app/vmselect: add sort_by_label(q, label) and sort_by_label_desc(q, label) functions

This is implementation of https://github.com/prometheus/prometheus/pull/1533 for VictoriaMetrics.

app/vmselect/promql/exec.go

@@ -71,7 +71,8 @@ func maySortResults(e metricsql.Expr, tss []*timeseries) bool {
 		return true
 	}
 	switch fe.Name {
-	case "sort", "sort_desc":
+	case "sort", "sort_desc",
+		"sort_by_label", "sort_by_label_desc":
 		return false
 	default:
 		return true

app/vmselect/promql/exec_test.go

@@ -1597,6 +1597,48 @@ func TestExecSuccess(t *testing.T) {
 		resultExpected := []netstorage.Result{r1, r2}
 		f(q, resultExpected)
 	})
+	t.Run(`sort_by_label()`, func(t *testing.T) {
+		t.Parallel()
+		q := `sort_by_label((
+			alias(1, "foo"),
+			alias(2, "bar"),
+		), "__name__")`
+		r1 := netstorage.Result{
+			MetricName: metricNameExpected,
+			Values:     []float64{2, 2, 2, 2, 2, 2},
+			Timestamps: timestampsExpected,
+		}
+		r1.MetricName.MetricGroup = []byte("bar")
+		r2 := netstorage.Result{
+			MetricName: metricNameExpected,
+			Values:     []float64{1, 1, 1, 1, 1, 1},
+			Timestamps: timestampsExpected,
+		}
+		r2.MetricName.MetricGroup = []byte("foo")
+		resultExpected := []netstorage.Result{r1, r2}
+		f(q, resultExpected)
+	})
+	t.Run(`sort_by_label_desc()`, func(t *testing.T) {
+		t.Parallel()
+		q := `sort_by_label_desc((
+			alias(1, "foo"),
+			alias(2, "bar"),
+		), "__name__")`
+		r1 := netstorage.Result{
+			MetricName: metricNameExpected,
+			Values:     []float64{1, 1, 1, 1, 1, 1},
+			Timestamps: timestampsExpected,
+		}
+		r1.MetricName.MetricGroup = []byte("foo")
+		r2 := netstorage.Result{
+			MetricName: metricNameExpected,
+			Values:     []float64{2, 2, 2, 2, 2, 2},
+			Timestamps: timestampsExpected,
+		}
+		r2.MetricName.MetricGroup = []byte("bar")
+		resultExpected := []netstorage.Result{r1, r2}
+		f(q, resultExpected)
+	})
 	t.Run(`a cmp scalar (leave MetricGroup)`, func(t *testing.T) {
 		t.Parallel()
 		q := `sort_desc((
@@ -5338,6 +5380,8 @@ func TestExecError(t *testing.T) {
 	f(`scalar()`)
 	f(`sort(1,2)`)
 	f(`sort_desc()`)
+	f(`sort_by_label()`)
+	f(`sort_by_label_desc()`)
 	f(`timestamp()`)
 	f(`vector()`)
 	f(`histogram_quantile()`)

app/vmselect/promql/transform.go

@@ -97,6 +97,8 @@ var transformFuncs = map[string]transformFunc{
 	"acos":               newTransformFuncOneArg(transformAcos),
 	"prometheus_buckets": transformPrometheusBuckets,
 	"histogram_share":    transformHistogramShare,
+	"sort_by_label":      newTransformFuncSortByLabel(false),
+	"sort_by_label_desc": newTransformFuncSortByLabel(true),
 }
 
 func getTransformFunc(s string) transformFunc {
@@ -1355,6 +1357,29 @@ func transformScalar(tfa *transformFuncArg) ([]*timeseries, error) {
 	return arg, nil
 }
 
+func newTransformFuncSortByLabel(isDesc bool) transformFunc {
+	return func(tfa *transformFuncArg) ([]*timeseries, error) {
+		args := tfa.args
+		if err := expectTransformArgsNum(args, 2); err != nil {
+			return nil, err
+		}
+		label, err := getString(args[1], 1)
+		if err != nil {
+			return nil, fmt.Errorf("cannot parse label name for sorting: %s", err)
+		}
+		rvs := args[0]
+		sort.SliceStable(rvs, func(i, j int) bool {
+			a := rvs[i].MetricName.GetTagValue(label)
+			b := rvs[j].MetricName.GetTagValue(label)
+			if isDesc {
+				return string(b) < string(a)
+			}
+			return string(a) < string(b)
+		})
+		return rvs, nil
+	}
+}
+
 func newTransformFuncSort(isDesc bool) transformFunc {
 	return func(tfa *transformFuncArg) ([]*timeseries, error) {
 		args := tfa.args
@@ -1367,7 +1392,7 @@ func newTransformFuncSort(isDesc bool) transformFunc {
 			b := rvs[j].Values
 			n := len(a) - 1
 			for n >= 0 {
-				if !math.IsNaN(a[n]) && !math.IsNaN(b[n]) {
+				if !math.IsNaN(a[n]) && !math.IsNaN(b[n]) && a[n] != b[n] {
 					break
 				}
 				n--
@@ -1375,11 +1400,10 @@ func newTransformFuncSort(isDesc bool) transformFunc {
 			if n < 0 {
 				return false
 			}
-			cmp := a[n] < b[n]
 			if isDesc {
-				cmp = !cmp
+				return b[n] < a[n]
 			}
-			return cmp
+			return a[n] < b[n]
 		})
 		return rvs, nil
 	}

docs/MetricsQL.md

@@ -53,6 +53,7 @@ This functionality can be tried at [an editable Grafana dashboard](http://play-g
   - `label_transform(q, label, regexp, replacement)` for replacing all the `regexp` occurences with `replacement` in the `label` values from `q`.
   - `label_value(q, label)` - returns numeric values for the given `label` from `q`.
 - `label_match(q, label, regexp)` and `label_mismatch(q, label, regexp)` for filtering time series with labels matching (or not matching) the given regexps.
+- `sort_by_label(q, label)` and `sort_by_label_desc(q, label)` for sorting time series by the given `label`.
 - `step()` function for returning the step in seconds used in the query.
 - `start()` and `end()` functions for returning the start and end timestamps of the `[start ... end]` range used in the query.
 - `integrate(m[d])` for returning integral over the given duration `d` for the given metric `m`.

lib/metricsql/transform.go

@@ -76,6 +76,8 @@ var transformFuncs = map[string]bool{
 	"acos":               true,
 	"prometheus_buckets": true,
 	"histogram_share":    true,
+	"sort_by_label":      true,
+	"sort_by_label_desc": true,
 }
 
 // IsTransformFunc returns whether funcName is known transform function.