1 /*- 2 * Copyright (c) 2003-2010 Tim Kientzle 3 * All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR 15 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 16 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 17 * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT, 18 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 19 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 20 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 21 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 22 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 23 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 24 */ 25 26 #ifndef ARCHIVE_STRING_H_INCLUDED 27 #define ARCHIVE_STRING_H_INCLUDED 28 29 #ifndef __LIBARCHIVE_BUILD 30 #ifndef __LIBARCHIVE_TEST 31 #error This header is only to be used internally to libarchive. 32 #endif 33 #endif 34 35 #include <stdarg.h> 36 #ifdef HAVE_STDLIB_H 37 #include <stdlib.h> /* required for wchar_t on some systems */ 38 #endif 39 #ifdef HAVE_STRING_H 40 #include <string.h> 41 #endif 42 #ifdef HAVE_WCHAR_H 43 #include <wchar.h> 44 #endif 45 46 #include "archive.h" 47 48 /* 49 * Basic resizable/reusable string support similar to Java's "StringBuffer." 50 * 51 * Unlike sbuf(9), the buffers here are fully reusable and track the 52 * length throughout. 53 */ 54 55 struct archive_string { 56 char *s; /* Pointer to the storage */ 57 size_t length; /* Length of 's' in characters */ 58 size_t buffer_length; /* Length of malloc-ed storage in bytes. */ 59 }; 60 61 struct archive_wstring { 62 wchar_t *s; /* Pointer to the storage */ 63 size_t length; /* Length of 's' in characters */ 64 size_t buffer_length; /* Length of malloc-ed storage in bytes. */ 65 }; 66 67 struct archive_string_conv; 68 69 /* Initialize an archive_string object on the stack or elsewhere. */ 70 #define archive_string_init(a) \ 71 do { (a)->s = NULL; (a)->length = 0; (a)->buffer_length = 0; } while(0) 72 73 /* Append a C char to an archive_string, resizing as necessary. */ 74 struct archive_string * 75 archive_strappend_char(struct archive_string *, char); 76 77 /* Ditto for a wchar_t and an archive_wstring. */ 78 struct archive_wstring * 79 archive_wstrappend_wchar(struct archive_wstring *, wchar_t); 80 81 /* Append a raw array to an archive_string, resizing as necessary */ 82 struct archive_string * 83 archive_array_append(struct archive_string *, const char *, size_t); 84 85 /* Convert a Unicode string to current locale and append the result. */ 86 /* Returns -1 if conversion fails. */ 87 int 88 archive_string_append_from_wcs(struct archive_string *, const wchar_t *, size_t); 89 90 91 /* Create a string conversion object. 92 * Return NULL and set a error message if the conversion is not supported 93 * on the platform. */ 94 struct archive_string_conv * 95 archive_string_conversion_to_charset(struct archive *, const char *, int); 96 struct archive_string_conv * 97 archive_string_conversion_from_charset(struct archive *, const char *, int); 98 /* Create the default string conversion object for reading/writing an archive. 99 * Return NULL if the conversion is unneeded. 100 * Note: On non Windows platform this always returns NULL. 101 */ 102 struct archive_string_conv * 103 archive_string_default_conversion_for_read(struct archive *); 104 struct archive_string_conv * 105 archive_string_default_conversion_for_write(struct archive *); 106 /* Dispose of a string conversion object. */ 107 void 108 archive_string_conversion_free(struct archive *); 109 const char * 110 archive_string_conversion_charset_name(struct archive_string_conv *); 111 void 112 archive_string_conversion_set_opt(struct archive_string_conv *, int); 113 #define SCONV_SET_OPT_UTF8_LIBARCHIVE2X 1 114 #define SCONV_SET_OPT_NORMALIZATION_C 2 115 #define SCONV_SET_OPT_NORMALIZATION_D 4 116 117 118 /* Copy one archive_string to another in locale conversion. 119 * Return -1 if conversion fails. */ 120 int 121 archive_strncpy_l(struct archive_string *, const void *, size_t, 122 struct archive_string_conv *); 123 124 /* Copy one archive_string to another in locale conversion. 125 * Return -1 if conversion fails. */ 126 int 127 archive_strncat_l(struct archive_string *, const void *, size_t, 128 struct archive_string_conv *); 129 130 131 /* Copy one archive_string to another */ 132 #define archive_string_copy(dest, src) \ 133 ((dest)->length = 0, archive_string_concat((dest), (src))) 134 #define archive_wstring_copy(dest, src) \ 135 ((dest)->length = 0, archive_wstring_concat((dest), (src))) 136 137 /* Concatenate one archive_string to another */ 138 void archive_string_concat(struct archive_string *dest, struct archive_string *src); 139 void archive_wstring_concat(struct archive_wstring *dest, struct archive_wstring *src); 140 141 /* Ensure that the underlying buffer is at least as large as the request. */ 142 struct archive_string * 143 archive_string_ensure(struct archive_string *, size_t); 144 struct archive_wstring * 145 archive_wstring_ensure(struct archive_wstring *, size_t); 146 147 /* Append C string, which may lack trailing \0. */ 148 /* The source is declared void * here because this gets used with 149 * "signed char *", "unsigned char *" and "char *" arguments. 150 * Declaring it "char *" as with some of the other functions just 151 * leads to a lot of extra casts. */ 152 struct archive_string * 153 archive_strncat(struct archive_string *, const void *, size_t); 154 struct archive_wstring * 155 archive_wstrncat(struct archive_wstring *, const wchar_t *, size_t); 156 157 /* Append a C string to an archive_string, resizing as necessary. */ 158 struct archive_string * 159 archive_strcat(struct archive_string *, const void *); 160 struct archive_wstring * 161 archive_wstrcat(struct archive_wstring *, const wchar_t *); 162 163 /* Copy a C string to an archive_string, resizing as necessary. */ 164 #define archive_strcpy(as,p) \ 165 archive_strncpy((as), (p), ((p) == NULL ? 0 : strlen(p))) 166 #define archive_wstrcpy(as,p) \ 167 archive_wstrncpy((as), (p), ((p) == NULL ? 0 : wcslen(p))) 168 #define archive_strcpy_l(as,p,lo) \ 169 archive_strncpy_l((as), (p), ((p) == NULL ? 0 : strlen(p)), (lo)) 170 171 /* Copy a C string to an archive_string with limit, resizing as necessary. */ 172 #define archive_strncpy(as,p,l) \ 173 ((as)->length=0, archive_strncat((as), (p), (l))) 174 #define archive_wstrncpy(as,p,l) \ 175 ((as)->length = 0, archive_wstrncat((as), (p), (l))) 176 177 /* Return length of string. */ 178 #define archive_strlen(a) ((a)->length) 179 180 /* Set string length to zero. */ 181 #define archive_string_empty(a) ((a)->length = 0) 182 #define archive_wstring_empty(a) ((a)->length = 0) 183 184 /* Release any allocated storage resources. */ 185 void archive_string_free(struct archive_string *); 186 void archive_wstring_free(struct archive_wstring *); 187 188 /* Like 'vsprintf', but resizes the underlying string as necessary. */ 189 /* Note: This only implements a small subset of standard printf functionality. */ 190 void archive_string_vsprintf(struct archive_string *, const char *, 191 va_list) __LA_PRINTF(2, 0); 192 void archive_string_sprintf(struct archive_string *, const char *, ...) 193 __LA_PRINTF(2, 3); 194 195 /* Translates from MBS to Unicode. */ 196 /* Returns non-zero if conversion failed in any way. */ 197 int archive_wstring_append_from_mbs(struct archive_wstring *dest, 198 const char *, size_t); 199 200 201 /* A "multistring" can hold Unicode, UTF8, or MBS versions of 202 * the string. If you set and read the same version, no translation 203 * is done. If you set and read different versions, the library 204 * will attempt to transparently convert. 205 */ 206 struct archive_mstring { 207 struct archive_string aes_mbs; 208 struct archive_string aes_utf8; 209 struct archive_wstring aes_wcs; 210 struct archive_string aes_mbs_in_locale; 211 /* Bitmap of which of the above are valid. Because we're lazy 212 * about malloc-ing and reusing the underlying storage, we 213 * can't rely on NULL pointers to indicate whether a string 214 * has been set. */ 215 int aes_set; 216 #define AES_SET_MBS 1 217 #define AES_SET_UTF8 2 218 #define AES_SET_WCS 4 219 }; 220 221 void archive_mstring_clean(struct archive_mstring *); 222 void archive_mstring_copy(struct archive_mstring *dest, struct archive_mstring *src); 223 int archive_mstring_get_mbs(struct archive *, struct archive_mstring *, const char **); 224 int archive_mstring_get_utf8(struct archive *, struct archive_mstring *, const char **); 225 int archive_mstring_get_wcs(struct archive *, struct archive_mstring *, const wchar_t **); 226 int archive_mstring_get_mbs_l(struct archive *, struct archive_mstring *, const char **, 227 size_t *, struct archive_string_conv *); 228 int archive_mstring_copy_mbs(struct archive_mstring *, const char *mbs); 229 int archive_mstring_copy_mbs_len(struct archive_mstring *, const char *mbs, 230 size_t); 231 int archive_mstring_copy_utf8(struct archive_mstring *, const char *utf8); 232 int archive_mstring_copy_wcs(struct archive_mstring *, const wchar_t *wcs); 233 int archive_mstring_copy_wcs_len(struct archive_mstring *, 234 const wchar_t *wcs, size_t); 235 int archive_mstring_copy_mbs_len_l(struct archive_mstring *, 236 const char *mbs, size_t, struct archive_string_conv *); 237 int archive_mstring_update_utf8(struct archive *, struct archive_mstring *aes, const char *utf8); 238 239 240 #endif 241