/src/tinysparql/subprojects/libstemmer_c-3.0.1/runtime/utilities.c

Source

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "header.h"

#define CREATE_SIZE 1

extern symbol * create_s(void) {
    symbol * p;
    void * mem = malloc(HEAD + (CREATE_SIZE + 1) * sizeof(symbol));
    if (mem == NULL) return NULL;
    p = (symbol *) (HEAD + (char *) mem);
    CAPACITY(p) = CREATE_SIZE;
    SET_SIZE(p, 0);
    return p;
}

extern void lose_s(symbol * p) {
    if (p == NULL) return;
    free((char *) p - HEAD);
}

/*
   new_p = skip_utf8(p, c, l, n); skips n characters forwards from p + c.
   new_p is the new position, or -1 on failure.

   -- used to implement hop and next in the utf8 case.
*/

extern int skip_utf8(const symbol * p, int c, int limit, int n) {
    int b;
    if (n < 0) return -1;
    for (; n > 0; n--) {
        if (c >= limit) return -1;
        b = p[c++];
        if (b >= 0xC0) {   /* 1100 0000 */
            while (c < limit) {
                b = p[c];
                if (b >= 0xC0 || b < 0x80) break;
                /* break unless b is 10------ */
                c++;
            }
        }
    }
    return c;
}

/*
   new_p = skip_b_utf8(p, c, lb, n); skips n characters backwards from p + c - 1
   new_p is the new position, or -1 on failure.

   -- used to implement hop and next in the utf8 case.
*/

extern int skip_b_utf8(const symbol * p, int c, int limit, int n) {
    int b;
    if (n < 0) return -1;
    for (; n > 0; n--) {
        if (c <= limit) return -1;
        b = p[--c];
        if (b >= 0x80) {   /* 1000 0000 */
            while (c > limit) {
                b = p[c];
                if (b >= 0xC0) break; /* 1100 0000 */
                c--;
            }
        }
    }
    return c;
}

/* Code for character groupings: utf8 cases */

static int get_utf8(const symbol * p, int c, int l, int * slot) {
    int b0, b1, b2;
    if (c >= l) return 0;
    b0 = p[c++];
    if (b0 < 0xC0 || c == l) {   /* 1100 0000 */
        *slot = b0;
        return 1;
    }
    b1 = p[c++] & 0x3F;
    if (b0 < 0xE0 || c == l) {   /* 1110 0000 */
        *slot = (b0 & 0x1F) << 6 | b1;
        return 2;
    }
    b2 = p[c++] & 0x3F;
    if (b0 < 0xF0 || c == l) {   /* 1111 0000 */
        *slot = (b0 & 0xF) << 12 | b1 << 6 | b2;
        return 3;
    }
    *slot = (b0 & 0x7) << 18 | b1 << 12 | b2 << 6 | (p[c] & 0x3F);
    return 4;
}

static int get_b_utf8(const symbol * p, int c, int lb, int * slot) {
    int a, b;
    if (c <= lb) return 0;
    b = p[--c];
    if (b < 0x80 || c == lb) {   /* 1000 0000 */
        *slot = b;
        return 1;
    }
    a = b & 0x3F;
    b = p[--c];
    if (b >= 0xC0 || c == lb) {   /* 1100 0000 */
        *slot = (b & 0x1F) << 6 | a;
        return 2;
    }
    a |= (b & 0x3F) << 6;
    b = p[--c];
    if (b >= 0xE0 || c == lb) {   /* 1110 0000 */
        *slot = (b & 0xF) << 12 | a;
        return 3;
    }
    *slot = (p[--c] & 0x7) << 18 | (b & 0x3F) << 12 | a;
    return 4;
}

extern int in_grouping_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        int w = get_utf8(z->p, z->c, z->l, & ch);
        if (!w) return -1;
        if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
            return w;
        z->c += w;
    } while (repeat);
    return 0;
}

extern int in_grouping_b_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        int w = get_b_utf8(z->p, z->c, z->lb, & ch);
        if (!w) return -1;
        if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
            return w;
        z->c -= w;
    } while (repeat);
    return 0;
}

extern int out_grouping_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        int w = get_utf8(z->p, z->c, z->l, & ch);
        if (!w) return -1;
        if (!(ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
            return w;
        z->c += w;
    } while (repeat);
    return 0;
}

extern int out_grouping_b_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        int w = get_b_utf8(z->p, z->c, z->lb, & ch);
        if (!w) return -1;
        if (!(ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
            return w;
        z->c -= w;
    } while (repeat);
    return 0;
}

/* Code for character groupings: non-utf8 cases */

extern int in_grouping(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        if (z->c >= z->l) return -1;
        ch = z->p[z->c];
        if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
            return 1;
        z->c++;
    } while (repeat);
    return 0;
}

extern int in_grouping_b(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        if (z->c <= z->lb) return -1;
        ch = z->p[z->c - 1];
        if (ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
            return 1;
        z->c--;
    } while (repeat);
    return 0;
}

extern int out_grouping(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        if (z->c >= z->l) return -1;
        ch = z->p[z->c];
        if (!(ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
            return 1;
        z->c++;
    } while (repeat);
    return 0;
}

extern int out_grouping_b(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
    do {
        int ch;
        if (z->c <= z->lb) return -1;
        ch = z->p[z->c - 1];
        if (!(ch > max || (ch -= min) < 0 || (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
            return 1;
        z->c--;
    } while (repeat);
    return 0;
}

extern int eq_s(struct SN_env * z, int s_size, const symbol * s) {
    if (z->l - z->c < s_size || memcmp(z->p + z->c, s, s_size * sizeof(symbol)) != 0) return 0;
    z->c += s_size; return 1;
}

extern int eq_s_b(struct SN_env * z, int s_size, const symbol * s) {
    if (z->c - z->lb < s_size || memcmp(z->p + z->c - s_size, s, s_size * sizeof(symbol)) != 0) return 0;
    z->c -= s_size; return 1;
}

extern int eq_v(struct SN_env * z, const symbol * p) {
    return eq_s(z, SIZE(p), p);
}

extern int eq_v_b(struct SN_env * z, const symbol * p) {
    return eq_s_b(z, SIZE(p), p);
}

extern int find_among(struct SN_env * z, const struct among * v, int v_size) {

    int i = 0;
    int j = v_size;

    int c = z->c; int l = z->l;
    const symbol * q = z->p + c;

    const struct among * w;

    int common_i = 0;
    int common_j = 0;

    int first_key_inspected = 0;

    while (1) {
        int k = i + ((j - i) >> 1);
        int diff = 0;
        int common = common_i < common_j ? common_i : common_j; /* smaller */
        w = v + k;
        {
            int i2; for (i2 = common; i2 < w->s_size; i2++) {
                if (c + common == l) { diff = -1; break; }
                diff = q[common] - w->s[i2];
                if (diff != 0) break;
                common++;
            }
        }
        if (diff < 0) {
            j = k;
            common_j = common;
        } else {
            i = k;
            common_i = common;
        }
        if (j - i <= 1) {
            if (i > 0) break; /* v->s has been inspected */
            if (j == i) break; /* only one item in v */

            /* - but now we need to go round once more to get
               v->s inspected. This looks messy, but is actually
               the optimal approach.  */

            if (first_key_inspected) break;
            first_key_inspected = 1;
        }
    }
    w = v + i;
    while (1) {
        if (common_i >= w->s_size) {
            z->c = c + w->s_size;
            if (w->function == NULL) return w->result;
            {
                int res = w->function(z);
                z->c = c + w->s_size;
                if (res) return w->result;
            }
        }
        if (!w->substring_i) return 0;
        w += w->substring_i;
    }
}

/* find_among_b is for backwards processing. Same comments apply */

extern int find_among_b(struct SN_env * z, const struct among * v, int v_size) {

    int i = 0;
    int j = v_size;

    int c = z->c; int lb = z->lb;
    const symbol * q = z->p + c - 1;

    const struct among * w;

    int common_i = 0;
    int common_j = 0;

    int first_key_inspected = 0;

    while (1) {
        int k = i + ((j - i) >> 1);
        int diff = 0;
        int common = common_i < common_j ? common_i : common_j;
        w = v + k;
        {
            int i2; for (i2 = w->s_size - 1 - common; i2 >= 0; i2--) {
                if (c - common == lb) { diff = -1; break; }
                diff = q[- common] - w->s[i2];
                if (diff != 0) break;
                common++;
            }
        }
        if (diff < 0) { j = k; common_j = common; }
                 else { i = k; common_i = common; }
        if (j - i <= 1) {
            if (i > 0) break;
            if (j == i) break;
            if (first_key_inspected) break;
            first_key_inspected = 1;
        }
    }
    w = v + i;
    while (1) {
        if (common_i >= w->s_size) {
            z->c = c - w->s_size;
            if (w->function == NULL) return w->result;
            {
                int res = w->function(z);
                z->c = c - w->s_size;
                if (res) return w->result;
            }
        }
        if (!w->substring_i) return 0;
        w += w->substring_i;
    }
}


/* Increase the size of the buffer pointed to by p to at least n symbols.
 * If insufficient memory, returns NULL and frees the old buffer.
 */
static symbol * increase_size(symbol * p, int n) {
    symbol * q;
    int new_size = n + 20;
    void * mem = realloc((char *) p - HEAD,
                         HEAD + (new_size + 1) * sizeof(symbol));
    if (mem == NULL) {
        lose_s(p);
        return NULL;
    }
    q = (symbol *) (HEAD + (char *)mem);
    CAPACITY(q) = new_size;
    return q;
}

/* to replace symbols between c_bra and c_ket in z->p by the
   s_size symbols at s.
   Returns 0 on success, -1 on error.
   Also, frees z->p (and sets it to NULL) on error.
*/
extern int replace_s(struct SN_env * z, int c_bra, int c_ket, int s_size, const symbol * s, int * adjptr)
{
    int adjustment;
    int len;
    if (z->p == NULL) {
        z->p = create_s();
        if (z->p == NULL) return -1;
    }
    adjustment = s_size - (c_ket - c_bra);
    len = SIZE(z->p);
    if (adjustment != 0) {
        if (adjustment + len > CAPACITY(z->p)) {
            z->p = increase_size(z->p, adjustment + len);
            if (z->p == NULL) return -1;
        }
        memmove(z->p + c_ket + adjustment,
                z->p + c_ket,
                (len - c_ket) * sizeof(symbol));
        SET_SIZE(z->p, adjustment + len);
        z->l += adjustment;
        if (z->c >= c_ket)
            z->c += adjustment;
        else if (z->c > c_bra)
            z->c = c_bra;
    }
    if (s_size) memmove(z->p + c_bra, s, s_size * sizeof(symbol));
    if (adjptr != NULL)
        *adjptr = adjustment;
    return 0;
}

static int slice_check(struct SN_env * z) {

    if (z->bra < 0 ||
        z->bra > z->ket ||
        z->ket > z->l ||
        z->p == NULL ||
        z->l > SIZE(z->p)) /* this line could be removed */
    {
#if 0
        fprintf(stderr, "faulty slice operation:\n");
        debug(z, -1, 0);
#endif
        return -1;
    }
    return 0;
}

extern int slice_from_s(struct SN_env * z, int s_size, const symbol * s) {
    if (slice_check(z)) return -1;
    return replace_s(z, z->bra, z->ket, s_size, s, NULL);
}

extern int slice_from_v(struct SN_env * z, const symbol * p) {
    return slice_from_s(z, SIZE(p), p);
}

extern int slice_del(struct SN_env * z) {
    return slice_from_s(z, 0, NULL);
}

extern int insert_s(struct SN_env * z, int bra, int ket, int s_size, const symbol * s) {
    int adjustment;
    if (replace_s(z, bra, ket, s_size, s, &adjustment))
        return -1;
    if (bra <= z->bra) z->bra += adjustment;
    if (bra <= z->ket) z->ket += adjustment;
    return 0;
}

extern int insert_v(struct SN_env * z, int bra, int ket, const symbol * p) {
    return insert_s(z, bra, ket, SIZE(p), p);
}

extern symbol * slice_to(struct SN_env * z, symbol * p) {
    if (slice_check(z)) {
        lose_s(p);
        return NULL;
    }
    {
        int len = z->ket - z->bra;
        if (CAPACITY(p) < len) {
            p = increase_size(p, len);
            if (p == NULL)
                return NULL;
        }
        memmove(p, z->p + z->bra, len * sizeof(symbol));
        SET_SIZE(p, len);
    }
    return p;
}

extern symbol * assign_to(struct SN_env * z, symbol * p) {
    int len = z->l;
    if (CAPACITY(p) < len) {
        p = increase_size(p, len);
        if (p == NULL)
            return NULL;
    }
    memmove(p, z->p, len * sizeof(symbol));
    SET_SIZE(p, len);
    return p;
}

extern int len_utf8(const symbol * p) {
    int size = SIZE(p);
    int len = 0;
    while (size--) {
        symbol b = *p++;
        if (b >= 0xC0 || b < 0x80) ++len;
    }
    return len;
}

#if 0
extern void debug(struct SN_env * z, int number, int line_count) {
    int i;
    int limit = SIZE(z->p);
    /*if (number >= 0) printf("%3d (line %4d): '", number, line_count);*/
    if (number >= 0) printf("%3d (line %4d): [%d]'", number, line_count,limit);
    for (i = 0; i <= limit; i++) {
        if (z->lb == i) printf("{");
        if (z->bra == i) printf("[");
        if (z->c == i) printf("|");
        if (z->ket == i) printf("]");
        if (z->l == i) printf("}");
        if (i < limit)
        {   int ch = z->p[i];
            if (ch == 0) ch = '#';
            printf("%c", ch);
        }
    }
    printf("'\n");
}
#endif

Coverage Report

Created: 2025-11-09 06:54

Line	Count	Source
1
2		#include <stdio.h>
3		#include <stdlib.h>
4		#include <string.h>
5
6		#include "header.h"
7
8	0	#define CREATE_SIZE 1
9
10	0	extern symbol * create_s(void) {
11	0	symbol * p;
12	0	void * mem = malloc(HEAD + (CREATE_SIZE + 1) * sizeof(symbol));
13	0	if (mem == NULL) return NULL;
14	0	p = (symbol ) (HEAD + (char ) mem);
15	0	CAPACITY(p) = CREATE_SIZE;
16	0	SET_SIZE(p, 0);
17	0	return p;
18	0	}
19
20	0	extern void lose_s(symbol * p) {
21	0	if (p == NULL) return;
22	0	free((char *) p - HEAD);
23	0	}
24
25		/*
26		new_p = skip_utf8(p, c, l, n); skips n characters forwards from p + c.
27		new_p is the new position, or -1 on failure.
28
29		-- used to implement hop and next in the utf8 case.
30		*/
31
32	0	extern int skip_utf8(const symbol * p, int c, int limit, int n) {
33	0	int b;
34	0	if (n < 0) return -1;
35	0	for (; n > 0; n--) {
36	0	if (c >= limit) return -1;
37	0	b = p[c++];
38	0	if (b >= 0xC0) { /* 1100 0000 */
39	0	while (c < limit) {
40	0	b = p[c];
41	0	if (b >= 0xC0 \|\| b < 0x80) break;
42		/* break unless b is 10------ */
43	0	c++;
44	0	}
45	0	}
46	0	}
47	0	return c;
48	0	}
49
50		/*
51		new_p = skip_b_utf8(p, c, lb, n); skips n characters backwards from p + c - 1
52		new_p is the new position, or -1 on failure.
53
54		-- used to implement hop and next in the utf8 case.
55		*/
56
57	0	extern int skip_b_utf8(const symbol * p, int c, int limit, int n) {
58	0	int b;
59	0	if (n < 0) return -1;
60	0	for (; n > 0; n--) {
61	0	if (c <= limit) return -1;
62	0	b = p[--c];
63	0	if (b >= 0x80) { /* 1000 0000 */
64	0	while (c > limit) {
65	0	b = p[c];
66	0	if (b >= 0xC0) break; /* 1100 0000 */
67	0	c--;
68	0	}
69	0	}
70	0	}
71	0	return c;
72	0	}
73
74		/* Code for character groupings: utf8 cases */
75
76	0	static int get_utf8(const symbol * p, int c, int l, int * slot) {
77	0	int b0, b1, b2;
78	0	if (c >= l) return 0;
79	0	b0 = p[c++];
80	0	if (b0 < 0xC0 \|\| c == l) { /* 1100 0000 */
81	0	*slot = b0;
82	0	return 1;
83	0	}
84	0	b1 = p[c++] & 0x3F;
85	0	if (b0 < 0xE0 \|\| c == l) { /* 1110 0000 */
86	0	*slot = (b0 & 0x1F) << 6 \| b1;
87	0	return 2;
88	0	}
89	0	b2 = p[c++] & 0x3F;
90	0	if (b0 < 0xF0 \|\| c == l) { /* 1111 0000 */
91	0	*slot = (b0 & 0xF) << 12 \| b1 << 6 \| b2;
92	0	return 3;
93	0	}
94	0	*slot = (b0 & 0x7) << 18 \| b1 << 12 \| b2 << 6 \| (p[c] & 0x3F);
95	0	return 4;
96	0	}
97
98	0	static int get_b_utf8(const symbol * p, int c, int lb, int * slot) {
99	0	int a, b;
100	0	if (c <= lb) return 0;
101	0	b = p[--c];
102	0	if (b < 0x80 \|\| c == lb) { /* 1000 0000 */
103	0	*slot = b;
104	0	return 1;
105	0	}
106	0	a = b & 0x3F;
107	0	b = p[--c];
108	0	if (b >= 0xC0 \|\| c == lb) { /* 1100 0000 */
109	0	*slot = (b & 0x1F) << 6 \| a;
110	0	return 2;
111	0	}
112	0	a \|= (b & 0x3F) << 6;
113	0	b = p[--c];
114	0	if (b >= 0xE0 \|\| c == lb) { /* 1110 0000 */
115	0	*slot = (b & 0xF) << 12 \| a;
116	0	return 3;
117	0	}
118	0	*slot = (p[--c] & 0x7) << 18 \| (b & 0x3F) << 12 \| a;
119	0	return 4;
120	0	}
121
122	0	extern int in_grouping_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
123	0	do {
124	0	int ch;
125	0	int w = get_utf8(z->p, z->c, z->l, & ch);
126	0	if (!w) return -1;
127	0	if (ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
128	0	return w;
129	0	z->c += w;
130	0	} while (repeat);
131	0	return 0;
132	0	}
133
134	0	extern int in_grouping_b_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
135	0	do {
136	0	int ch;
137	0	int w = get_b_utf8(z->p, z->c, z->lb, & ch);
138	0	if (!w) return -1;
139	0	if (ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
140	0	return w;
141	0	z->c -= w;
142	0	} while (repeat);
143	0	return 0;
144	0	}
145
146	0	extern int out_grouping_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
147	0	do {
148	0	int ch;
149	0	int w = get_utf8(z->p, z->c, z->l, & ch);
150	0	if (!w) return -1;
151	0	if (!(ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
152	0	return w;
153	0	z->c += w;
154	0	} while (repeat);
155	0	return 0;
156	0	}
157
158	0	extern int out_grouping_b_U(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
159	0	do {
160	0	int ch;
161	0	int w = get_b_utf8(z->p, z->c, z->lb, & ch);
162	0	if (!w) return -1;
163	0	if (!(ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
164	0	return w;
165	0	z->c -= w;
166	0	} while (repeat);
167	0	return 0;
168	0	}
169
170		/* Code for character groupings: non-utf8 cases */
171
172	0	extern int in_grouping(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
173	0	do {
174	0	int ch;
175	0	if (z->c >= z->l) return -1;
176	0	ch = z->p[z->c];
177	0	if (ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
178	0	return 1;
179	0	z->c++;
180	0	} while (repeat);
181	0	return 0;
182	0	}
183
184	0	extern int in_grouping_b(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
185	0	do {
186	0	int ch;
187	0	if (z->c <= z->lb) return -1;
188	0	ch = z->p[z->c - 1];
189	0	if (ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0)
190	0	return 1;
191	0	z->c--;
192	0	} while (repeat);
193	0	return 0;
194	0	}
195
196	0	extern int out_grouping(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
197	0	do {
198	0	int ch;
199	0	if (z->c >= z->l) return -1;
200	0	ch = z->p[z->c];
201	0	if (!(ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
202	0	return 1;
203	0	z->c++;
204	0	} while (repeat);
205	0	return 0;
206	0	}
207
208	0	extern int out_grouping_b(struct SN_env * z, const unsigned char * s, int min, int max, int repeat) {
209	0	do {
210	0	int ch;
211	0	if (z->c <= z->lb) return -1;
212	0	ch = z->p[z->c - 1];
213	0	if (!(ch > max \|\| (ch -= min) < 0 \|\| (s[ch >> 3] & (0X1 << (ch & 0X7))) == 0))
214	0	return 1;
215	0	z->c--;
216	0	} while (repeat);
217	0	return 0;
218	0	}
219
220	0	extern int eq_s(struct SN_env * z, int s_size, const symbol * s) {
221	0	if (z->l - z->c < s_size \|\| memcmp(z->p + z->c, s, s_size * sizeof(symbol)) != 0) return 0;
222	0	z->c += s_size; return 1;
223	0	}
224
225	0	extern int eq_s_b(struct SN_env * z, int s_size, const symbol * s) {
226	0	if (z->c - z->lb < s_size \|\| memcmp(z->p + z->c - s_size, s, s_size * sizeof(symbol)) != 0) return 0;
227	0	z->c -= s_size; return 1;
228	0	}
229
230	0	extern int eq_v(struct SN_env * z, const symbol * p) {
231	0	return eq_s(z, SIZE(p), p);
232	0	}
233
234	0	extern int eq_v_b(struct SN_env * z, const symbol * p) {
235	0	return eq_s_b(z, SIZE(p), p);
236	0	}
237
238	0	extern int find_among(struct SN_env * z, const struct among * v, int v_size) {
239
240	0	int i = 0;
241	0	int j = v_size;
242
243	0	int c = z->c; int l = z->l;
244	0	const symbol * q = z->p + c;
245
246	0	const struct among * w;
247
248	0	int common_i = 0;
249	0	int common_j = 0;
250
251	0	int first_key_inspected = 0;
252
253	0	while (1) {
254	0	int k = i + ((j - i) >> 1);
255	0	int diff = 0;
256	0	int common = common_i < common_j ? common_i : common_j; /* smaller */
257	0	w = v + k;
258	0	{
259	0	int i2; for (i2 = common; i2 < w->s_size; i2++) {
260	0	if (c + common == l) { diff = -1; break; }
261	0	diff = q[common] - w->s[i2];
262	0	if (diff != 0) break;
263	0	common++;
264	0	}
265	0	}
266	0	if (diff < 0) {
267	0	j = k;
268	0	common_j = common;
269	0	} else {
270	0	i = k;
271	0	common_i = common;
272	0	}
273	0	if (j - i <= 1) {
274	0	if (i > 0) break; /* v->s has been inspected */
275	0	if (j == i) break; /* only one item in v */
276
277		/* - but now we need to go round once more to get
278		v->s inspected. This looks messy, but is actually
279		the optimal approach. */
280
281	0	if (first_key_inspected) break;
282	0	first_key_inspected = 1;
283	0	}
284	0	}
285	0	w = v + i;
286	0	while (1) {
287	0	if (common_i >= w->s_size) {
288	0	z->c = c + w->s_size;
289	0	if (w->function == NULL) return w->result;
290	0	{
291	0	int res = w->function(z);
292	0	z->c = c + w->s_size;
293	0	if (res) return w->result;
294	0	}
295	0	}
296	0	if (!w->substring_i) return 0;
297	0	w += w->substring_i;
298	0	}
299	0	}
300
301		/* find_among_b is for backwards processing. Same comments apply */
302
303	0	extern int find_among_b(struct SN_env * z, const struct among * v, int v_size) {
304
305	0	int i = 0;
306	0	int j = v_size;
307
308	0	int c = z->c; int lb = z->lb;
309	0	const symbol * q = z->p + c - 1;
310
311	0	const struct among * w;
312
313	0	int common_i = 0;
314	0	int common_j = 0;
315
316	0	int first_key_inspected = 0;
317
318	0	while (1) {
319	0	int k = i + ((j - i) >> 1);
320	0	int diff = 0;
321	0	int common = common_i < common_j ? common_i : common_j;
322	0	w = v + k;
323	0	{
324	0	int i2; for (i2 = w->s_size - 1 - common; i2 >= 0; i2--) {
325	0	if (c - common == lb) { diff = -1; break; }
326	0	diff = q[- common] - w->s[i2];
327	0	if (diff != 0) break;
328	0	common++;
329	0	}
330	0	}
331	0	if (diff < 0) { j = k; common_j = common; }
332	0	else { i = k; common_i = common; }
333	0	if (j - i <= 1) {
334	0	if (i > 0) break;
335	0	if (j == i) break;
336	0	if (first_key_inspected) break;
337	0	first_key_inspected = 1;
338	0	}
339	0	}
340	0	w = v + i;
341	0	while (1) {
342	0	if (common_i >= w->s_size) {
343	0	z->c = c - w->s_size;
344	0	if (w->function == NULL) return w->result;
345	0	{
346	0	int res = w->function(z);
347	0	z->c = c - w->s_size;
348	0	if (res) return w->result;
349	0	}
350	0	}
351	0	if (!w->substring_i) return 0;
352	0	w += w->substring_i;
353	0	}
354	0	}
355
356
357		/* Increase the size of the buffer pointed to by p to at least n symbols.
358		* If insufficient memory, returns NULL and frees the old buffer.
359		*/
360	0	static symbol * increase_size(symbol * p, int n) {
361	0	symbol * q;
362	0	int new_size = n + 20;
363	0	void * mem = realloc((char *) p - HEAD,
364	0	HEAD + (new_size + 1) * sizeof(symbol));
365	0	if (mem == NULL) {
366	0	lose_s(p);
367	0	return NULL;
368	0	}
369	0	q = (symbol ) (HEAD + (char )mem);
370	0	CAPACITY(q) = new_size;
371	0	return q;
372	0	}
373
374		/* to replace symbols between c_bra and c_ket in z->p by the
375		s_size symbols at s.
376		Returns 0 on success, -1 on error.
377		Also, frees z->p (and sets it to NULL) on error.
378		*/
379		extern int replace_s(struct SN_env * z, int c_bra, int c_ket, int s_size, const symbol * s, int * adjptr)
380	0	{
381	0	int adjustment;
382	0	int len;
383	0	if (z->p == NULL) {
384	0	z->p = create_s();
385	0	if (z->p == NULL) return -1;
386	0	}
387	0	adjustment = s_size - (c_ket - c_bra);
388	0	len = SIZE(z->p);
389	0	if (adjustment != 0) {
390	0	if (adjustment + len > CAPACITY(z->p)) {
391	0	z->p = increase_size(z->p, adjustment + len);
392	0	if (z->p == NULL) return -1;
393	0	}
394	0	memmove(z->p + c_ket + adjustment,
395	0	z->p + c_ket,
396	0	(len - c_ket) * sizeof(symbol));
397	0	SET_SIZE(z->p, adjustment + len);
398	0	z->l += adjustment;
399	0	if (z->c >= c_ket)
400	0	z->c += adjustment;
401	0	else if (z->c > c_bra)
402	0	z->c = c_bra;
403	0	}
404	0	if (s_size) memmove(z->p + c_bra, s, s_size * sizeof(symbol));
405	0	if (adjptr != NULL)
406	0	*adjptr = adjustment;
407	0	return 0;
408	0	}
409
410	0	static int slice_check(struct SN_env * z) {
411
412	0	if (z->bra < 0 \|\|
413	0	z->bra > z->ket \|\|
414	0	z->ket > z->l \|\|
415	0	z->p == NULL \|\|
416	0	z->l > SIZE(z->p)) /* this line could be removed */
417	0	{
418		#if 0
419		fprintf(stderr, "faulty slice operation:\n");
420		debug(z, -1, 0);
421		#endif
422	0	return -1;
423	0	}
424	0	return 0;
425	0	}
426
427	0	extern int slice_from_s(struct SN_env * z, int s_size, const symbol * s) {
428	0	if (slice_check(z)) return -1;
429	0	return replace_s(z, z->bra, z->ket, s_size, s, NULL);
430	0	}
431
432	0	extern int slice_from_v(struct SN_env * z, const symbol * p) {
433	0	return slice_from_s(z, SIZE(p), p);
434	0	}
435
436	0	extern int slice_del(struct SN_env * z) {
437	0	return slice_from_s(z, 0, NULL);
438	0	}
439
440	0	extern int insert_s(struct SN_env * z, int bra, int ket, int s_size, const symbol * s) {
441	0	int adjustment;
442	0	if (replace_s(z, bra, ket, s_size, s, &adjustment))
443	0	return -1;
444	0	if (bra <= z->bra) z->bra += adjustment;
445	0	if (bra <= z->ket) z->ket += adjustment;
446	0	return 0;
447	0	}
448
449	0	extern int insert_v(struct SN_env * z, int bra, int ket, const symbol * p) {
450	0	return insert_s(z, bra, ket, SIZE(p), p);
451	0	}
452
453	0	extern symbol * slice_to(struct SN_env * z, symbol * p) {
454	0	if (slice_check(z)) {
455	0	lose_s(p);
456	0	return NULL;
457	0	}
458	0	{
459	0	int len = z->ket - z->bra;
460	0	if (CAPACITY(p) < len) {
461	0	p = increase_size(p, len);
462	0	if (p == NULL)
463	0	return NULL;
464	0	}
465	0	memmove(p, z->p + z->bra, len * sizeof(symbol));
466	0	SET_SIZE(p, len);
467	0	}
468	0	return p;
469	0	}
470
471	0	extern symbol * assign_to(struct SN_env * z, symbol * p) {
472	0	int len = z->l;
473	0	if (CAPACITY(p) < len) {
474	0	p = increase_size(p, len);
475	0	if (p == NULL)
476	0	return NULL;
477	0	}
478	0	memmove(p, z->p, len * sizeof(symbol));
479	0	SET_SIZE(p, len);
480	0	return p;
481	0	}
482
483	0	extern int len_utf8(const symbol * p) {
484	0	int size = SIZE(p);
485	0	int len = 0;
486	0	while (size--) {
487	0	symbol b = *p++;
488	0	if (b >= 0xC0 \|\| b < 0x80) ++len;
489	0	}
490	0	return len;
491	0	}
492
493		#if 0
494		extern void debug(struct SN_env * z, int number, int line_count) {
495		int i;
496		int limit = SIZE(z->p);
497		/if (number >= 0) printf("%3d (line %4d): '", number, line_count);/
498		if (number >= 0) printf("%3d (line %4d): [%d]'", number, line_count,limit);
499		for (i = 0; i <= limit; i++) {
500		if (z->lb == i) printf("{");
501		if (z->bra == i) printf("[");
502		if (z->c == i) printf("\|");
503		if (z->ket == i) printf("]");
504		if (z->l == i) printf("}");
505		if (i < limit)
506		{ int ch = z->p[i];
507		if (ch == 0) ch = '#';
508		printf("%c", ch);
509		}
510		}
511		printf("'\n");
512		}
513		#endif