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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407
|
/* flint_utils.h: Generic functions for flint
*
* Copyright 1999,2000,2001 BrightStation PLC
* Copyright 2002 Ananova Ltd
* Copyright 2002,2003,2004,2006,2008,2009 Olly Betts
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation; either version 2 of the
* License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301
* USA
*/
#ifndef OM_HGUARD_FLINT_UTILS_H
#define OM_HGUARD_FLINT_UTILS_H
#include "omassert.h"
#include <xapian/types.h>
#include <string>
using namespace std;
typedef unsigned char om_byte;
typedef unsigned int om_uint32;
typedef int om_int32;
/** FIXME: the pack and unpack int methods store in low-byte-first order
* - it might be easier to implement efficient specialisations with
* high-byte-first order.
*/
/** Reads an unsigned integer from a string starting at a given position.
*
* @param src A pointer to a pointer to the data to read. The
* character pointer will be updated to point to the
* next character to read, or 0 if the method ran out of
* data. (It is only set to 0 in case of an error).
* @param src_end A pointer to the byte after the end of the data to
* read the integer from.
* @param resultptr A pointer to a place to store the result. If an
* error occurs, the value stored in this location is
* undefined. If this pointer is 0, the result is not
* stored, and the method simply skips over the result.
*
* @result True if an integer was successfully read. False if the read
* failed. Failure may either be due to the data running out (in
* which case *src will equal 0), or due to the value read
* overflowing the size of result (in which case *src will point
* to wherever the value ends, despite the overflow).
*/
template<class T>
bool
F_unpack_uint(const char ** src,
const char * src_end,
T * resultptr)
{
// Check unsigned
STATIC_ASSERT_UNSIGNED_TYPE(T);
// Check byte is what it's meant to be
STATIC_ASSERT(sizeof(om_byte) == 1);
unsigned int shift = 0;
T result = 0;
while (true) {
if ((*src) == src_end) {
*src = 0;
return false;
}
om_byte part = static_cast<om_byte>(**src);
(*src)++;
// if new byte might cause overflow, and it does
if (((shift > (sizeof(T) - 1) * 8 + 1) &&
((part & 0x7f) << (shift % 8)) >= 0x100) ||
(shift >= sizeof(T) * 8)) {
// Overflowed - move to end of this integer
while (true) {
if ((part & 0x80) == 0) return false;
if ((*src) == src_end) {
*src = 0;
return false;
}
part = static_cast<om_byte>(**src);
(*src)++;
}
}
result += T(part & 0x7f) << shift;
shift += 7;
if ((part & 0x80) == 0) {
if (resultptr) *resultptr = result;
return true;
}
}
}
/** Generates a packed representation of an integer.
*
* @param value The integer to represent.
*
* @result A string containing the representation of the integer.
*/
template<class T>
string
F_pack_uint(T value)
{
// Check unsigned
STATIC_ASSERT_UNSIGNED_TYPE(T);
if (value == 0) return string(1, '\0');
string result;
while (value != 0) {
om_byte part = static_cast<om_byte>(value & 0x7f);
value = value >> 7;
if (value) part |= 0x80;
result.append(1u, char(part));
}
return result;
}
/** Generates a packed representation of a bool.
*
* This is a specialisation of the template above.
*
* @param value The bool to represent.
*
* @result A string containing the representation of the bool.
*/
template<>
inline string
F_pack_uint<bool>(bool value)
{
return string(1, static_cast<char>(value));
}
/** Reads an unsigned integer from a string starting at a given position.
* This encoding requires that we know the encoded length from out-of-band
* information (so is suitable when only one integer is encoded, or for
* the last integer encoded).
*
* @param src A pointer to a pointer to the data to read.
* @param src_end A pointer to the byte after the end of the data to
* read the integer from.
* @param resultptr A pointer to a place to store the result. If an
* error occurs, the value stored in this location is
* undefined. If this pointer is 0, the result is not
* stored, and the method simply skips over the result.
*
* @result True if an integer was successfully read. False if the read
* failed. Failure can hapen if the value read overflows
* the size of result.
*/
template<class T>
bool
F_unpack_uint_last(const char ** src, const char * src_end, T * resultptr)
{
// Check unsigned
STATIC_ASSERT_UNSIGNED_TYPE(T);
// Check byte is what it's meant to be
STATIC_ASSERT(sizeof(om_byte) == 1);
if (src_end - *src > int(sizeof(T))) {
// Would overflow
*src = src_end;
return false;
}
T result = 0;
int shift = 0;
while (*src != src_end) {
result |= static_cast<T>(static_cast<om_byte>(**src)) << shift;
++(*src);
shift += 8;
}
*resultptr = result;
return true;
}
/** Generates a packed representation of an integer.
* This encoding requires that we know the encoded length from out-of-band
* information (so is suitable when only one integer is encoded, or for
* the last integer encoded).
*
* @param value The integer to represent.
*
* @result A string containing the representation of the integer.
*/
template<class T>
string
F_pack_uint_last(T value)
{
// Check unsigned
STATIC_ASSERT_UNSIGNED_TYPE(T);
string result;
while (value) {
result += char(value);
value >>= 8;
}
return result;
}
/** Generate a packed representation of an integer, preserving sort order.
*
* This representation is less compact than the usual one, and has a limit
* of 256 bytes on the length of the integer. However, this is unlikely to
* ever be a problem.
*
* @param value The integer to represent.
*
* @result A string containing the representation of the integer.
*/
template<class T>
string
F_pack_uint_preserving_sort(T value)
{
// Check unsigned
STATIC_ASSERT_UNSIGNED_TYPE(T);
string result;
while (value != 0) {
om_byte part = static_cast<om_byte>(value & 0xff);
value = value >> 8;
result.insert(string::size_type(0), 1u, char(part));
}
result.insert(string::size_type(0), 1u, char(result.size()));
return result;
}
/** Unpack a unsigned integer, store in sort preserving order.
*
* @param src A pointer to a pointer to the data to read. The
* character pointer will be updated to point to the
* next character to read, or 0 if the method ran out of
* data. (It is only set to 0 in case of an error).
* @param src_end A pointer to the byte after the end of the data to
* read the integer from.
* @param resultptr A pointer to a place to store the result. If an
* error occurs, the value stored in this location is
* undefined. If this pointer is 0, the result is not
* stored, and the method simply skips over the result.
*
* @result True if an integer was successfully read. False if the read
* failed. Failure may either be due to the data running out (in
* which case *src will equal 0), or due to the value read
* overflowing the size of result (in which case *src will point
* to wherever the value ends, despite the overflow).
*/
template<class T>
bool
F_unpack_uint_preserving_sort(const char ** src,
const char * src_end,
T * resultptr)
{
if (*src == src_end) {
*src = 0;
return false;
}
unsigned int length = static_cast<om_byte>(**src);
(*src)++;
if (length > sizeof(T)) {
*src += length;
if (*src > src_end) {
*src = 0;
}
return false;
}
// Can't be overflow now.
T result = 0;
while (length > 0) {
result = result << 8;
result += static_cast<om_byte>(**src);
(*src)++;
length--;
}
*resultptr = result;
return true;
}
inline bool
F_unpack_string(const char ** src,
const char * src_end,
string & result)
{
string::size_type length;
if (!F_unpack_uint(src, src_end, &length)) {
return false;
}
if (src_end - *src < 0 ||
string::size_type(src_end - *src) < length) {
src = 0;
return false;
}
result.assign(*src, length);
*src += length;
return true;
}
inline string
F_pack_string(const string & value)
{
return F_pack_uint(value.size()) + value;
}
/** Pack a string into a representation which preserves sort order.
* We do this by replacing zero bytes in the string with a zero byte
* followed by byte value 0xff, and then appending two zero bytes to
* the end.
*/
inline string
F_pack_string_preserving_sort(string value)
{
string::size_type i = 0, j;
while ((j = value.find('\0', i)) != string::npos) {
value.replace(j, 1, "\0\xff", 2);
i = j + 2;
}
value += '\0'; // FIXME temp...
return value + '\0'; // Note - next byte mustn't be '\xff'...
}
inline bool
F_unpack_string_preserving_sort(const char ** src,
const char * src_end,
string & result)
{
result.resize(0);
while (*src < src_end) {
const char *begin = *src;
while (**src) {
++(*src);
if (*src == src_end) return false;
}
result += string(begin, *src - begin);
++(*src);
if (*src == src_end) return false;
if (**src != '\xff') {
++(*src); // FIXME temp
return true;
}
result += '\0';
++(*src);
}
return false;
}
inline bool
F_unpack_bool(const char ** src,
const char * src_end,
bool * resultptr)
{
if (*src == src_end) {
*src = 0;
return false;
}
switch (*((*src)++)) {
case '0':
if (resultptr) *resultptr = false;
return true;
case '1':
if (resultptr) *resultptr = true;
return true;
}
*src = 0;
return false;
}
inline string
F_pack_bool(bool value)
{
return value ? "1" : "0";
}
/** Convert a document id to a key (suitable when the docid is the only
* component of the key).
*/
inline string
flint_docid_to_key(Xapian::docid did)
{
return F_pack_uint_preserving_sort(did);
}
#endif /* OM_HGUARD_FLINT_UTILS_H */
|