summaryrefslogtreecommitdiffstats
path: root/src/storage/ngram.cpp
blob: 1c99948da59f1cd1d4dc52e11e171a2431277ac7 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
/* 
 *  libpinyin
 *  Library to deal with pinyin.
 *  
 *  Copyright (C) 2006-2007 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 <errno.h>
#include <glib.h>
#include <glib/gstdio.h>
#include "memory_chunk.h"
#include "novel_types.h"
#include "ngram.h"

using namespace pinyin;

struct SingleGramItem{
    phrase_token_t m_token;
    guint32 m_freq;
};

SingleGram::SingleGram(){
    m_chunk.set_size(sizeof(guint32));
    memset(m_chunk.begin(), 0, sizeof(guint32));
}

SingleGram::SingleGram(void * buffer, size_t length, bool copy){
    if (copy)
        m_chunk.set_content(0, buffer, length);
    else
        m_chunk.set_chunk(buffer, length, NULL);
}

bool SingleGram::get_total_freq(guint32 & total) const{
    char * buf_begin = (char *)m_chunk.begin();
    total = *((guint32 *)buf_begin);
    return true;
}

bool SingleGram::set_total_freq(guint32 total){
    char * buf_begin = (char *)m_chunk.begin();
    *((guint32 *)buf_begin) = total;
    return true;
}

guint32 SingleGram::get_length(){
    /* get the number of items. */
    const SingleGramItem * begin = (const SingleGramItem *)
        ((const char *)(m_chunk.begin()) + sizeof(guint32));
    const SingleGramItem * end = (const SingleGramItem *) m_chunk.end();

    const guint32 length = end - begin;

    if (0 == length) {
        /* no items here, total freq should be zero. */
        guint32 total_freq = 0;
        assert(get_total_freq(total_freq));
        assert(0 == total_freq);
    }

    return length;
}

guint32 SingleGram::mask_out(phrase_token_t mask, phrase_token_t value){
    guint32 removed_items = 0;

    guint32 total_freq = 0;
    assert(get_total_freq(total_freq));

    const SingleGramItem * begin = (const SingleGramItem *)
        ((const char *)(m_chunk.begin()) + sizeof(guint32));
    const SingleGramItem * end = (const SingleGramItem *) m_chunk.end();

    for (const SingleGramItem * cur = begin; cur != end; ++cur) {
        if ((cur->m_token & mask) != value)
            continue;

        total_freq -= cur->m_freq;
        size_t offset = sizeof(guint32) +
            sizeof(SingleGramItem) * (cur - begin);
        m_chunk.remove_content(offset, sizeof(SingleGramItem));

        /* update chunk end. */
        end = (const SingleGramItem *) m_chunk.end();
        ++removed_items;
        --cur;
    }

    assert(set_total_freq(total_freq));
    return removed_items;
}

bool SingleGram::prune(){
    assert(false);
#if 0
    SingleGramItem * begin = (SingleGramItem *)
	((const char *)(m_chunk.begin()) + sizeof(guint32));
    SingleGramItem * end = (SingleGramItem *)m_chunk.end();
    
    size_t nitem = 0;
    for ( SingleGramItem * cur = begin; cur != end; ++cur){
	cur->m_freq--;
	nitem++;
	if ( cur->m_freq == 0 ){
	    size_t offset = sizeof(guint32) + (cur - begin)
		* sizeof(SingleGramItem) ;
	    m_chunk.remove_content(offset, sizeof(SingleGramItem));
	}
    }
    guint32 total_freq;
    assert(get_total_freq(total_freq));
    assert(set_total_freq(total_freq - nitem));
#endif
	return true;
}

static bool token_less_than(const SingleGramItem & lhs,const SingleGramItem & rhs){
    return lhs.m_token < rhs.m_token;
}

bool SingleGram::retrieve_all(/* out */ BigramPhraseWithCountArray array)
    const {
    const SingleGramItem * begin = (const SingleGramItem *)
        ((const char *)(m_chunk.begin()) + sizeof(guint32));
    const SingleGramItem * end = (const SingleGramItem *) m_chunk.end();

    guint32 total_freq;
    BigramPhraseItemWithCount bigram_item_with_count;
    assert(get_total_freq(total_freq));

    for ( const SingleGramItem * cur_item = begin; cur_item != end; ++cur_item){
        bigram_item_with_count.m_token = cur_item->m_token;
        bigram_item_with_count.m_count = cur_item->m_freq;
        bigram_item_with_count.m_freq = cur_item->m_freq / (gfloat)total_freq;
        g_array_append_val(array, bigram_item_with_count);
    }

    return true;
}

bool SingleGram::search(/* in */ PhraseIndexRange * range,
			/* out */ BigramPhraseArray array) const {
    const SingleGramItem * begin = (const SingleGramItem *)
	((const char *)(m_chunk.begin()) + sizeof(guint32));
    const SingleGramItem * end = (const SingleGramItem *)m_chunk.end();

    SingleGramItem compare_item;
    compare_item.m_token = range->m_range_begin;
    const SingleGramItem * cur_item = std_lite::lower_bound(begin, end, compare_item, token_less_than);

    guint32 total_freq;
    BigramPhraseItem bigram_item;
    assert(get_total_freq(total_freq));

    for ( ; cur_item != end; ++cur_item){
	if ( cur_item->m_token >= range->m_range_end )
	    break;
	bigram_item.m_token = cur_item->m_token;
	bigram_item.m_freq = cur_item->m_freq / (gfloat)total_freq;
	g_array_append_val(array, bigram_item);
    }

    return true;
}

bool SingleGram::insert_freq( /* in */ phrase_token_t token,
                              /* in */ guint32 freq){
    SingleGramItem * begin = (SingleGramItem *)
        ((const char *)(m_chunk.begin()) + sizeof(guint32));
    SingleGramItem * end = (SingleGramItem *) m_chunk.end();
    SingleGramItem compare_item;
    compare_item.m_token = token;
    SingleGramItem * cur_item = std_lite::lower_bound(begin, end, compare_item, token_less_than);

    SingleGramItem insert_item;
    insert_item.m_token = token;
    insert_item.m_freq = freq;
    for ( ; cur_item != end; ++cur_item ){
        if ( cur_item->m_token > token ){
            size_t offset = sizeof(guint32) +
                sizeof(SingleGramItem) * (cur_item - begin);
            m_chunk.insert_content(offset, &insert_item,
                                   sizeof(SingleGramItem));
            return true;
        }
        if ( cur_item->m_token == token ){
            return false;
        }
    }
    m_chunk.insert_content(m_chunk.size(), &insert_item,
                           sizeof(SingleGramItem));
    return true;
}

bool SingleGram::remove_freq( /* in */ phrase_token_t token,
                              /* out */ guint32 & freq){
    freq = 0;
    const SingleGramItem * begin = (const SingleGramItem *)
        ((const char *)(m_chunk.begin()) + sizeof(guint32));
    const SingleGramItem * end = (const SingleGramItem *)m_chunk.end();
    SingleGramItem compare_item;
    compare_item.m_token = token;
    const SingleGramItem * cur_item = std_lite::lower_bound(begin, end, compare_item, token_less_than);

    for ( ; cur_item != end; ++cur_item ){
        if ( cur_item->m_token > token )
            return false;
        if ( cur_item->m_token == token ){
            freq = cur_item -> m_freq;
            size_t offset = sizeof(guint32) +
                sizeof(SingleGramItem) * (cur_item - begin);
            m_chunk.remove_content(offset, sizeof(SingleGramItem));
            return true;
        }
    }
    return false;
}

bool SingleGram::get_freq(/* in */ phrase_token_t token,
                          /* out */ guint32 & freq) const {
    freq = 0;
    const SingleGramItem * begin = (const SingleGramItem *)
	((const char *)(m_chunk.begin()) + sizeof(guint32));
    const SingleGramItem * end = (const SingleGramItem *)m_chunk.end();
    SingleGramItem compare_item;
    compare_item.m_token = token;
    const SingleGramItem * cur_item = std_lite::lower_bound(begin, end, compare_item, token_less_than);
    
    for ( ; cur_item != end; ++cur_item){
	if ( cur_item->m_token > token )
	    return false;
	if ( cur_item->m_token == token ){
	    freq = cur_item -> m_freq;
	    return true;
	}
    }
    return false;
}

bool SingleGram::set_freq( /* in */ phrase_token_t token,
			   /* in */ guint32 freq){
    SingleGramItem * begin = (SingleGramItem *)
	((const char *)(m_chunk.begin()) + sizeof(guint32));
    SingleGramItem * end = (SingleGramItem *)m_chunk.end();
    SingleGramItem compare_item;
    compare_item.m_token = token;
    SingleGramItem * cur_item = std_lite::lower_bound(begin, end, compare_item, token_less_than);
    
    for ( ;cur_item != end; ++cur_item){
	if ( cur_item->m_token > token ){
	    return false;
	}
	if ( cur_item->m_token == token ){
	    cur_item -> m_freq = freq;
	    return true;
	}
    }
    return false;
}


namespace pinyin{

/* merge origin system info and delta user info */
bool merge_single_gram(SingleGram * merged, const SingleGram * system,
                       const SingleGram * user){
    if (NULL == system && NULL == user)
        return false;

    MemoryChunk & merged_chunk = merged->m_chunk;

    if (NULL == system) {
        merged_chunk.set_chunk(user->m_chunk.begin(),
                               user->m_chunk.size(), NULL);
        return true;
    }

    if (NULL == user) {
        merged_chunk.set_chunk(system->m_chunk.begin(),
                               system->m_chunk.size(), NULL);
        return true;
    }

    /* clear merged. */
    merged_chunk.set_size(sizeof(guint32));

    /* merge the origin info and delta info */
    guint32 system_total, user_total;
    assert(system->get_total_freq(system_total));
    assert(user->get_total_freq(user_total));
    const guint32 merged_total = system_total + user_total;
    merged_chunk.set_content(0, &merged_total, sizeof(guint32));

    const SingleGramItem * cur_system = (const SingleGramItem *)
        (((const char *)(system->m_chunk.begin())) + sizeof(guint32));
    const SingleGramItem * system_end = (const SingleGramItem *)
        system->m_chunk.end();

    const SingleGramItem * cur_user = (const SingleGramItem *)
        (((const char *)(user->m_chunk.begin())) + sizeof(guint32));
    const SingleGramItem * user_end = (const SingleGramItem *)
        user->m_chunk.end();

    while (cur_system < system_end && cur_user < user_end) {

        if (cur_system->m_token < cur_user->m_token) {
            /* do append operation here */
            merged_chunk.append_content(cur_system, sizeof(SingleGramItem));
            cur_system++;
        } else if (cur_system->m_token > cur_user->m_token) {
            /* do append operation here */
            merged_chunk.append_content(cur_user, sizeof(SingleGramItem));
            cur_user++;
        } else {
            assert(cur_system->m_token == cur_user->m_token);

            SingleGramItem merged_item;
            merged_item.m_token = cur_system->m_token;
            merged_item.m_freq = cur_system->m_freq + cur_user->m_freq;

            merged_chunk.append_content(&merged_item, sizeof(SingleGramItem));
            cur_system++; cur_user++;
        }
    }

    /* add remained items. */
    while (cur_system < system_end) {
        merged_chunk.append_content(cur_system, sizeof(SingleGramItem));
        cur_system++;
    }

    while (cur_user < user_end) {
        merged_chunk.append_content(cur_user, sizeof(SingleGramItem));
        cur_user++;
    }

    return true;
}

};