| 1 | /* |
| 2 | * This file is part of DisOrder. |
| 3 | * Copyright (C) 2004, 2006 Richard Kettlewell |
| 4 | * |
| 5 | * This program is free software; you can redistribute it and/or modify |
| 6 | * it under the terms of the GNU General Public License as published by |
| 7 | * the Free Software Foundation; either version 2 of the License, or |
| 8 | * (at your option) any later version. |
| 9 | * |
| 10 | * This program is distributed in the hope that it will be useful, but |
| 11 | * WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 13 | * General Public License for more details. |
| 14 | * |
| 15 | * You should have received a copy of the GNU General Public License |
| 16 | * along with this program; if not, write to the Free Software |
| 17 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 |
| 18 | * USA |
| 19 | */ |
| 20 | |
| 21 | #include <config.h> |
| 22 | #include "types.h" |
| 23 | |
| 24 | #include <ctype.h> |
| 25 | #include <string.h> |
| 26 | #include <errno.h> |
| 27 | |
| 28 | #include "mem.h" |
| 29 | #include "split.h" |
| 30 | #include "log.h" |
| 31 | #include "charset.h" |
| 32 | #include "vector.h" |
| 33 | |
| 34 | static inline int space(int c) { |
| 35 | return (c == ' ' |
| 36 | || c == '\t' |
| 37 | || c == '\n' |
| 38 | || c == '\r'); |
| 39 | } |
| 40 | |
| 41 | static void no_error_handler(const char attribute((unused)) *msg, |
| 42 | void attribute((unused)) *u) { |
| 43 | } |
| 44 | |
| 45 | /* TODO: handle combining characters attached to delimiters in some |
| 46 | * sane way (might include reporting an error) */ |
| 47 | |
| 48 | char **split(const char *p, |
| 49 | int *np, |
| 50 | unsigned flags, |
| 51 | void (*error_handler)(const char *msg, void *u), |
| 52 | void *u) { |
| 53 | char *f, *g; |
| 54 | const char *q; |
| 55 | struct vector v; |
| 56 | size_t l; |
| 57 | int qc; |
| 58 | |
| 59 | if(!error_handler) error_handler = no_error_handler; |
| 60 | vector_init(&v); |
| 61 | while(*p && !(*p == '#' && (flags & SPLIT_COMMENTS))) { |
| 62 | if(space(*p)) { |
| 63 | ++p; |
| 64 | continue; |
| 65 | } |
| 66 | if((flags & SPLIT_QUOTES) && (*p == '"' || *p == '\'')) { |
| 67 | qc = *p++; |
| 68 | l = 0; |
| 69 | for(q = p; *q && *q != qc; ++q) { |
| 70 | if(*q == '\\' && q[1]) |
| 71 | ++q; |
| 72 | ++l; |
| 73 | } |
| 74 | if(!*q) { |
| 75 | error_handler("unterminated quoted string", u); |
| 76 | return 0; |
| 77 | } |
| 78 | f = g = xmalloc_noptr(l + 1); |
| 79 | for(q = p; *q != qc;) { |
| 80 | if(*q == '\\') { |
| 81 | ++q; |
| 82 | switch(*q) { |
| 83 | case '\\': |
| 84 | case '"': |
| 85 | case '\'': |
| 86 | *g++ = *q++; |
| 87 | break; |
| 88 | case 'n': |
| 89 | ++q; |
| 90 | *g++ = '\n'; |
| 91 | break; |
| 92 | default: |
| 93 | error_handler("illegal escape sequence", u); |
| 94 | return 0; |
| 95 | } |
| 96 | } else |
| 97 | *g++ = *q++; |
| 98 | } |
| 99 | *g = 0; |
| 100 | p = q + 1; |
| 101 | } else { |
| 102 | for(q = p; *q && !space(*q); ++q) |
| 103 | ; |
| 104 | l = q - p; |
| 105 | f = xstrndup(p, l); |
| 106 | p = q; |
| 107 | } |
| 108 | vector_append(&v, f); |
| 109 | } |
| 110 | vector_terminate(&v); |
| 111 | if(np) |
| 112 | *np = v.nvec; |
| 113 | return v.vec; |
| 114 | } |
| 115 | |
| 116 | /* TODO handle initial combining characters sanely */ |
| 117 | |
| 118 | const char *quoteutf8(const char *s) { |
| 119 | size_t len = 3 + strlen(s); |
| 120 | const char *t; |
| 121 | char *r, *q; |
| 122 | |
| 123 | /* see if we need to quote */ |
| 124 | if(*s) { |
| 125 | for(t = s; *t; t++) |
| 126 | if((unsigned char)*t <= ' ' |
| 127 | || *t == '"' |
| 128 | || *t == '\\' |
| 129 | || *t == '\'' |
| 130 | || *t == '#') |
| 131 | break; |
| 132 | if(!*t) |
| 133 | return s; |
| 134 | } |
| 135 | |
| 136 | /* we rely on ASCII characters only ever representing themselves in UTF-8. */ |
| 137 | for(t = s; *t; t++) { |
| 138 | switch(*t) { |
| 139 | case '"': |
| 140 | case '\\': |
| 141 | case '\n': |
| 142 | ++len; |
| 143 | break; |
| 144 | } |
| 145 | } |
| 146 | q = r = xmalloc_noptr(len); |
| 147 | *q++ = '"'; |
| 148 | for(t = s; *t; t++) { |
| 149 | switch(*t) { |
| 150 | case '"': |
| 151 | case '\\': |
| 152 | *q++ = '\\'; |
| 153 | /* fall through */ |
| 154 | default: |
| 155 | *q++ = *t; |
| 156 | break; |
| 157 | case '\n': |
| 158 | *q++ = '\\'; |
| 159 | *q++ = 'n'; |
| 160 | break; |
| 161 | } |
| 162 | } |
| 163 | *q++ = '"'; |
| 164 | *q = 0; |
| 165 | return r; |
| 166 | } |
| 167 | |
| 168 | /* |
| 169 | Local Variables: |
| 170 | c-basic-offset:2 |
| 171 | comment-column:40 |
| 172 | End: |
| 173 | */ |