Dates


Dates


Bluge includes support indexing and searching DateTime (time.Time) values. It is important to understand that one should only use the Date field type when one needs to perform DateRange queries. If you only intend to perform exact matches, you should convert your date to a string representation and use the keyword analyzer.

Indexing

To start we’ll create a document with identifier a:

doc := bluge.NewDocument("a")

Now, let’s add a DateTime value to this document:

now := time.Now()
doc.AddField(bluge.NewDateTimeField("updated", now)

The first argument to the NewDateTimeField name is the field name age. The second argument is the time.Time value to be indexed.

NOTE: Because Bluge’s internal representation encodes signed 64-bit nanoseconds relative to the Unix epoch (1970-01-01 00:00:00 +0000 UTC), accurate encoding of time.Time values has nanosecond precision in the range (1677-09-21 00:12:43.145224192 +0000 UTC to 2262-04-11 23:47:16.854775807 +0000 UTC)

Searching

To search for documents containing a DateTime value within a range, we use the NewDateRangeQuery. Here is an example, if we wanted to search for values after 2020-01-01 and before 2021-01-01:

start2020, _ := time.Parse(time.RFC3339, "2020-01-01T00:00:00Z")
start2021, _ := time.Parse(time.RFC3339, "2021-01-01T00:00:00Z")
q := bluge.NewDateRangeQuery(start2020, start20201)

The first argument is the minimum date time (inclusive), and the second argument is the maximum date time (exclusive). If you need to control the inclusive/exclusive boundaries, use the longer form:

NewDateRangeInclusiveQuery(start, end time.Time, startInclusive, endInclusive bool)

Partially Open Ranges

To perform a DateRangeQuery with a partially open range, the time.Time zero-value can be used as a minimum or maximum value.