testing.diff() function

testing.diff() produces a diff between two streams.

The function matches tables from each stream based on group keys. For each matched table, it produces a diff. Any added or removed rows are added to the table as a row. An additional string column with the name diff is created and contains a - if the row was present in the got table and not in the want table or + if the opposite is true.

diff() function emits at least one row if the tables are different and no rows if the tables are the same. The exact diff produced may change. diff() can be used to perform in-line diffs in a query.

Function type signature

(
    <-got: stream[A],
    want: stream[A],
    ?epsilon: B,
    ?nansEqual: C,
    ?verbose: D,
) => stream[{A with _diff: string}]

{{% caption %}} For more information, see Function type signatures. {{% /caption %}}

Parameters

got

Stream containing data to test. Default is piped-forward data (<-).

want

({{< req >}}) Stream that contains data to test against.

epsilon

Specify how far apart two float values can be, but still considered equal. Defaults to 0.000000001.

verbose

Include detailed differences in output. Default is false.

nansEqual

Consider NaN float values equal. Default is false.

Examples

• Output a diff between two streams of tables
• Return a diff between a stream of tables an the expected output

Output a diff between two streams of tables

import "sampledata"
import "testing"

want = sampledata.int()
got =
    sampledata.int()
        |> map(fn: (r) => ({r with _value: if r._value > 15 then r._value + 1 else r._value}))

testing.diff(got: got, want: want)

{{< expand-wrapper >}} {{% expand "View example input" %}}

Input data

{{% /expand %}} {{< /expand-wrapper >}}

Return a diff between a stream of tables an the expected output

import "testing"

want = from(bucket: "backup-example-bucket") |> range(start: -5m)

from(bucket: "example-bucket")
    |> range(start: -5m)
    |> testing.diff(want: want)