aboutsummaryrefslogtreecommitdiffstats
path: root/libparser.c
blob: 9ee22a12668fec2f15ffce3828c90d10a2406f4f (plain) (blame)
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
/* See LICENSE file for copyright and license details. */
#include "libparser.h"
#include <libsimple.h>


struct context {
	const struct libparser_rule *const *rules;
	struct libparser_unit *cache;
	const char *data;
	size_t length;
	size_t position;
	int done;
	int exception;
};


static void
free_unit(struct libparser_unit *unit, struct context *ctx)
{
	struct libparser_unit *prev;
	while (unit) {
		free_unit(unit->in, ctx);
		prev = unit;
		unit = unit->next;
		prev->next = ctx->cache;
		ctx->cache = prev;
	}
}


static struct libparser_unit *
try_match(const char *rule, const union libparser_sentence *sentence, struct context *ctx)
{
	struct libparser_unit *unit, *next;
	struct libparser_unit **head;
	unsigned char c;
	size_t i;

	if (!ctx->cache) {
		unit = ecalloc(1, sizeof(*unit));
	} else {
		unit = ctx->cache;
		ctx->cache = unit->next;
		unit->in = unit->next = NULL;
	}

	unit->rule = rule;
	unit->start = ctx->position;

	switch (sentence->type) {
	case LIBPARSER_SENTENCE_TYPE_CONCATENATION:
		unit->in = try_match(NULL, sentence->binary.left, ctx);
		if (!unit->in)
			goto mismatch;
		if (ctx->done)
			break;
		unit->in->next = try_match(NULL, sentence->binary.right, ctx);
		if (!unit->in->next) {
			free_unit(unit->in, ctx);
			goto mismatch;
		}
		if (!unit->in->next->rule || unit->in->next->rule[0] == '_') {
			unit->in->next->next = ctx->cache;
			ctx->cache = unit->in->next;
			unit->in->next = unit->in->next->in;
		}
		if (!unit->in->rule || unit->in->rule[0] == '_') {
			next = unit->in->next;
			unit->in->next = ctx->cache;
			ctx->cache = unit->in;
			unit->in = unit->in->in;
			if (unit->in) {
				for (head = &unit->in->next; *head; head = &(*head)->next);
				*head = next;
			} else {
				unit->in = next;
			}
		}
		break;

	case LIBPARSER_SENTENCE_TYPE_ALTERNATION:
		unit->in = try_match(NULL, sentence->binary.left, ctx);
		if (!unit->in) {
			unit->in = try_match(NULL, sentence->binary.right, ctx);
			if (!unit->in)
				goto mismatch;
		}
	prone:
		if (unit->in && (!unit->in->rule || unit->in->rule[0] == '_')) {
			unit->in->next = ctx->cache;
			ctx->cache = unit->in;
			unit->in = unit->in->in;
		}
		break;

	case LIBPARSER_SENTENCE_TYPE_OPTIONAL:
		unit->in = try_match(NULL, sentence->unary.sentence, ctx);
		goto prone;

	case LIBPARSER_SENTENCE_TYPE_REPEATED:
		head = &unit->in;
		while ((*head = try_match(NULL, sentence->unary.sentence, ctx))) {
			if (!(*head)->rule || (*head)->rule[0] == '_') {
				(*head)->next = ctx->cache;
				ctx->cache = *head;
				*head = (*head)->in;
				while (*head)
					head = &(*head)->next;
			} else {
				head = &(*head)->next;
			}
			if (ctx->done)
				break;
		}
		break;

	case LIBPARSER_SENTENCE_TYPE_STRING:
		if (sentence->string.length > ctx->length - ctx->position)
			goto mismatch;
		if (memcmp(&ctx->data[ctx->position], sentence->string.string, sentence->string.length))
			goto mismatch;
		ctx->position += sentence->string.length;
		break;

	case LIBPARSER_SENTENCE_TYPE_CHAR_RANGE:
		if (ctx->position == ctx->length)
			goto mismatch;
		c = ((const unsigned char *)ctx->data)[ctx->position];
		if (sentence->char_range.low > c || c > sentence->char_range.high)
			goto mismatch;
		ctx->position += 1;
		break;

	case LIBPARSER_SENTENCE_TYPE_RULE:
		for (i = 0; ctx->rules[i]; i++)
			if (!strcmp(ctx->rules[i]->name, sentence->rule.rule))
				break;
		if (!ctx->rules[i])
			abort();
		unit->in = try_match(ctx->rules[i]->name, ctx->rules[i]->sentence, ctx);
		if (!unit->in)
			goto mismatch;
		goto prone;

	case LIBPARSER_SENTENCE_TYPE_EXCEPTION:
		ctx->done = 1;
		ctx->exception = 1;
		break;

	case LIBPARSER_SENTENCE_TYPE_EOF:
		if (ctx->position != ctx->length)
			goto mismatch;
		ctx->done = 1;
		break;

	default:
		abort();
	}

	unit->end = ctx->position;
	return unit;

mismatch:
	ctx->position = unit->start;
	unit->next = ctx->cache;
	ctx->cache = unit;
	return NULL;
}


struct libparser_unit *
libparser_parse_file(const struct libparser_rule *const rules[], const char *data, size_t length, int *exceptionp)
{
	struct libparser_unit *ret, *t;
	struct context ctx;
	size_t i;

	ctx.rules = rules;
	ctx.cache = NULL;
	ctx.data = data;
	ctx.length = length;
	ctx.position = 0;
	ctx.done = 0;
	ctx.exception = 0;

	for (i = 0; rules[i]; i++)
		if (!strcmp(rules[i]->name, "@start"))
			break;
	if (!rules[i])
		abort();

	ret = try_match(rules[i]->name, rules[i]->sentence, &ctx);
	*exceptionp = ctx.exception;

	while (ctx.cache) {
		t = ctx.cache;
		ctx.cache = t->next;
		free(t);
	}

	return ret;
}