ZoneEqualityComparer Options EnumerationNoda Time
Options to use when comparing time zones for equality. Each option makes the comparison more restrictive.

Namespace: NodaTime.TimeZones
Assembly: NodaTime (in NodaTime.dll) Version: 1.3.0.0 (1.3.1)
Syntax

[FlagsAttribute]
public enum Options
Members

  Member nameValueDescription
OnlyMatchWallOffset0 The default comparison, which only cares about the wall offset at any particular instant, within the interval of the comparer. In other words, if GetUtcOffset(Instant) returns the same value for all instants in the interval, the comparer will consider the zones to be equal.
MatchOffsetComponents1 Instead of only comparing wall offsets, the standard/savings split is also considered. So when this option is used, two zones which both have a wall offset of +2 at one instant would be considered unequal if one of those offsets was +1 standard, +1 savings and the other was +2 standard with no daylight saving.
MatchNames2 Compare the names of zone intervals as well as offsets.
MatchAllTransitions4 This option prevents adjacent zone intervals from being coalesced, even if they are otherwise considered equivalent according to other options.
MatchStartAndEndTransitions8 Includes the transitions into the first zone interval and out of the last zone interval as part of the comparison, even if they do not affect the offset or name for any instant within the operating interval.
StrictestMatch15 The combination of all available match options.
Remarks

By default, the comparer only compares the wall offset (total of standard offset and any daylight saving offset) at every instant within the interval over which the comparer operates. In practice, this is done by comparing each ZoneInterval which includes an instant within the interval (using GetZoneIntervals(Interval)). For most purposes, this is all that's required: from the simple perspective of a time zone being just a function from instants to local time, the default option of OnlyMatchWallOffset effectively checks that the function gives the same result across the two time zones being compared, for any given instant within the interval.

It's possible for a time zone to have a transition from one ZoneInterval to another which doesn't adjust the offset: it might just change the name, or the balance between standard offset to daylight saving offset. (As an example, at midnight local time on October 27th 1968, the Europe/London time zone went from a standard offset of 0 and a daylight saving offset of 1 hour to a standard offset of 1 and a daylight saving offset of 0... which left the clocks unchanged.) This transition is irrelevant to the default options, so the two zone intervals involved are effectively coalesced.

The options available change what sort of comparison is performed - which can also change which zone intervals can be coalesced. For example, by specifying just the MatchAllTransitions option, you would indicate that even though you don't care about the name within a zone interval or how the wall offset is calculated, you do care about the fact that there was a transition at all, and when it occurred. With that option enabled, zone intervals are never coalesced and the transition points within the operating interval are checked.

Similarly, the MatchStartAndEndTransitions option is the only one where instants outside the operating interval are relevant. For example, consider a comparer which operates over the interval [2000-01-01T00:00:00Z, 2011-01-01T00:00:00Z). Normally, anything that happens before the year 2000 (UTC) would be irrelevant - but with this option enabled, the transitions of the first and last zone intervals are part of the comparison... so if one time zone has a zone interval 1999-09-01T00:00:00Z to 2000-03-01T00:00:00Z and the other has a zone interval 1999-10-15T00:00:00Z to 2000-03-01T00:00:Z, the two zones would be considered unequal, despite the fact that the only instants observing the difference occur outside the operating interval.

Version Information

Available since: 1.1.0
Supported in the PCL? Yes
See Also