rok commented on a change in pull request #10457:
URL: https://github.com/apache/arrow/pull/10457#discussion_r657531497
##########
File path: cpp/src/arrow/compute/kernels/scalar_temporal_test.cc
##########
@@ -143,39 +142,204 @@ TEST(ScalarTemporalTest,
TestTemporalComponentExtractionWithDifferentUnits) {
CheckScalarUnary("quarter", unit, times, int64(), quarter);
CheckScalarUnary("hour", unit, times, int64(), hour);
CheckScalarUnary("minute", unit, times, int64(), minute);
- CheckScalarUnary("second", unit, times, float64(), second);
+ CheckScalarUnary("second", unit, times, int64(), second);
CheckScalarUnary("millisecond", unit, times, int64(), zeros);
CheckScalarUnary("microsecond", unit, times, int64(), zeros);
CheckScalarUnary("nanosecond", unit, times, int64(), zeros);
CheckScalarUnary("subsecond", unit, times, float64(), zeros);
}
}
-TEST(ScalarTemporalTest, TestZonedTemporalComponentExtraction) {
- std::string timezone = "Asia/Kolkata";
- const char* times = R"(["1970-01-01T00:00:59", null])";
+TEST(ScalarTemporalTest, TestOutsideNanosecondRange) {
+ const char* times = R"(["1677-09-20T00:00:59.123456",
"2262-04-13T23:23:23.999999"])";
- for (auto u : internal::AllTimeUnits()) {
- auto unit = timestamp(u, timezone);
- auto timestamps = ArrayFromJSON(unit, times);
-
- ASSERT_RAISES(NotImplemented, Year(timestamps));
- ASSERT_RAISES(NotImplemented, Month(timestamps));
- ASSERT_RAISES(NotImplemented, Day(timestamps));
- ASSERT_RAISES(NotImplemented, DayOfWeek(timestamps));
- ASSERT_RAISES(NotImplemented, DayOfYear(timestamps));
- ASSERT_RAISES(NotImplemented, ISOYear(timestamps));
- ASSERT_RAISES(NotImplemented, ISOWeek(timestamps));
- ASSERT_RAISES(NotImplemented, ISOCalendar(timestamps));
- ASSERT_RAISES(NotImplemented, Quarter(timestamps));
- ASSERT_RAISES(NotImplemented, Hour(timestamps));
- ASSERT_RAISES(NotImplemented, Minute(timestamps));
- ASSERT_RAISES(NotImplemented, Second(timestamps));
- ASSERT_RAISES(NotImplemented, Millisecond(timestamps));
- ASSERT_RAISES(NotImplemented, Microsecond(timestamps));
- ASSERT_RAISES(NotImplemented, Nanosecond(timestamps));
- ASSERT_RAISES(NotImplemented, Subsecond(timestamps));
- }
+ auto unit = timestamp(TimeUnit::MICRO);
+ auto iso_calendar_type =
+ struct_({field("iso_year", int64()), field("iso_week", int64()),
+ field("iso_day_of_week", int64())});
+
+ auto year = "[1677, 2262]";
+ auto month = "[9, 4]";
+ auto day = "[20, 13]";
+ auto day_of_week = "[0, 6]";
+ auto day_of_year = "[263, 103]";
+ auto iso_year = "[1677, 2262]";
+ auto iso_week = "[38, 15]";
+ auto iso_calendar =
+ ArrayFromJSON(iso_calendar_type,
+ R"([{"iso_year": 1677, "iso_week": 38, "iso_day_of_week":
1},
+ {"iso_year": 2262, "iso_week": 15,
"iso_day_of_week": 7}])");
+ auto quarter = "[3, 2]";
+ auto hour = "[0, 23]";
+ auto minute = "[0, 23]";
+ auto second = "[59, 23]";
+ auto millisecond = "[123, 999]";
+ auto microsecond = "[456, 999]";
+ auto nanosecond = "[0, 0]";
+ auto subsecond = "[0.123456, 0.999999]";
+
+ CheckScalarUnary("year", unit, times, int64(), year);
+ CheckScalarUnary("month", unit, times, int64(), month);
+ CheckScalarUnary("day", unit, times, int64(), day);
+ CheckScalarUnary("day_of_week", unit, times, int64(), day_of_week);
+ CheckScalarUnary("day_of_year", unit, times, int64(), day_of_year);
+ CheckScalarUnary("iso_year", unit, times, int64(), iso_year);
+ CheckScalarUnary("iso_week", unit, times, int64(), iso_week);
+ CheckScalarUnary("iso_calendar", ArrayFromJSON(unit, times), iso_calendar);
+ CheckScalarUnary("quarter", unit, times, int64(), quarter);
+ CheckScalarUnary("hour", unit, times, int64(), hour);
+ CheckScalarUnary("minute", unit, times, int64(), minute);
+ CheckScalarUnary("second", unit, times, int64(), second);
+ CheckScalarUnary("millisecond", unit, times, int64(), millisecond);
+ CheckScalarUnary("microsecond", unit, times, int64(), microsecond);
+ CheckScalarUnary("nanosecond", unit, times, int64(), nanosecond);
+ CheckScalarUnary("subsecond", unit, times, float64(), subsecond);
+}
+
+#ifndef _WIN32
Review comment:
We're using `tz.h` library which needs an updated timezone database to
correctly handle timezoned timestamps. See [installation
instructions](https://howardhinnant.github.io/date/tz.html#Installation).
We have the following options (if I understand correctly) for getting a
timezone database:
1. local (non-windows) OS timezone database - no work required.
2. arrow bundled folder - we could bundle the database at build time for
windows. Database would slowly go stale.
3. download it from IANA Time Zone Database at runtime - `tz.h` gets the
database at runtime, but curl (and 7-zip on windows) are required.
4. local user-provided folder - user could provide a location at buildtime.
Nice to have.
@kszucs - where would be a good place to implement 2?
Does 3 look like something we could do?
For now it would probably be best to create a Jira for this and keep these
test disabled on windows.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
