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
|
/*
* libpinyin
* Library to deal with pinyin.
*
* Copyright (C) 2010 Peng Wu
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
#include <stdio.h>
#include <string.h>
#include <locale.h>
#include <glib.h>
#include "pinyin_internal.h"
/* graph shortest path sentence segment. */
/* Note:
* Currently libpinyin only supports ucs4 characters, as this is a
* pre-processor tool for raw corpus, it will skip all sentences
* which contains non-ucs4 characters.
*/
struct SegmentStep{
phrase_token_t m_handle;
ucs4_t * m_phrase;
size_t m_phrase_len;
//use formula W = number of words. Zero handle means one word.
guint m_nword;
//backtrace information, -1 one step backward.
gint m_backward_nstep;
public:
SegmentStep(){
m_handle = 0;
m_phrase = NULL;
m_phrase_len = 0;
m_nword = UINT_MAX;
m_backward_nstep = -0;
}
};
bool backtrace(GArray * steps, glong phrase_len, GArray * strings);
//Note: do not free phrase, as it is used by strings (array of segment).
bool segment(PhraseLargeTable * phrases, //Lookup Phrase
ucs4_t * phrase,
glong phrase_len,
GArray * strings /* Array of Segment *. */){
/* Prepare for shortest path segment dynamic programming. */
GArray * steps = g_array_new(TRUE, TRUE, sizeof(SegmentStep));
SegmentStep step;
for ( glong i = 0; i < phrase_len + 1; ++i ){
g_array_append_val(steps, step);
}
SegmentStep * first_step = &g_array_index(steps, SegmentStep, 0);
first_step->m_nword = 0;
for ( glong i = 0; i < phrase_len + 1; ++i ) {
SegmentStep * step_begin = &g_array_index(steps, SegmentStep, i);
size_t nword = step_begin->m_nword;
for ( glong k = i + 1; k < phrase_len + 1; ++k ) {
size_t len = k - i;
ucs4_t * cur_phrase = phrase + i;
phrase_token_t token = 0;
int result = phrases->search(len, cur_phrase, token);
if ( !(result & SEARCH_OK) ){
token = 0;
if ( 1 != len )
continue;
}
++nword;
SegmentStep * step_end = &g_array_index(steps, SegmentStep, k);
if ( nword < step_end->m_nword ) {
step_end->m_handle = token;
step_end->m_phrase = cur_phrase;
step_end->m_phrase_len = len;
step_end->m_nword = nword;
step_end->m_backward_nstep = i - k;
}
if ( !(result & SEARCH_CONTINUED) )
break;
}
}
return backtrace(steps, phrase_len, strings);
}
bool backtrace(GArray * steps, glong phrase_len, GArray * strings){
//backtracing to get the result.
size_t cur_step = phrase_len;
g_array_set_size(strings, 0);
while ( cur_step ){
SegmentStep * step = &g_array_index(steps, SegmentStep, cur_step);
g_array_append_val(strings, *step);
cur_step = cur_step + step->m_backward_nstep;
//intended to avoid leaking internal informations
step->m_nword = 0; step->m_backward_nstep = 0;
}
//reverse the strings
for ( size_t i = 0; i < strings->len / 2; ++i ) {
SegmentStep * head, * tail;
head = &g_array_index(strings, SegmentStep, i);
tail = &g_array_index(strings, SegmentStep, strings->len - 1 - i );
SegmentStep tmp;
tmp = *head;
*head = *tail;
*tail = tmp;
}
g_array_free(steps, TRUE);
return true;
}
void print_help(){
printf("Usage: spseg [--generate-extra-enter]\n");
}
int main(int argc, char * argv[]){
int i = 1;
bool gen_extra_enter = false;
setlocale(LC_ALL, "");
//deal with options.
while ( i < argc ){
if ( strcmp ("--help", argv[i]) == 0) {
print_help();
exit(0);
} else if (strcmp("--generate-extra-enter", argv[i]) == 0) {
gen_extra_enter = true;
} else {
print_help();
exit(EINVAL);
}
++i;
}
//init phrase table
PhraseLargeTable phrase_table;
MemoryChunk * chunk = new MemoryChunk;
chunk->load("phrase_index.bin");
phrase_table.load(chunk);
char * linebuf = NULL;
size_t size = 0;
ssize_t read;
while( (read = getline(&linebuf, &size, stdin)) != -1 ){
if ( '\n' == linebuf[strlen(linebuf) - 1] ) {
linebuf[strlen(linebuf) - 1] = '\0';
}
//check non-ucs4 characters
const glong num_of_chars = g_utf8_strlen(linebuf, -1);
glong len = 0;
ucs4_t * sentence = g_utf8_to_ucs4(linebuf, -1, NULL, &len, NULL);
if ( len != num_of_chars ) {
fprintf(stderr, "non-ucs4 characters encountered:%s.\n", linebuf);
printf("\n");
continue;
}
//do segment stuff
GArray * strings = g_array_new(TRUE, TRUE, sizeof(SegmentStep));
segment(&phrase_table, sentence, len, strings);
//print out the split phrase
for ( glong i = 0; i < strings->len; ++i ) {
SegmentStep * step = &g_array_index(strings, SegmentStep, i);
char * string = g_ucs4_to_utf8( step->m_phrase, step->m_phrase_len, NULL, NULL, NULL);
printf("%s\n", string);
g_free(string);
}
/* print extra enter */
if ( gen_extra_enter )
printf("\n");
g_array_free(strings, TRUE);
g_free(sentence);
}
/* print enter at file tail */
printf("\n");
return 0;
}
|
