[Temporal] Part 2.1 Add TemporalParser to parse ISO8601

Parser based on
https://tc39.es/proposal-temporal/#sec-temporal-iso8601grammar

Bug: v8:11544
Change-Id: I18eafc86da86005d5aee7b672c145fcf38a3ef5e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3271827
Reviewed-by: Adam Klein <adamk@chromium.org>
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Commit-Queue: Frank Tang <ftang@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78049}

BUILD.bazel

@@ -1987,6 +1987,8 @@ filegroup(
         "src/tasks/operations-barrier.h",
         "src/tasks/task-utils.cc",
         "src/tasks/task-utils.h",
+        "src/temporal/temporal-parser.cc",
+        "src/temporal/temporal-parser.h",
         "src/torque/runtime-macro-shims.h",
         "src/third_party/siphash/halfsiphash.cc",
         "src/third_party/siphash/halfsiphash.h",

BUILD.gn

@@ -3370,6 +3370,7 @@ v8_header_set("v8_internal_headers") {
     "src/tasks/cancelable-task.h",
     "src/tasks/operations-barrier.h",
     "src/tasks/task-utils.h",
+    "src/temporal/temporal-parser.h",
     "src/third_party/siphash/halfsiphash.h",
     "src/third_party/utf8-decoder/utf8-decoder.h",
     "src/torque/runtime-macro-shims.h",
@@ -4379,6 +4380,7 @@ v8_source_set("v8_base_without_compiler") {
     "src/tasks/cancelable-task.cc",
     "src/tasks/operations-barrier.cc",
     "src/tasks/task-utils.cc",
+    "src/temporal/temporal-parser.cc",
     "src/third_party/siphash/halfsiphash.cc",
     "src/tracing/trace-event.cc",
     "src/tracing/traced-value.cc",

src/temporal/OWNERS

@@ -0,0 +1,2 @@
+ftang@chromium.org
+syg@chromium.org

src/temporal/temporal-parser.h

@@ -0,0 +1,147 @@
+// Copyright 2021 the V8 project authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#ifndef V8_TEMPORAL_TEMPORAL_PARSER_H_
+#define V8_TEMPORAL_TEMPORAL_PARSER_H_
+
+#include "src/execution/isolate.h"
+
+namespace v8 {
+namespace internal {
+
+/**
+ * ParsedISO8601Result contains the parsed result of ISO 8601 grammar
+ * documented in #sec-temporal-iso8601grammar
+ * for TemporalInstantString, TemporalZonedDateTimeString,
+ * TemporalCalendarString, TemporalDateString, TemporalDateTimeString,
+ * TemporalMonthDayString, TemporalRelativeToString, TemporalTimeString,
+ * TemporalTimeZoneString, and TemporalYearMonthString. For all the fields
+ * represented by int32_t, a special value kMinInt31 is used to represent the
+ * field is "undefined" after parsing.
+ */
+struct ParsedISO8601Result {
+  int32_t date_year;    // DateYear production
+  int32_t date_month;   // DateMonth production
+  int32_t date_day;     // DateDay production
+  int32_t time_hour;    // TimeHour production
+  int32_t time_minute;  // TimeMinute production
+  int32_t time_second;  // TimeSecond production
+  int32_t
+      time_nanosecond;  // TimeFractionalPart production stored in nanosecond
+  int32_t tzuo_sign;    // TimeZoneUTCOffsetSign production
+  int32_t tzuo_hour;    // TimeZoneUTCOffsetHour production
+  int32_t tzuo_minute;  // TimeZoneUTCOffsetMinute production
+  int32_t tzuo_second;  // TimeZoneUTCOffsetSecond production
+  int32_t
+      tzuo_nanosecond;  // TimeZoneUTCOffsetFractionalPart stored in nanosecond
+  bool utc_designator;  // UTCDesignator is presented
+  int32_t tzi_name_start;   // Starting offset of TimeZoneIANAName in the input
+                            // string.
+  int32_t tzi_name_length;  // Length of TimeZoneIANAName production
+  int32_t calendar_name_start;  // Starting offset of CalendarName production in
+                                // the input string.
+  int32_t calendar_name_length;  // Length of CalendarName production.
+
+  ParsedISO8601Result()
+      : date_year(kMinInt31),
+        date_month(kMinInt31),
+        date_day(kMinInt31),
+        time_hour(kMinInt31),
+        time_minute(kMinInt31),
+        time_second(kMinInt31),
+        time_nanosecond(kMinInt31),
+        tzuo_sign(kMinInt31),
+        tzuo_hour(kMinInt31),
+        tzuo_minute(kMinInt31),
+        tzuo_second(kMinInt31),
+        tzuo_nanosecond(kMinInt31),
+        utc_designator(false),
+        tzi_name_start(0),
+        tzi_name_length(0),
+        calendar_name_start(0),
+        calendar_name_length(0) {}
+
+  bool date_year_is_undefined() const { return date_year == kMinInt31; }
+  bool date_month_is_undefined() const { return date_month == kMinInt31; }
+  bool date_day_is_undefined() const { return date_day == kMinInt31; }
+  bool time_hour_is_undefined() const { return time_hour == kMinInt31; }
+  bool time_minute_is_undefined() const { return time_minute == kMinInt31; }
+  bool time_second_is_undefined() const { return time_second == kMinInt31; }
+  bool time_nanosecond_is_undefined() const {
+    return time_nanosecond == kMinInt31;
+  }
+  bool tzuo_hour_is_undefined() const { return tzuo_hour == kMinInt31; }
+  bool tzuo_minute_is_undefined() const { return tzuo_minute == kMinInt31; }
+  bool tzuo_second_is_undefined() const { return tzuo_second == kMinInt31; }
+  bool tzuo_sign_is_undefined() const { return tzuo_sign == kMinInt31; }
+  bool tzuo_nanosecond_is_undefined() const {
+    return tzuo_nanosecond == kMinInt31;
+  }
+};
+
+/**
+ * ParsedISO8601Duration contains the parsed result of ISO 8601 grammar
+ * documented in #prod-TemporalDurationString
+ * for TemporalDurationString.
+ */
+struct ParsedISO8601Duration {
+  int64_t sign;              // Sign production
+  int64_t years;             // DurationYears production
+  int64_t months;            // DurationMonths production
+  int64_t weeks;             // DurationWeeks production
+  int64_t days;              // DurationDays production
+  int64_t whole_hours;       // DurationWholeHours production
+  int64_t hours_fraction;    // DurationHoursFraction, in unit of 1e-9 hours
+  int64_t whole_minutes;     // DurationWholeMinutes production
+  int64_t minutes_fraction;  // DurationMinuteFraction, in unit of 1e-9 minutes
+  int64_t whole_seconds;     // DurationWholeSeconds production
+  int64_t seconds_fraction;  // DurationSecondFraction, in unit of nanosecond (
+                             // 1e-9 seconds).
+
+  ParsedISO8601Duration()
+      : sign(1),
+        years(0),
+        months(0),
+        weeks(0),
+        days(0),
+        whole_hours(0),
+        hours_fraction(0),
+        whole_minutes(0),
+        minutes_fraction(0),
+        whole_seconds(0),
+        seconds_fraction(0) {}
+};
+
+/**
+ * TemporalParser is low level parsing functions to support the implementation
+ * of various ParseTemporal*String Abstract Operations listed after
+ * #sec-temporal-parsetemporalinstantstring.
+ * All the methods take an Isolate, a Handle<String> as input, and also a
+ * pointer to a bool to answer the "satisfy the syntax of a Temporal*String"
+ * question and return the parsed result.
+ */
+class V8_EXPORT_PRIVATE TemporalParser {
+ public:
+#define DEFINE_PARSE_METHOD(R, NAME)                 \
+  V8_WARN_UNUSED_RESULT static Maybe<R> Parse##NAME( \
+      Isolate* isolate, Handle<String> iso_string, bool* satisfy)
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalDateString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalDateTimeString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalTimeString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalYearMonthString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalMonthDayString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalInstantString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalZonedDateTimeString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalTimeZoneString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalRelativeToString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TemporalCalendarString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Duration, TemporalDurationString);
+  DEFINE_PARSE_METHOD(ParsedISO8601Result, TimeZoneNumericUTCOffset);
+};
+#undef DEFINE_PARSE_METHOD
+
+}  // namespace internal
+}  // namespace v8
+
+#endif  // V8_TEMPORAL_TEMPORAL_PARSER_H_

test/cctest/BUILD.gn

@@ -277,6 +277,7 @@ v8_source_set("cctest_sources") {
     "test-swiss-name-dictionary-infra.cc",
     "test-swiss-name-dictionary.cc",
     "test-symbols.cc",
+    "test-temporal-parser.cc",
     "test-thread-termination.cc",
     "test-threads.cc",
     "test-trace-event.cc",