summaryrefslogtreecommitdiff
path: root/dict.c
blob: eaa40063cac82eecf535d8161615e4b86a5018c1 (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
/*
 * dict.c: dictionary of reusable strings, just used to avoid allocation
 *         and freeing operations.
 *
 * Copyright (C) 2003 Daniel Veillard.
 *
 * Permission to use, copy, modify, and distribute this software for any
 * purpose with or without fee is hereby granted, provided that the above
 * copyright notice and this permission notice appear in all copies.
 *
 * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED
 * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
 * MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE AUTHORS AND
 * CONTRIBUTORS ACCEPT NO RESPONSIBILITY IN ANY CONCEIVABLE MANNER.
 *
 * Author: daniel@veillard.com
 */

#define IN_LIBXML
#include "libxml.h"

#include <string.h>
#include <libxml/tree.h>
#include <libxml/dict.h>
#include <libxml/xmlmemory.h>
#include <libxml/xmlerror.h>
#include <libxml/globals.h>

#define MAX_HASH_LEN 4
#define MIN_DICT_SIZE 128
#define MAX_DICT_HASH 8 * 2048

/* #define ALLOW_REMOVAL */
/* #define DEBUG_GROW */

/*
 * An entry in the dictionnary
 */
typedef struct _xmlDictEntry xmlDictEntry;
typedef xmlDictEntry *xmlDictEntryPtr;
struct _xmlDictEntry {
    struct _xmlDictEntry *next;
    xmlChar *name;
    int len;
    int valid;
};

/*
 * The entire dictionnary
 */
struct _xmlDict {
    struct _xmlDictEntry *dict;
    int size;
    int nbElems;
};

/*
 * xmlDictComputeKey:
 * Calculate the hash key
 */
static unsigned long
xmlDictComputeKey(xmlDictPtr dict, const xmlChar *name, int namelen) {
    unsigned long value = 0L;
    char ch;
    int len = 0;
    
    if (name == NULL) return(0);
    value += 30 * (*name);
    if (namelen > 10) {
        value += name[namelen - 1];
        namelen = 10;
    }
    switch (namelen) {
        case 10: value += name[9];
        case 9: value += name[8];
        case 8: value += name[7];
        case 7: value += name[6];
        case 6: value += name[5];
        case 5: value += name[4];
        case 4: value += name[3];
        case 3: value += name[2];
        case 2: value += name[1];
        case 1: value += name[0];
        default: break;
    }
#if 0
    while ((len++ < namelen) && ((ch = *name++) != 0)) {
	value += (unsigned long)ch;
    }
#endif
#if 0
    while ((len++ < namelen) && ((ch = *name++) != 0)) {
	value = value ^ ((value << 5) + (value >> 3) + (unsigned long)ch);
    }
#endif
    return (value % dict->size);
}

/**
 * xmlDictCreate:
 *
 * Create a new dictionary
 *
 * Returns the newly created object, or NULL if an error occured.
 */
xmlDictPtr
xmlDictCreate(void) {
    xmlDictPtr dict;
  
    dict = xmlMalloc(sizeof(xmlDict));
    if (dict) {
        dict->size = MIN_DICT_SIZE;
	dict->nbElems = 0;
        dict->dict = xmlMalloc(MIN_DICT_SIZE * sizeof(xmlDictEntry));
        if (dict->dict) {
  	    memset(dict->dict, 0, MIN_DICT_SIZE * sizeof(xmlDictEntry));
  	    return(dict);
        }
        xmlFree(dict);
    }
    return(NULL);
}

/**
 * xmlDictGrow:
 * @dict: the dictionnary
 * @size: the new size of the dictionnary
 *
 * resize the dictionnary
 *
 * Returns 0 in case of success, -1 in case of failure
 */
static int
xmlDictGrow(xmlDictPtr dict, int size) {
    unsigned long key;
    int oldsize, i;
    xmlDictEntryPtr iter, next;
    struct _xmlDictEntry *olddict;
#ifdef DEBUG_GROW
    unsigned long nbElem = 0;
#endif
  
    if (dict == NULL)
	return(-1);
    if (size < 8)
        return(-1);
    if (size > 8 * 2048)
	return(-1);

    oldsize = dict->size;
    olddict = dict->dict;
    if (olddict == NULL)
        return(-1);
  
    dict->dict = xmlMalloc(size * sizeof(xmlDictEntry));
    if (dict->dict == NULL) {
	dict->dict = olddict;
	return(-1);
    }
    memset(dict->dict, 0, size * sizeof(xmlDictEntry));
    dict->size = size;

    /*	If the two loops are merged, there would be situations where
	a new entry needs to allocated and data copied into it from 
	the main dict. So instead, we run through the array twice, first
	copying all the elements in the main array (where we can't get
	conflicts) and then the rest, so we only free (and don't allocate)
    */
    for (i = 0; i < oldsize; i++) {
	if (olddict[i].valid == 0) 
	    continue;
	key = xmlDictComputeKey(dict, olddict[i].name, olddict[i].len);
	memcpy(&(dict->dict[key]), &(olddict[i]), sizeof(xmlDictEntry));
	dict->dict[key].next = NULL;
#ifdef DEBUG_GROW
	nbElem++;
#endif
    }

    for (i = 0; i < oldsize; i++) {
	iter = olddict[i].next;
	while (iter) {
	    next = iter->next;

	    /*
	     * put back the entry in the new dict
	     */

	    key = xmlDictComputeKey(dict, iter->name, iter->len);
	    if (dict->dict[key].valid == 0) {
		memcpy(&(dict->dict[key]), iter, sizeof(xmlDictEntry));
		dict->dict[key].next = NULL;
		dict->dict[key].valid = 1;
		xmlFree(iter);
	    } else {
	    	iter->next = dict->dict[key].next;
	    	dict->dict[key].next = iter;
	    }

#ifdef DEBUG_GROW
	    nbElem++;
#endif

	    iter = next;
	}
    }

    xmlFree(olddict);

#ifdef DEBUG_GROW
    xmlGenericError(xmlGenericErrorContext,
	    "xmlDictGrow : from %d to %d, %d elems\n", oldsize, size, nbElem);
#endif

    return(0);
}

/**
 * xmlDictFree:
 * @dict: the dictionnary
 *
 * Free the hash @dict and its contents. The userdata is
 * deallocated with @f if provided.
 */
void
xmlDictFree(xmlDictPtr dict) {
    int i;
    xmlDictEntryPtr iter;
    xmlDictEntryPtr next;
    int inside_dict = 0;

    if (dict == NULL)
	return;
    if (dict->dict) {
	for(i = 0; ((i < dict->size) && (dict->nbElems > 0)); i++) {
	    iter = &(dict->dict[i]);
	    if (iter->valid == 0)
		continue;
	    inside_dict = 1;
	    while (iter) {
		next = iter->next;
		if (iter->name)
		    xmlFree(iter->name);
		if (!inside_dict)
		    xmlFree(iter);
		dict->nbElems--;
		inside_dict = 0;
		iter = next;
	    }
	    inside_dict = 0;
	}
	xmlFree(dict->dict);
    }
    xmlFree(dict);
}

/**
 * xmlDictLookup:
 * @dict: the dictionnary
 * @name: the name of the userdata
 * @len: the length of the name
 *
 * Add the @name to the hash @dict if not present.
 *
 * Returns the internal copy of the name or NULL in case of internal error
 */
const xmlChar *
xmlDictLookup(xmlDictPtr dict, const xmlChar *name, int len) {
    unsigned long key, nbi = 0;
    xmlDictEntryPtr entry;
    xmlDictEntryPtr insert;
    const xmlChar *ret;

    if ((dict == NULL) || (name == NULL) || (len <= 0))
	return(NULL);

    /*
     * Check for duplicate and insertion location.
     */
    key = xmlDictComputeKey(dict, name, len);
    if (dict->dict[key].valid == 0) {
	insert = NULL;
    } else {
	for (insert = &(dict->dict[key]); insert->next != NULL;
	     insert = insert->next) {
	    if ((insert->len == len) &&
	        (!xmlStrncmp(insert->name, name, len)))
		return(insert->name);
	    nbi++;
	}
	if ((insert->len == len) &&
	    (!xmlStrncmp(insert->name, name, len)))
	    return(insert->name);
    }

    if (insert == NULL) {
	entry = &(dict->dict[key]);
    } else {
	entry = xmlMalloc(sizeof(xmlDictEntry));
	if (entry == NULL)
	     return(NULL);
    }

    ret = entry->name = xmlStrndup(name, len);
    entry->len = len;
    entry->next = NULL;
    entry->valid = 1;


    if (insert != NULL) 
	insert->next = entry;

    dict->nbElems++;

    if ((nbi > MAX_HASH_LEN) &&
        (dict->size <= ((MAX_DICT_HASH / 2) / MAX_HASH_LEN)))
	xmlDictGrow(dict, MAX_HASH_LEN * 2 * dict->size);
    /* Note that entry may have been freed at this point by xmlDictGrow */

    return(ret);
}

/**
 * xmlDictSize:
 * @dict: the dictionnary
 *
 * Query the number of elements installed in the hash @dict.
 *
 * Returns the number of elements in the dictionnary or
 * -1 in case of error
 */
int
xmlDictSize(xmlDictPtr dict) {
    if (dict == NULL)
	return(-1);
    return(dict->nbElems);
}