From df018368188a30571ea4ab02442c50bc55ddd354 Mon Sep 17 00:00:00 2001
From: Aliaksandr Valialkin <valyala@gmail.com>
Date: Fri, 17 Jul 2020 15:14:58 +0300
Subject: [PATCH] app/vmselect/promql: keep all labels for time series from
 `any()` call

---
 app/vmselect/promql/aggr.go      | 20 +++++++++++++++++---
 app/vmselect/promql/exec_test.go |  5 +++++
 docs/MetricsQL.md                |  2 +-
 3 files changed, 23 insertions(+), 4 deletions(-)

diff --git a/app/vmselect/promql/aggr.go b/app/vmselect/promql/aggr.go
index f53fd9f4b..4ddb96c60 100644
--- a/app/vmselect/promql/aggr.go
+++ b/app/vmselect/promql/aggr.go
@@ -43,7 +43,7 @@ var aggrFuncs = map[string]aggrFunc{
 	"bottomk_max":    newAggrFuncRangeTopK(maxValue, true),
 	"bottomk_avg":    newAggrFuncRangeTopK(avgValue, true),
 	"bottomk_median": newAggrFuncRangeTopK(medianValue, true),
-	"any":            newAggrFunc(aggrFuncAny),
+	"any":            aggrFuncAny,
 	"outliersk":      aggrFuncOutliersK,
 }
 
@@ -77,6 +77,8 @@ func removeGroupTags(metricName *storage.MetricName, modifier *metricsql.Modifie
 		metricName.RemoveTagsOn(modifier.Args)
 	case "without":
 		metricName.RemoveTagsIgnoring(modifier.Args)
+		// Reset metric group as Prometheus does on `aggr(...) without (...)` call.
+		metricName.ResetMetricGroup()
 	default:
 		logger.Panicf("BUG: unknown group modifier: %q", groupOp)
 	}
@@ -120,8 +122,20 @@ func aggrFuncExt(afe func(tss []*timeseries) []*timeseries, argOrig []*timeserie
 	return rvs, nil
 }
 
-func aggrFuncAny(tss []*timeseries) []*timeseries {
-	return tss[:1]
+func aggrFuncAny(afa *aggrFuncArg) ([]*timeseries, error) {
+	args := afa.args
+	if err := expectTransformArgsNum(args, 1); err != nil {
+		return nil, err
+	}
+	afe := func(tss []*timeseries) []*timeseries {
+		return tss[:1]
+	}
+	limit := afa.ae.Limit
+	if limit > 1 {
+		// Only a single time series per group must be returned
+		limit = 1
+	}
+	return aggrFuncExt(afe, args[0], &afa.ae.Modifier, limit, true)
 }
 
 func aggrFuncSum(tss []*timeseries) []*timeseries {
diff --git a/app/vmselect/promql/exec_test.go b/app/vmselect/promql/exec_test.go
index 654d7ad26..461953434 100644
--- a/app/vmselect/promql/exec_test.go
+++ b/app/vmselect/promql/exec_test.go
@@ -3825,6 +3825,10 @@ func TestExecSuccess(t *testing.T) {
 			Values:     []float64{10, 10, 10, 10, 10, 10},
 			Timestamps: timestampsExpected,
 		}
+		r.MetricName.Tags = []storage.Tag{{
+			Key:   []byte("foo"),
+			Value: []byte("bar"),
+		}}
 		resultExpected := []netstorage.Result{r}
 		f(q, resultExpected)
 	})
@@ -5613,6 +5617,7 @@ func TestExecError(t *testing.T) {
 	f(`sum()`)
 	f(`count_values()`)
 	f(`quantile()`)
+	f(`any()`)
 	f(`topk()`)
 	f(`topk_min()`)
 	f(`topk_max()`)
diff --git a/docs/MetricsQL.md b/docs/MetricsQL.md
index 0e905fc6b..ca69567fd 100644
--- a/docs/MetricsQL.md
+++ b/docs/MetricsQL.md
@@ -76,7 +76,7 @@ This functionality can be tried at [an editable Grafana dashboard](http://play-g
 - `median_over_time(m[d])` - calculates median values for `m` over `d` time window. Shorthand to `quantile_over_time(0.5, m[d])`.
 - `median(q)` - median aggregate. Shorthand to `quantile(0.5, q)`.
 - `limitk(k, q)` - limits the number of time series returned from `q` to `k`.
-- `any(q) by (x)` - returns any time series from `q` for each group in `x`. Note that `any()` removes all the labels except of those listed in `by (x)`.
+- `any(q) by (x)` - returns any time series from `q` for each group in `x`.
   Use `limitk(1, q)` if you need retaining all the labels from `q`.
 - `keep_last_value(q)` - fills missing data (gaps) in `q` with the previous non-empty value.
 - `keep_next_value(q)` - fills missing data (gaps) in `q` with the next non-empty value.