Ben Murdoch | 8b112d2 | 2011-06-08 16:22:53 +0100 | [diff] [blame] | 1 | // Copyright 2011 the V8 project authors. All rights reserved. |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 2 | // Redistribution and use in source and binary forms, with or without |
| 3 | // modification, are permitted provided that the following conditions are |
| 4 | // met: |
| 5 | // |
| 6 | // * Redistributions of source code must retain the above copyright |
| 7 | // notice, this list of conditions and the following disclaimer. |
| 8 | // * Redistributions in binary form must reproduce the above |
| 9 | // copyright notice, this list of conditions and the following |
| 10 | // disclaimer in the documentation and/or other materials provided |
| 11 | // with the distribution. |
| 12 | // * Neither the name of Google Inc. nor the names of its |
| 13 | // contributors may be used to endorse or promote products derived |
| 14 | // from this software without specific prior written permission. |
| 15 | // |
| 16 | // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
| 17 | // "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
| 18 | // LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR |
| 19 | // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT |
| 20 | // OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, |
| 21 | // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
| 22 | // LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, |
| 23 | // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY |
| 24 | // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| 25 | // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
| 26 | // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 27 | |
| 28 | #ifndef V8_DATEPARSER_INL_H_ |
| 29 | #define V8_DATEPARSER_INL_H_ |
| 30 | |
Steve Block | 3ce2e20 | 2009-11-05 08:53:23 +0000 | [diff] [blame] | 31 | #include "dateparser.h" |
| 32 | |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 33 | namespace v8 { |
| 34 | namespace internal { |
| 35 | |
| 36 | template <typename Char> |
Ben Murdoch | 8b112d2 | 2011-06-08 16:22:53 +0100 | [diff] [blame] | 37 | bool DateParser::Parse(Vector<Char> str, |
| 38 | FixedArray* out, |
| 39 | UnicodeCache* unicode_cache) { |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 40 | ASSERT(out->length() >= OUTPUT_SIZE); |
Ben Murdoch | 8b112d2 | 2011-06-08 16:22:53 +0100 | [diff] [blame] | 41 | InputReader<Char> in(unicode_cache, str); |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 42 | DateStringTokenizer<Char> scanner(&in); |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 43 | TimeZoneComposer tz; |
| 44 | TimeComposer time; |
| 45 | DayComposer day; |
| 46 | |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 47 | // Specification: |
| 48 | // Accept ES5 ISO 8601 date-time-strings or legacy dates compatible |
| 49 | // with Safari. |
| 50 | // ES5 ISO 8601 dates: |
| 51 | // [('-'|'+')yy]yyyy[-MM[-DD]][THH:mm[:ss[.sss]][Z|(+|-)hh:mm]] |
| 52 | // where yyyy is in the range 0000..9999 and |
| 53 | // +/-yyyyyy is in the range -999999..+999999 - |
| 54 | // but -000000 is invalid (year zero must be positive), |
| 55 | // MM is in the range 01..12, |
| 56 | // DD is in the range 01..31, |
| 57 | // MM and DD defaults to 01 if missing,, |
| 58 | // HH is generally in the range 00..23, but can be 24 if mm, ss |
| 59 | // and sss are zero (or missing), representing midnight at the |
| 60 | // end of a day, |
| 61 | // mm and ss are in the range 00..59, |
| 62 | // sss is in the range 000..999, |
| 63 | // hh is in the range 00..23, |
| 64 | // mm, ss, and sss default to 00 if missing, and |
| 65 | // timezone defaults to Z if missing. |
| 66 | // Extensions: |
| 67 | // We also allow sss to have more or less than three digits (but at |
| 68 | // least one). |
| 69 | // We allow hh:mm to be specified as hhmm. |
| 70 | // Legacy dates: |
| 71 | // Any unrecognized word before the first number is ignored. |
| 72 | // Parenthesized text is ignored. |
| 73 | // An unsigned number followed by ':' is a time value, and is |
| 74 | // added to the TimeComposer. A number followed by '::' adds a second |
| 75 | // zero as well. A number followed by '.' is also a time and must be |
| 76 | // followed by milliseconds. |
| 77 | // Any other number is a date component and is added to DayComposer. |
| 78 | // A month name (or really: any word having the same first three letters |
| 79 | // as a month name) is recorded as a named month in the Day composer. |
| 80 | // A word recognizable as a time-zone is recorded as such, as is |
| 81 | // '(+|-)(hhmm|hh:)'. |
| 82 | // Legacy dates don't allow extra signs ('+' or '-') or umatched ')' |
| 83 | // after a number has been read (before the first number, any garbage |
| 84 | // is allowed). |
| 85 | // Intersection of the two: |
| 86 | // A string that matches both formats (e.g. 1970-01-01) will be |
| 87 | // parsed as an ES5 date-time string - which means it will default |
| 88 | // to UTC time-zone. That's unavoidable if following the ES5 |
| 89 | // specification. |
| 90 | // After a valid "T" has been read while scanning an ES5 datetime string, |
| 91 | // the input can no longer be a valid legacy date, since the "T" is a |
| 92 | // garbage string after a number has been read. |
| 93 | |
| 94 | // First try getting as far as possible with as ES5 Date Time String. |
| 95 | DateToken next_unhandled_token = ParseES5DateTime(&scanner, &day, &time, &tz); |
| 96 | if (next_unhandled_token.IsInvalid()) return false; |
| 97 | bool has_read_number = !day.IsEmpty(); |
| 98 | // If there's anything left, continue with the legacy parser. |
| 99 | for (DateToken token = next_unhandled_token; |
| 100 | !token.IsEndOfInput(); |
| 101 | token = scanner.Next()) { |
| 102 | if (token.IsNumber()) { |
| 103 | has_read_number = true; |
| 104 | int n = token.number(); |
| 105 | if (scanner.SkipSymbol(':')) { |
| 106 | if (scanner.SkipSymbol(':')) { |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 107 | // n + "::" |
| 108 | if (!time.IsEmpty()) return false; |
| 109 | time.Add(n); |
| 110 | time.Add(0); |
| 111 | } else { |
| 112 | // n + ":" |
| 113 | if (!time.Add(n)) return false; |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 114 | if (scanner.Peek().IsSymbol('.')) scanner.Next(); |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 115 | } |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 116 | } else if (scanner.SkipSymbol('.') && time.IsExpecting(n)) { |
Steve Block | 6ded16b | 2010-05-10 14:33:55 +0100 | [diff] [blame] | 117 | time.Add(n); |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 118 | if (!scanner.Peek().IsNumber()) return false; |
| 119 | int n = ReadMilliseconds(scanner.Next()); |
| 120 | if (n < 0) return false; |
Steve Block | 6ded16b | 2010-05-10 14:33:55 +0100 | [diff] [blame] | 121 | time.AddFinal(n); |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 122 | } else if (tz.IsExpecting(n)) { |
| 123 | tz.SetAbsoluteMinute(n); |
| 124 | } else if (time.IsExpecting(n)) { |
| 125 | time.AddFinal(n); |
Steve Block | 5915150 | 2010-09-22 15:07:15 +0100 | [diff] [blame] | 126 | // Require end, white space, "Z", "+" or "-" immediately after |
| 127 | // finalizing time. |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 128 | DateToken peek = scanner.Peek(); |
| 129 | if (!peek.IsEndOfInput() && |
| 130 | !peek.IsWhiteSpace() && |
| 131 | !peek.IsKeywordZ() && |
| 132 | !peek.IsAsciiSign()) return false; |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 133 | } else { |
| 134 | if (!day.Add(n)) return false; |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 135 | scanner.SkipSymbol('-'); |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 136 | } |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 137 | } else if (token.IsKeyword()) { |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 138 | // Parse a "word" (sequence of chars. >= 'A'). |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 139 | KeywordType type = token.keyword_type(); |
| 140 | int value = token.keyword_value(); |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 141 | if (type == AM_PM && !time.IsEmpty()) { |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 142 | time.SetHourOffset(value); |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 143 | } else if (type == MONTH_NAME) { |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 144 | day.SetNamedMonth(value); |
| 145 | scanner.SkipSymbol('-'); |
| 146 | } else if (type == TIME_ZONE_NAME && has_read_number) { |
| 147 | tz.Set(value); |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 148 | } else { |
| 149 | // Garbage words are illegal if a number has been read. |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 150 | if (has_read_number) return false; |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 151 | } |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 152 | } else if (token.IsAsciiSign() && (tz.IsUTC() || !time.IsEmpty())) { |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 153 | // Parse UTC offset (only after UTC or time). |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 154 | tz.SetSign(token.ascii_sign()); |
| 155 | // The following number may be empty. |
| 156 | int n = 0; |
| 157 | if (scanner.Peek().IsNumber()) { |
| 158 | n = scanner.Next().number(); |
| 159 | } |
| 160 | has_read_number = true; |
| 161 | |
| 162 | if (scanner.Peek().IsSymbol(':')) { |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 163 | tz.SetAbsoluteHour(n); |
| 164 | tz.SetAbsoluteMinute(kNone); |
| 165 | } else { |
| 166 | tz.SetAbsoluteHour(n / 100); |
| 167 | tz.SetAbsoluteMinute(n % 100); |
| 168 | } |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 169 | } else if ((token.IsAsciiSign() || token.IsSymbol(')')) && |
| 170 | has_read_number) { |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 171 | // Extra sign or ')' is illegal if a number has been read. |
| 172 | return false; |
| 173 | } else { |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 174 | // Ignore other characters and whitespace. |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 175 | } |
| 176 | } |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 177 | |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 178 | return day.Write(out) && time.Write(out) && tz.Write(out); |
| 179 | } |
| 180 | |
Ben Murdoch | 3fb3ca8 | 2011-12-02 17:19:32 +0000 | [diff] [blame] | 181 | |
| 182 | template<typename CharType> |
| 183 | DateParser::DateToken DateParser::DateStringTokenizer<CharType>::Scan() { |
| 184 | int pre_pos = in_->position(); |
| 185 | if (in_->IsEnd()) return DateToken::EndOfInput(); |
| 186 | if (in_->IsAsciiDigit()) { |
| 187 | int n = in_->ReadUnsignedNumeral(); |
| 188 | int length = in_->position() - pre_pos; |
| 189 | return DateToken::Number(n, length); |
| 190 | } |
| 191 | if (in_->Skip(':')) return DateToken::Symbol(':'); |
| 192 | if (in_->Skip('-')) return DateToken::Symbol('-'); |
| 193 | if (in_->Skip('+')) return DateToken::Symbol('+'); |
| 194 | if (in_->Skip('.')) return DateToken::Symbol('.'); |
| 195 | if (in_->Skip(')')) return DateToken::Symbol(')'); |
| 196 | if (in_->IsAsciiAlphaOrAbove()) { |
| 197 | ASSERT(KeywordTable::kPrefixLength == 3); |
| 198 | uint32_t buffer[3] = {0, 0, 0}; |
| 199 | int length = in_->ReadWord(buffer, 3); |
| 200 | int index = KeywordTable::Lookup(buffer, length); |
| 201 | return DateToken::Keyword(KeywordTable::GetType(index), |
| 202 | KeywordTable::GetValue(index), |
| 203 | length); |
| 204 | } |
| 205 | if (in_->SkipWhiteSpace()) { |
| 206 | return DateToken::WhiteSpace(in_->position() - pre_pos); |
| 207 | } |
| 208 | if (in_->SkipParentheses()) { |
| 209 | return DateToken::Unknown(); |
| 210 | } |
| 211 | in_->Next(); |
| 212 | return DateToken::Unknown(); |
| 213 | } |
| 214 | |
| 215 | |
| 216 | template <typename Char> |
| 217 | DateParser::DateToken DateParser::ParseES5DateTime( |
| 218 | DateStringTokenizer<Char>* scanner, |
| 219 | DayComposer* day, |
| 220 | TimeComposer* time, |
| 221 | TimeZoneComposer* tz) { |
| 222 | ASSERT(day->IsEmpty()); |
| 223 | ASSERT(time->IsEmpty()); |
| 224 | ASSERT(tz->IsEmpty()); |
| 225 | |
| 226 | // Parse mandatory date string: [('-'|'+')yy]yyyy[':'MM[':'DD]] |
| 227 | if (scanner->Peek().IsAsciiSign()) { |
| 228 | // Keep the sign token, so we can pass it back to the legacy |
| 229 | // parser if we don't use it. |
| 230 | DateToken sign_token = scanner->Next(); |
| 231 | if (!scanner->Peek().IsFixedLengthNumber(6)) return sign_token; |
| 232 | int sign = sign_token.ascii_sign(); |
| 233 | int year = scanner->Next().number(); |
| 234 | if (sign < 0 && year == 0) return sign_token; |
| 235 | day->Add(sign * year); |
| 236 | } else if (scanner->Peek().IsFixedLengthNumber(4)) { |
| 237 | day->Add(scanner->Next().number()); |
| 238 | } else { |
| 239 | return scanner->Next(); |
| 240 | } |
| 241 | if (scanner->SkipSymbol('-')) { |
| 242 | if (!scanner->Peek().IsFixedLengthNumber(2) || |
| 243 | !DayComposer::IsMonth(scanner->Peek().number())) return scanner->Next(); |
| 244 | day->Add(scanner->Next().number()); |
| 245 | if (scanner->SkipSymbol('-')) { |
| 246 | if (!scanner->Peek().IsFixedLengthNumber(2) || |
| 247 | !DayComposer::IsDay(scanner->Peek().number())) return scanner->Next(); |
| 248 | day->Add(scanner->Next().number()); |
| 249 | } |
| 250 | } |
| 251 | // Check for optional time string: 'T'HH':'mm[':'ss['.'sss]]Z |
| 252 | if (!scanner->Peek().IsKeywordType(TIME_SEPARATOR)) { |
| 253 | if (!scanner->Peek().IsEndOfInput()) return scanner->Next(); |
| 254 | } else { |
| 255 | // ES5 Date Time String time part is present. |
| 256 | scanner->Next(); |
| 257 | if (!scanner->Peek().IsFixedLengthNumber(2) || |
| 258 | !Between(scanner->Peek().number(), 0, 24)) { |
| 259 | return DateToken::Invalid(); |
| 260 | } |
| 261 | // Allow 24:00[:00[.000]], but no other time starting with 24. |
| 262 | bool hour_is_24 = (scanner->Peek().number() == 24); |
| 263 | time->Add(scanner->Next().number()); |
| 264 | if (!scanner->SkipSymbol(':')) return DateToken::Invalid(); |
| 265 | if (!scanner->Peek().IsFixedLengthNumber(2) || |
| 266 | !TimeComposer::IsMinute(scanner->Peek().number()) || |
| 267 | (hour_is_24 && scanner->Peek().number() > 0)) { |
| 268 | return DateToken::Invalid(); |
| 269 | } |
| 270 | time->Add(scanner->Next().number()); |
| 271 | if (scanner->SkipSymbol(':')) { |
| 272 | if (!scanner->Peek().IsFixedLengthNumber(2) || |
| 273 | !TimeComposer::IsSecond(scanner->Peek().number()) || |
| 274 | (hour_is_24 && scanner->Peek().number() > 0)) { |
| 275 | return DateToken::Invalid(); |
| 276 | } |
| 277 | time->Add(scanner->Next().number()); |
| 278 | if (scanner->SkipSymbol('.')) { |
| 279 | if (!scanner->Peek().IsNumber() || |
| 280 | (hour_is_24 && scanner->Peek().number() > 0)) { |
| 281 | return DateToken::Invalid(); |
| 282 | } |
| 283 | // Allow more or less than the mandated three digits. |
| 284 | time->Add(ReadMilliseconds(scanner->Next())); |
| 285 | } |
| 286 | } |
| 287 | // Check for optional timezone designation: 'Z' | ('+'|'-')hh':'mm |
| 288 | if (scanner->Peek().IsKeywordZ()) { |
| 289 | scanner->Next(); |
| 290 | tz->Set(0); |
| 291 | } else if (scanner->Peek().IsSymbol('+') || |
| 292 | scanner->Peek().IsSymbol('-')) { |
| 293 | tz->SetSign(scanner->Next().symbol() == '+' ? 1 : -1); |
| 294 | if (scanner->Peek().IsFixedLengthNumber(4)) { |
| 295 | // hhmm extension syntax. |
| 296 | int hourmin = scanner->Next().number(); |
| 297 | int hour = hourmin / 100; |
| 298 | int min = hourmin % 100; |
| 299 | if (!TimeComposer::IsHour(hour) || !TimeComposer::IsMinute(min)) { |
| 300 | return DateToken::Invalid(); |
| 301 | } |
| 302 | tz->SetAbsoluteHour(hour); |
| 303 | tz->SetAbsoluteMinute(min); |
| 304 | } else { |
| 305 | // hh:mm standard syntax. |
| 306 | if (!scanner->Peek().IsFixedLengthNumber(2) || |
| 307 | !TimeComposer::IsHour(scanner->Peek().number())) { |
| 308 | return DateToken::Invalid(); |
| 309 | } |
| 310 | tz->SetAbsoluteHour(scanner->Next().number()); |
| 311 | if (!scanner->SkipSymbol(':')) return DateToken::Invalid(); |
| 312 | if (!scanner->Peek().IsFixedLengthNumber(2) || |
| 313 | !TimeComposer::IsMinute(scanner->Peek().number())) { |
| 314 | return DateToken::Invalid(); |
| 315 | } |
| 316 | tz->SetAbsoluteMinute(scanner->Next().number()); |
| 317 | } |
| 318 | } |
| 319 | if (!scanner->Peek().IsEndOfInput()) return DateToken::Invalid(); |
| 320 | } |
| 321 | // Successfully parsed ES5 Date Time String. Default to UTC if no TZ given. |
| 322 | if (tz->IsEmpty()) tz->Set(0); |
| 323 | day->set_iso_date(); |
| 324 | return DateToken::EndOfInput(); |
| 325 | } |
| 326 | |
| 327 | |
Steve Block | a7e24c1 | 2009-10-30 11:49:00 +0000 | [diff] [blame] | 328 | } } // namespace v8::internal |
| 329 | |
| 330 | #endif // V8_DATEPARSER_INL_H_ |