Allow downstream projects to use faster regexp (#10251)

* Add benchmark for FastRegexMatcher

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* Use modified regexp package with optimisations

See https://github.com/grafana/regexp/tree/speedup#readme

Includes the following changes proposed upstream:
* [regexp: allow patterns with no alternates to be one-pass](https://go-review.googlesource.com/c/go/+/353711)
* [regexp: speed up onepass prefix check](https://go-review.googlesource.com/c/go/+/354909)
* [regexp: handle prefix string with fold-case](https://go-review.googlesource.com/c/go/+/358756)
* [regexp: avoid copying each instruction executed](https://go-review.googlesource.com/c/go/+/355789)
* [regexp: allow prefix string anchored at beginning](https://go-review.googlesource.com/c/go/+/377294)

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

* Use regexp code identical to upstream Go

Change `grafana/regexp` import to use `main` branch.

This means Prometheus is not using the proposed optimisations, but
downstream users of Prometheus code are able to `replace` the library
with the `speedup` branch which does.

Signed-off-by: Bryan Boreham <bjboreham@gmail.com>

go.mod

@@ -31,6 +31,7 @@ require (
 	github.com/google/pprof v0.0.0-20211214055906-6f57359322fd
 	github.com/gophercloud/gophercloud v0.24.0
 	github.com/gorilla/mux v1.8.0 // indirect
+	github.com/grafana/regexp v0.0.0-20220202152315-e74e38789280
 	github.com/grpc-ecosystem/grpc-gateway v1.16.0
 	github.com/hashicorp/consul/api v1.12.0
 	github.com/hashicorp/go-hclog v0.12.2 // indirect

go.sum

@@ -672,6 +672,8 @@ github.com/gorilla/mux v1.8.0/go.mod h1:DVbg23sWSpFRCP0SfiEN6jmj59UnW/n46BH5rLB7
 github.com/gorilla/websocket v0.0.0-20170926233335-4201258b820c/go.mod h1:E7qHFY5m1UJ88s3WnNqhKjPHQ0heANvMoAMk2YaljkQ=
 github.com/gorilla/websocket v1.4.0/go.mod h1:E7qHFY5m1UJ88s3WnNqhKjPHQ0heANvMoAMk2YaljkQ=
 github.com/gorilla/websocket v1.4.2/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/adAjf1fMHhE=
+github.com/grafana/regexp v0.0.0-20220202152315-e74e38789280 h1:MOND6wXrwVXEzmL2bZ+Jcbgycwt1LD5q6NQbqz/Nlic=
+github.com/grafana/regexp v0.0.0-20220202152315-e74e38789280/go.mod h1:M5qHK+eWfAv8VR/265dIuEpL3fNfeC21tXXp9itM24A=
 github.com/gregjones/httpcache v0.0.0-20180305231024-9cad4c3443a7/go.mod h1:FecbI9+v66THATjSRHfNgh1IVFe/9kFxbXtjV0ctIMA=
 github.com/grpc-ecosystem/go-grpc-middleware v1.0.0/go.mod h1:FiyG127CGDf3tlThmgyCl78X/SZQqEOJBCDaAfeWzPs=
 github.com/grpc-ecosystem/go-grpc-middleware v1.0.1-0.20190118093823-f849b5445de4/go.mod h1:FiyG127CGDf3tlThmgyCl78X/SZQqEOJBCDaAfeWzPs=

model/labels/regexp.go

@@ -14,9 +14,10 @@
 package labels
 
 import (
-	"regexp"
-	"regexp/syntax"
 	"strings"
+
+	"github.com/grafana/regexp"
+	"github.com/grafana/regexp/syntax"
 )
 
 type FastRegexMatcher struct {

model/labels/regexp_test.go

@@ -14,9 +14,10 @@
 package labels
 
 import (
-	"regexp/syntax"
+	"strings"
 	"testing"
 
+	"github.com/grafana/regexp/syntax"
 	"github.com/stretchr/testify/require"
 )
 
@@ -96,3 +97,42 @@ func TestOptimizeConcatRegex(t *testing.T) {
 		require.Equal(t, c.contains, contains)
 	}
 }
+
+func BenchmarkFastRegexMatcher(b *testing.B) {
+	var (
+		x = strings.Repeat("x", 50)
+		y = "foo" + x
+		z = x + "foo"
+	)
+	regexes := []string{
+		"foo",
+		"^foo",
+		"(foo|bar)",
+		"foo.*",
+		".*foo",
+		"^.*foo$",
+		"^.+foo$",
+		".*",
+		".+",
+		"foo.+",
+		".+foo",
+		".*foo.*",
+		"(?i:foo)",
+		"(prometheus|api_prom)_api_v1_.+",
+		"((fo(bar))|.+foo)",
+	}
+	for _, r := range regexes {
+		r := r
+		b.Run(r, func(b *testing.B) {
+			m, err := NewFastRegexMatcher(r)
+			require.NoError(b, err)
+			b.ResetTimer()
+			for i := 0; i < b.N; i++ {
+				_ = m.MatchString(x)
+				_ = m.MatchString(y)
+				_ = m.MatchString(z)
+			}
+		})
+
+	}
+}