( k++; k<ana.size(); k++ )
|
759
|
+
if ((c = ana[k].lower_char()) != Label::epsilon)
|
760
|
+
break;
|
761
|
+
if (c != Label::epsilon)
|
762
|
+
sym = write_char(c);
|
763
|
+
}
|
764
|
+
if (is_verb &&
|
765
|
+
(strcmp(sym,"<PPres>") == 0 || strcmp(sym,"<PPast>") == 0))
|
766
|
+
continue; // don't consider participles as complex
|
767
|
+
if (!is_verb &&
|
768
|
+
(strcmp(sym,"<Sup>") == 0 || strcmp(sym,"<Comp>") == 0))
|
769
|
+
continue;
|
770
|
+
}
|
771
|
+
}
|
772
|
+
score--;
|
773
|
+
}
|
774
|
+
return score;
|
775
|
+
}
|
776
|
+
|
777
|
+
|
778
|
+
|
779
|
+
/*******************************************************************/
|
780
|
+
/* */
|
781
|
+
/* Alphabet::disambiguate */
|
782
|
+
/* */
|
783
|
+
/*******************************************************************/
|
784
|
+
|
785
|
+
void Alphabet::disambiguate( vector<Analysis> &analyses )
|
786
|
+
|
787
|
+
{
|
788
|
+
// compute the scores
|
789
|
+
int bestscore=INT_MIN;
|
790
|
+
vector<int> score;
|
791
|
+
|
792
|
+
for( size_t i=0; i<analyses.size(); i++ ) {
|
793
|
+
score.push_back(compute_score(analyses[i]));
|
794
|
+
if (bestscore < score[i])
|
795
|
+
bestscore = score[i];
|
796
|
+
}
|
797
|
+
|
798
|
+
// delete suboptimal analyses
|
799
|
+
size_t k=0;
|
800
|
+
for( size_t i=0; i<analyses.size(); i++ )
|
801
|
+
if (score[i] == bestscore)
|
802
|
+
analyses[k++] = analyses[i];
|
803
|
+
analyses.resize(k);
|
804
|
+
}
|
805
|
+
|
806
|
+
|
807
|
+
|
808
|
+
/*******************************************************************/
|
809
|
+
/* */
|
810
|
+
/* Alphabet::print_analysis */
|
811
|
+
/* */
|
812
|
+
/*******************************************************************/
|
813
|
+
|
814
|
+
char *Alphabet::print_analysis( Analysis &ana, bool both_layers )
|
815
|
+
|
816
|
+
{
|
817
|
+
vector<char> ch;
|
818
|
+
|
819
|
+
// for each transition
|
820
|
+
for( size_t i=0; i<ana.size(); i++ ) {
|
821
|
+
|
822
|
+
// get the transition label
|
823
|
+
Label l=ana[i];
|
824
|
+
const char *s;
|
825
|
+
|
826
|
+
// either print the analysis symbol or the whole label
|
827
|
+
if (both_layers)
|
828
|
+
s = write_label(l);
|
829
|
+
else if (l.lower_char() != Label::epsilon)
|
830
|
+
s = write_char(l.lower_char());
|
831
|
+
else
|
832
|
+
continue;
|
833
|
+
|
834
|
+
// copy the characters to the character array
|
835
|
+
while (*s)
|
836
|
+
ch.push_back(*(s++));
|
837
|
+
}
|
838
|
+
ch.push_back(0); // terminate the string
|
839
|
+
|
840
|
+
static char *result=NULL;
|
841
|
+
if (result != NULL)
|
842
|
+
delete[] result;
|
843
|
+
result = new char[ch.size()];
|
844
|
+
for( size_t i=0; i<ch.size(); i++ )
|
845
|
+
result[i] = ch[i];
|
846
|
+
|
847
|
+
return result;
|
848
|
+
}
|
849
|
+
|
850
|
+
|
851
|
+
/*******************************************************************/
|
852
|
+
/* */
|
853
|
+
/* Alphabet::operator== */
|
854
|
+
/* */
|
855
|
+
/*******************************************************************/
|
856
|
+
|
857
|
+
bool Alphabet::operator==(const Alphabet &alpha) const
|
858
|
+
|
859
|
+
{
|
860
|
+
for ( SymbolMap::const_iterator it = this->sm.begin(); it != this->sm.end(); it++ )
|
861
|
+
{
|
862
|
+
SymbolMap::const_iterator alpha_it = alpha.sm.find(it->first);
|
863
|
+
if ( alpha_it == alpha.sm.end() )
|
864
|
+
return false;
|
865
|
+
if ( alpha_it->second == it->second )
|
866
|
+
return false;
|
867
|
+
}
|
868
|
+
for ( SymbolMap::const_iterator alpha_it = alpha.sm.begin(); alpha_it != alpha.sm.end(); alpha_it++ )
|
869
|
+
{
|
870
|
+
SymbolMap::const_iterator it = this->sm.find(alpha_it->first);
|
871
|
+
if ( it == this->sm.end() )
|
872
|
+
return false;
|
873
|
+
if ( it->second == alpha_it->second )
|
874
|
+
return false;
|
875
|
+
}
|
876
|
+
return true;
|
877
|
+
}
|
878
|
+
|
879
|
+
}
|