2 * Copyright (C) 2013-2014 Jo-Philipp Wich <jow@openwrt.org>
4 * Permission to use, copy, modify, and/or distribute this software for any
5 * purpose with or without fee is hereby granted, provided that the above
6 * copyright notice and this permission notice appear in all copies.
8 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9 * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10 * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11 * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14 * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
31 int (*parse)(const char *buf, struct jp_opcode *op);
38 (((x) >= 'a') ? (10 + (x) - 'a') : \
39 (((x) >= 'A') ? (10 + (x) - 'A') : dec(x)))
42 * Stores the given codepoint as a utf8 multibyte sequence into the given
43 * output buffer and substracts the required amount of bytes from the given
46 * Returns false if the multibyte sequence would not fit into the buffer,
51 utf8enc(char **out, int *rem, int code)
53 if (code > 0 && code <= 0x7F)
58 *(*out++) = code; (*rem)--;
61 else if (code > 0 && code <= 0x7FF)
66 *(*out)++ = ((code >> 6) & 0x1F) | 0xC0; (*rem)--;
67 *(*out)++ = ( code & 0x3F) | 0x80; (*rem)--;
70 else if (code > 0 && code <= 0xFFFF)
75 *(*out)++ = ((code >> 12) & 0x0F) | 0xE0; (*rem)--;
76 *(*out)++ = ((code >> 6) & 0x3F) | 0x80; (*rem)--;
77 *(*out)++ = ( code & 0x3F) | 0x80; (*rem)--;
80 else if (code > 0 && code <= 0x10FFFF)
85 *(*out)++ = ((code >> 18) & 0x07) | 0xF0; (*rem)--;
86 *(*out)++ = ((code >> 12) & 0x3F) | 0x80; (*rem)--;
87 *(*out)++ = ((code >> 6) & 0x3F) | 0x80; (*rem)--;
88 *(*out)++ = ( code & 0x3F) | 0x80; (*rem)--;
97 * Parses a string literal from the given buffer.
99 * Returns a negative value on error, otherwise the amount of consumed
100 * characters from the given buffer.
103 * -1 Unterminated string
104 * -2 Invalid escape sequence
105 * -3 String literal too long
109 parse_string(const char *buf, struct jp_opcode *op)
112 char str[128] = { 0 };
114 const char *in = buf;
116 int rem = sizeof(str) - 1;
121 /* continuation of escape sequence */
127 if (isxdigit(in[1]) && isxdigit(in[2]) &&
128 isxdigit(in[3]) && isxdigit(in[4]))
130 if (!utf8enc(&out, &rem,
131 hex(in[1]) * 16 * 16 * 16 +
132 hex(in[2]) * 16 * 16 +
146 else if (in[0] == 'x')
148 if (isxdigit(in[1]) && isxdigit(in[2]))
150 if (!utf8enc(&out, &rem, hex(in[1]) * 16 + hex(in[2])))
161 /* \377, \77 or \7 */
162 else if (in[0] >= '0' && in[0] <= '7')
165 if (in[1] >= '0' && in[1] <= '7' &&
166 in[2] >= '0' && in[2] <= '7')
168 code = dec(in[0]) * 8 * 8 +
175 if (!utf8enc(&out, &rem, code))
182 else if (in[1] >= '0' && in[1] <= '7')
184 if (!utf8enc(&out, &rem, dec(in[0]) * 8 + dec(in[1])))
193 if (!utf8enc(&out, &rem, dec(in[0])))
200 /* single character escape */
208 case 'a': *out = '\a'; break;
209 case 'b': *out = '\b'; break;
210 case 'e': *out = '\e'; break;
211 case 'f': *out = '\f'; break;
212 case 'n': *out = '\n'; break;
213 case 'r': *out = '\r'; break;
214 case 't': *out = '\t'; break;
215 case 'v': *out = '\v'; break;
216 default: *out = *in; break;
226 /* begin of escape sequence */
227 else if (*in == '\\')
233 /* terminating quote */
236 op->str = strdup(str);
237 return (in - buf) + 2;
255 * Parses a label from the given buffer.
257 * Returns a negative value on error, otherwise the amount of consumed
258 * characters from the given buffer.
265 parse_label(const char *buf, struct jp_opcode *op)
267 char str[128] = { 0 };
269 const char *in = buf;
270 int rem = sizeof(str) - 1;
272 while (*in == '_' || isalnum(*in))
280 if (!strcmp(str, "true") || !strcmp(str, "false"))
282 op->num = (str[0] == 't');
287 op->str = strdup(str);
295 * Parses a number literal from the given buffer.
297 * Returns a negative value on error, otherwise the amount of consumed
298 * characters from the given buffer.
301 * -2 Invalid number character
305 parse_number(const char *buf, struct jp_opcode *op)
308 int n = strtol(buf, &e, 10);
318 static const struct token tokens[] = {
328 { T_BROPEN, "[", 1 },
329 { T_BRCLOSE, "]", 1 },
331 { T_PCLOSE, ")", 1 },
339 { T_WILDCARD, "*", 1 },
340 { T_STRING, "'", 1, parse_string },
341 { T_STRING, "\"", 1, parse_string },
342 { T_LABEL, "_", 1, parse_label },
343 { T_LABEL, "az", 0, parse_label },
344 { T_LABEL, "AZ", 0, parse_label },
345 { T_NUMBER, "-", 1, parse_number },
346 { T_NUMBER, "09", 0, parse_number },
349 const char *tokennames[23] = {
365 [T_WILDCARD] = "'*'",
369 [T_NUMBER] = "Number",
370 [T_STRING] = "String",
377 match_token(const char *ptr, struct jp_opcode *op)
380 const struct token *tok;
382 for (i = 0, tok = &tokens[0];
383 i < sizeof(tokens) / sizeof(tokens[0]);
384 i++, tok = &tokens[i])
386 if ((tok->plen > 0 && !strncmp(ptr, tok->pat, tok->plen)) ||
387 (tok->plen == 0 && *ptr >= tok->pat[0] && *ptr <= tok->pat[1]))
389 op->type = tok->type;
392 return tok->parse(ptr, op);
402 jp_get_token(struct jp_state *s, const char *input, int *mlen)
404 struct jp_opcode op = { 0 };
406 *mlen = match_token(input, &op);
408 if (*mlen < 0 || op.type == 0)
411 return jp_alloc_op(s, op.type, op.num, op.str, NULL);