| 1 | // |
| 2 | // DateTimeParser.cpp |
| 3 | // |
| 4 | // Library: Foundation |
| 5 | // Package: DateTime |
| 6 | // Module: DateTimeParser |
| 7 | // |
| 8 | // Copyright (c) 2004-2006, Applied Informatics Software Engineering GmbH. |
| 9 | // and Contributors. |
| 10 | // |
| 11 | // SPDX-License-Identifier: BSL-1.0 |
| 12 | // |
| 13 | |
| 14 | |
| 15 | #include "Poco/DateTimeParser.h" |
| 16 | #include "Poco/DateTimeFormat.h" |
| 17 | #include "Poco/DateTime.h" |
| 18 | #include "Poco/Exception.h" |
| 19 | #include "Poco/Ascii.h" |
| 20 | |
| 21 | |
| 22 | namespace Poco { |
| 23 | |
| 24 | |
| 25 | #define SKIP_JUNK() \ |
| 26 | while (it != end && !Ascii::isDigit(*it)) ++it |
| 27 | |
| 28 | |
| 29 | #define SKIP_DIGITS() \ |
| 30 | while (it != end && Ascii::isDigit(*it)) ++it |
| 31 | |
| 32 | |
| 33 | #define PARSE_NUMBER(var) \ |
| 34 | while (it != end && Ascii::isDigit(*it)) var = var*10 + ((*it++) - '0') |
| 35 | |
| 36 | |
| 37 | #define PARSE_NUMBER_N(var, n) \ |
| 38 | { int i = 0; while (i++ < n && it != end && Ascii::isDigit(*it)) var = var*10 + ((*it++) - '0'); } |
| 39 | |
| 40 | |
| 41 | #define PARSE_FRACTIONAL_N(var, n) \ |
| 42 | { int i = 0; while (i < n && it != end && Ascii::isDigit(*it)) { var = var*10 + ((*it++) - '0'); i++; } while (i++ < n) var *= 10; } |
| 43 | |
| 44 | |
| 45 | void DateTimeParser::parse(const std::string& fmt, |
| 46 | const std::string& str, |
| 47 | DateTime& dateTime, |
| 48 | int& timeZoneDifferential) |
| 49 | { |
| 50 | if (fmt.empty() || str.empty() || (DateTimeFormat::hasFormat(fmt) && !DateTimeFormat::isValid(str))) |
| 51 | throw SyntaxException("Invalid DateTimeString:" + str); |
| 52 | |
| 53 | int year = 0; |
| 54 | int month = 0; |
| 55 | int day = 0; |
| 56 | int hour = 0; |
| 57 | int minute = 0; |
| 58 | int second = 0; |
| 59 | int millis = 0; |
| 60 | int micros = 0; |
| 61 | int tzd = 0; |
| 62 | |
| 63 | std::string::const_iterator it = str.begin(); |
| 64 | std::string::const_iterator end = str.end(); |
| 65 | std::string::const_iterator itf = fmt.begin(); |
| 66 | std::string::const_iterator endf = fmt.end(); |
| 67 | |
| 68 | while (itf != endf && it != end) |
| 69 | { |
| 70 | if (*itf == '%') |
| 71 | { |
| 72 | if (++itf != endf) |
| 73 | { |
| 74 | switch (*itf) |
| 75 | { |
| 76 | case 'w': |
| 77 | case 'W': |
| 78 | while (it != end && Ascii::isSpace(*it)) ++it; |
| 79 | while (it != end && Ascii::isAlpha(*it)) ++it; |
| 80 | break; |
| 81 | case 'b': |
| 82 | case 'B': |
| 83 | month = parseMonth(it, end); |
| 84 | break; |
| 85 | case 'd': |
| 86 | case 'e': |
| 87 | case 'f': |
| 88 | SKIP_JUNK(); |
| 89 | PARSE_NUMBER_N(day, 2); |
| 90 | break; |
| 91 | case 'm': |
| 92 | case 'n': |
| 93 | case 'o': |
| 94 | SKIP_JUNK(); |
| 95 | PARSE_NUMBER_N(month, 2); |
| 96 | break; |
| 97 | case 'y': |
| 98 | SKIP_JUNK(); |
| 99 | PARSE_NUMBER_N(year, 2); |
| 100 | if (year >= 69) |
| 101 | year += 1900; |
| 102 | else |
| 103 | year += 2000; |
| 104 | break; |
| 105 | case 'Y': |
| 106 | SKIP_JUNK(); |
| 107 | PARSE_NUMBER_N(year, 4); |
| 108 | break; |
| 109 | case 'r': |
| 110 | SKIP_JUNK(); |
| 111 | PARSE_NUMBER(year); |
| 112 | if (year < 1000) |
| 113 | { |
| 114 | if (year >= 69) |
| 115 | year += 1900; |
| 116 | else |
| 117 | year += 2000; |
| 118 | } |
| 119 | break; |
| 120 | case 'H': |
| 121 | case 'h': |
| 122 | SKIP_JUNK(); |
| 123 | PARSE_NUMBER_N(hour, 2); |
| 124 | break; |
| 125 | case 'a': |
| 126 | case 'A': |
| 127 | hour = parseAMPM(it, end, hour); |
| 128 | break; |
| 129 | case 'M': |
| 130 | SKIP_JUNK(); |
| 131 | PARSE_NUMBER_N(minute, 2); |
| 132 | break; |
| 133 | case 'S': |
| 134 | SKIP_JUNK(); |
| 135 | PARSE_NUMBER_N(second, 2); |
| 136 | break; |
| 137 | case 's': |
| 138 | SKIP_JUNK(); |
| 139 | PARSE_NUMBER_N(second, 2); |
| 140 | if (it != end && (*it == '.' || *it == ',')) |
| 141 | { |
| 142 | ++it; |
| 143 | PARSE_FRACTIONAL_N(millis, 3); |
| 144 | PARSE_FRACTIONAL_N(micros, 3); |
| 145 | SKIP_DIGITS(); |
| 146 | } |
| 147 | break; |
| 148 | case 'i': |
| 149 | SKIP_JUNK(); |
| 150 | PARSE_NUMBER_N(millis, 3); |
| 151 | break; |
| 152 | case 'c': |
| 153 | SKIP_JUNK(); |
| 154 | PARSE_NUMBER_N(millis, 1); |
| 155 | millis *= 100; |
| 156 | break; |
| 157 | case 'F': |
| 158 | SKIP_JUNK(); |
| 159 | PARSE_FRACTIONAL_N(millis, 3); |
| 160 | PARSE_FRACTIONAL_N(micros, 3); |
| 161 | SKIP_DIGITS(); |
| 162 | break; |
| 163 | case 'z': |
| 164 | case 'Z': |
| 165 | tzd = parseTZD(it, end); |
| 166 | break; |
| 167 | } |
| 168 | ++itf; |
| 169 | } |
| 170 | } |
| 171 | else ++itf; |
| 172 | } |
| 173 | if (month == 0) month = 1; |
| 174 | if (day == 0) day = 1; |
| 175 | if (DateTime::isValid(year, month, day, hour, minute, second, millis, micros)) |
| 176 | dateTime.assign(year, month, day, hour, minute, second, millis, micros); |
| 177 | else |
| 178 | throw SyntaxException("date/time component out of range" ); |
| 179 | timeZoneDifferential = tzd; |
| 180 | } |
| 181 | |
| 182 | |
| 183 | DateTime DateTimeParser::parse(const std::string& fmt, const std::string& str, int& timeZoneDifferential) |
| 184 | { |
| 185 | DateTime result; |
| 186 | parse(fmt, str, result, timeZoneDifferential); |
| 187 | return result; |
| 188 | } |
| 189 | |
| 190 | |
| 191 | bool DateTimeParser::tryParse(const std::string& fmt, |
| 192 | const std::string& str, |
| 193 | DateTime& dateTime, |
| 194 | int& timeZoneDifferential) |
| 195 | { |
| 196 | try |
| 197 | { |
| 198 | parse(fmt, str, dateTime, timeZoneDifferential); |
| 199 | } |
| 200 | catch (Exception&) |
| 201 | { |
| 202 | return false; |
| 203 | } |
| 204 | return true; |
| 205 | } |
| 206 | |
| 207 | |
| 208 | void DateTimeParser::parse(const std::string& str, DateTime& dateTime, int& timeZoneDifferential) |
| 209 | { |
| 210 | if (!tryParse(str, dateTime, timeZoneDifferential)) |
| 211 | throw SyntaxException("Unsupported or invalid date/time format" ); |
| 212 | } |
| 213 | |
| 214 | |
| 215 | DateTime DateTimeParser::parse(const std::string& str, int& timeZoneDifferential) |
| 216 | { |
| 217 | DateTime result; |
| 218 | if (tryParse(str, result, timeZoneDifferential)) |
| 219 | return result; |
| 220 | else |
| 221 | throw SyntaxException("Unsupported or invalid date/time format" ); |
| 222 | } |
| 223 | |
| 224 | |
| 225 | bool DateTimeParser::tryParse(const std::string& str, DateTime& dateTime, int& timeZoneDifferential) |
| 226 | { |
| 227 | if (str.length() < 4) return false; |
| 228 | |
| 229 | if (str[3] == ',') |
| 230 | return tryParse("%w, %e %b %r %H:%M:%S %Z" , str, dateTime, timeZoneDifferential); |
| 231 | else if (str[3] == ' ') |
| 232 | return tryParse(DateTimeFormat::ASCTIME_FORMAT, str, dateTime, timeZoneDifferential); |
| 233 | else if (str.find(',') < 10) |
| 234 | return tryParse("%W, %e %b %r %H:%M:%S %Z" , str, dateTime, timeZoneDifferential); |
| 235 | else if (Ascii::isDigit(str[0])) |
| 236 | { |
| 237 | if (str.find(' ') != std::string::npos || str.length() == 10) |
| 238 | return tryParse(DateTimeFormat::SORTABLE_FORMAT, str, dateTime, timeZoneDifferential); |
| 239 | else if (str.find('.') != std::string::npos || str.find(',') != std::string::npos) |
| 240 | return tryParse(DateTimeFormat::ISO8601_FRAC_FORMAT, str, dateTime, timeZoneDifferential); |
| 241 | else |
| 242 | return tryParse(DateTimeFormat::ISO8601_FORMAT, str, dateTime, timeZoneDifferential); |
| 243 | } |
| 244 | else return false; |
| 245 | } |
| 246 | |
| 247 | |
| 248 | int DateTimeParser::parseTZD(std::string::const_iterator& it, const std::string::const_iterator& end) |
| 249 | { |
| 250 | struct Zone |
| 251 | { |
| 252 | const char* designator; |
| 253 | int timeZoneDifferential; |
| 254 | }; |
| 255 | |
| 256 | static Zone zones[] = |
| 257 | { |
| 258 | {"Z" , 0}, |
| 259 | {"UT" , 0}, |
| 260 | {"GMT" , 0}, |
| 261 | {"BST" , 1*3600}, |
| 262 | {"IST" , 1*3600}, |
| 263 | {"WET" , 0}, |
| 264 | {"WEST" , 1*3600}, |
| 265 | {"CET" , 1*3600}, |
| 266 | {"CEST" , 2*3600}, |
| 267 | {"EET" , 2*3600}, |
| 268 | {"EEST" , 3*3600}, |
| 269 | {"MSK" , 3*3600}, |
| 270 | {"MSD" , 4*3600}, |
| 271 | {"NST" , -3*3600-1800}, |
| 272 | {"NDT" , -2*3600-1800}, |
| 273 | {"AST" , -4*3600}, |
| 274 | {"ADT" , -3*3600}, |
| 275 | {"EST" , -5*3600}, |
| 276 | {"EDT" , -4*3600}, |
| 277 | {"CST" , -6*3600}, |
| 278 | {"CDT" , -5*3600}, |
| 279 | {"MST" , -7*3600}, |
| 280 | {"MDT" , -6*3600}, |
| 281 | {"PST" , -8*3600}, |
| 282 | {"PDT" , -7*3600}, |
| 283 | {"AKST" , -9*3600}, |
| 284 | {"AKDT" , -8*3600}, |
| 285 | {"HST" , -10*3600}, |
| 286 | {"AEST" , 10*3600}, |
| 287 | {"AEDT" , 11*3600}, |
| 288 | {"ACST" , 9*3600+1800}, |
| 289 | {"ACDT" , 10*3600+1800}, |
| 290 | {"AWST" , 8*3600}, |
| 291 | {"AWDT" , 9*3600} |
| 292 | }; |
| 293 | |
| 294 | int tzd = 0; |
| 295 | while (it != end && Ascii::isSpace(*it)) ++it; |
| 296 | if (it != end) |
| 297 | { |
| 298 | if (Ascii::isAlpha(*it)) |
| 299 | { |
| 300 | std::string designator; |
| 301 | designator += *it++; |
| 302 | if (it != end && Ascii::isAlpha(*it)) designator += *it++; |
| 303 | if (it != end && Ascii::isAlpha(*it)) designator += *it++; |
| 304 | if (it != end && Ascii::isAlpha(*it)) designator += *it++; |
| 305 | for (unsigned i = 0; i < sizeof(zones)/sizeof(Zone); ++i) |
| 306 | { |
| 307 | if (designator == zones[i].designator) |
| 308 | { |
| 309 | tzd = zones[i].timeZoneDifferential; |
| 310 | break; |
| 311 | } |
| 312 | } |
| 313 | } |
| 314 | if (it != end && (*it == '+' || *it == '-')) |
| 315 | { |
| 316 | int sign = *it == '+' ? 1 : -1; |
| 317 | ++it; |
| 318 | int hours = 0; |
| 319 | PARSE_NUMBER_N(hours, 2); |
| 320 | if (it != end && *it == ':') ++it; |
| 321 | int minutes = 0; |
| 322 | PARSE_NUMBER_N(minutes, 2); |
| 323 | tzd += sign*(hours*3600 + minutes*60); |
| 324 | } |
| 325 | } |
| 326 | return tzd; |
| 327 | } |
| 328 | |
| 329 | |
| 330 | int DateTimeParser::parseMonth(std::string::const_iterator& it, const std::string::const_iterator& end) |
| 331 | { |
| 332 | std::string month; |
| 333 | while (it != end && (Ascii::isSpace(*it) || Ascii::isPunct(*it))) ++it; |
| 334 | bool isFirst = true; |
| 335 | while (it != end && Ascii::isAlpha(*it)) |
| 336 | { |
| 337 | char ch = (*it++); |
| 338 | if (isFirst) { month += Ascii::toUpper(ch); isFirst = false; } |
| 339 | else month += Ascii::toLower(ch); |
| 340 | } |
| 341 | if (month.length() < 3) throw SyntaxException("Month name must be at least three characters long" , month); |
| 342 | for (int i = 0; i < 12; ++i) |
| 343 | { |
| 344 | if (DateTimeFormat::MONTH_NAMES[i].find(month) == 0) |
| 345 | return i + 1; |
| 346 | } |
| 347 | throw SyntaxException("Not a valid month name" , month); |
| 348 | } |
| 349 | |
| 350 | |
| 351 | int DateTimeParser::parseDayOfWeek(std::string::const_iterator& it, const std::string::const_iterator& end) |
| 352 | { |
| 353 | std::string dow; |
| 354 | while (it != end && (Ascii::isSpace(*it) || Ascii::isPunct(*it))) ++it; |
| 355 | bool isFirst = true; |
| 356 | while (it != end && Ascii::isAlpha(*it)) |
| 357 | { |
| 358 | char ch = (*it++); |
| 359 | if (isFirst) { dow += Ascii::toUpper(ch); isFirst = false; } |
| 360 | else dow += Ascii::toLower(ch); |
| 361 | } |
| 362 | if (dow.length() < 3) throw SyntaxException("Weekday name must be at least three characters long" , dow); |
| 363 | for (int i = 0; i < 7; ++i) |
| 364 | { |
| 365 | if (DateTimeFormat::WEEKDAY_NAMES[i].find(dow) == 0) |
| 366 | return i; |
| 367 | } |
| 368 | throw SyntaxException("Not a valid weekday name" , dow); |
| 369 | } |
| 370 | |
| 371 | |
| 372 | int DateTimeParser::parseAMPM(std::string::const_iterator& it, const std::string::const_iterator& end, int hour) |
| 373 | { |
| 374 | std::string ampm; |
| 375 | while (it != end && (Ascii::isSpace(*it) || Ascii::isPunct(*it))) ++it; |
| 376 | while (it != end && Ascii::isAlpha(*it)) |
| 377 | { |
| 378 | char ch = (*it++); |
| 379 | ampm += Ascii::toUpper(ch); |
| 380 | } |
| 381 | if (ampm == "AM" ) |
| 382 | { |
| 383 | if (hour == 12) |
| 384 | return 0; |
| 385 | else |
| 386 | return hour; |
| 387 | } |
| 388 | else if (ampm == "PM" ) |
| 389 | { |
| 390 | if (hour < 12) |
| 391 | return hour + 12; |
| 392 | else |
| 393 | return hour; |
| 394 | } |
| 395 | else throw SyntaxException("Not a valid AM/PM designator" , ampm); |
| 396 | } |
| 397 | |
| 398 | |
| 399 | } // namespace Poco |
| 400 | |