test(filebeat): compress journal test files (#42142)

* test(filebeat): compress journal test files

save around ~32MB from the module zip, relevant for program using beats as
a dependency.
Compress journal files with gzip and extract them to test tempdir

* lint: fix linter issues

filebeat/input/journald/input_filtering_test.go

@@ -22,14 +22,16 @@ package journald
 import (
 	"context"
 	"encoding/json"
-	"path"
+	"path/filepath"
 	"testing"
 	"time"
 
 	"github.com/elastic/elastic-agent-libs/mapstr"
 )
 
 func TestInputSyslogIdentifier(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "input-multiline-parser.journal.gz"))
+
 	tests := map[string]struct {
 		identifiers      []string
 		expectedMessages []string
@@ -53,7 +55,7 @@ func TestInputSyslogIdentifier(t *testing.T) {
 		t.Run(name, func(t *testing.T) {
 			env := newInputTestingEnvironment(t)
 			inp := env.mustCreateInput(mapstr.M{
-				"paths":              []string{path.Join("testdata", "input-multiline-parser.journal")},
+				"paths":              []string{out},
 				"syslog_identifiers": testCase.identifiers,
 			})
 
@@ -73,6 +75,7 @@ func TestInputSyslogIdentifier(t *testing.T) {
 }
 
 func TestInputUnits(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "input-multiline-parser.journal.gz"))
 	tests := map[string]struct {
 		units            []string
 		kernel           bool
@@ -110,7 +113,7 @@ func TestInputUnits(t *testing.T) {
 		t.Run(name, func(t *testing.T) {
 			env := newInputTestingEnvironment(t)
 			inp := env.mustCreateInput(mapstr.M{
-				"paths":  []string{path.Join("testdata", "input-multiline-parser.journal")},
+				"paths":  []string{out},
 				"units":  testCase.units,
 				"kernel": testCase.kernel,
 			})
@@ -131,6 +134,7 @@ func TestInputUnits(t *testing.T) {
 }
 
 func TestInputIncludeMatches(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "input-multiline-parser.journal.gz"))
 	tests := map[string]struct {
 		includeMatches   map[string]interface{}
 		expectedMessages []string
@@ -168,7 +172,7 @@ func TestInputIncludeMatches(t *testing.T) {
 		t.Run(name, func(t *testing.T) {
 			env := newInputTestingEnvironment(t)
 			inp := env.mustCreateInput(mapstr.M{
-				"paths":           []string{path.Join("testdata", "input-multiline-parser.journal")},
+				"paths":           []string{out},
 				"include_matches": testCase.includeMatches,
 			})
 
@@ -190,6 +194,7 @@ func TestInputIncludeMatches(t *testing.T) {
 // TestInputSeek test the output of various seek modes while reading
 // from input-multiline-parser.journal.
 func TestInputSeek(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "input-multiline-parser.journal.gz"))
 	// Uncomment the following line to see all logs during the test execution
 	// logp.DevelopmentSetup()
 	timeAfterFirstEvent := time.Date(2021, time.November, 22, 17, 10, 20, 0, time.UTC).In(time.Local)
@@ -256,13 +261,13 @@ func TestInputSeek(t *testing.T) {
 				if err := json.Unmarshal([]byte(testCase.cursor), &tmp); err != nil {
 					t.Fatal(err)
 				}
-				if err := store.Set("journald::testdata/input-multiline-parser.journal", tmp); err != nil {
+				if err := store.Set("journald::"+out, tmp); err != nil {
 					t.Fatal(err)
 				}
 			}
 
 			conf := mapstr.M{
-				"paths": []string{path.Join("testdata", "input-multiline-parser.journal")},
+				"paths": []string{out},
 			}
 
 			conf.DeepUpdate(testCase.config)

filebeat/input/journald/input_parsers_test.go

@@ -21,7 +21,7 @@ package journald
 
 import (
 	"context"
-	"path"
+	"path/filepath"
 	"testing"
 
 	"github.com/elastic/elastic-agent-libs/mapstr"
@@ -31,9 +31,11 @@ import (
 // it only tests a single parser, but that is enough to ensure
 // we're correctly using the parsers
 func TestInputParsers(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "ndjson-parser.journal.gz"))
+
 	env := newInputTestingEnvironment(t)
 	inp := env.mustCreateInput(mapstr.M{
-		"paths": []string{path.Join("testdata", "ndjson-parser.journal")},
+		"paths": []string{out},
 		"parsers": []mapstr.M{
 			{
 				"ndjson": mapstr.M{

filebeat/input/journald/input_test.go

@@ -20,13 +20,15 @@
 package journald
 
 import (
+	"compress/gzip"
 	"context"
 	_ "embed"
 	"encoding/json"
 	"fmt"
+	"io"
 	"os"
-	"path"
 	"path/filepath"
+	"strings"
 	"testing"
 
 	"github.com/stretchr/testify/require"
@@ -40,9 +42,11 @@ import (
 )
 
 func TestInputCanReadAllBoots(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "multiple-boots.journal.gz"))
+
 	env := newInputTestingEnvironment(t)
 	cfg := mapstr.M{
-		"paths": []string{path.Join("testdata", "multiple-boots.journal")},
+		"paths": []string{out},
 	}
 	inp := env.mustCreateInput(cfg)
 
@@ -54,6 +58,7 @@ func TestInputCanReadAllBoots(t *testing.T) {
 }
 
 func TestInputFieldsTranslation(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "input-multiline-parser.journal.gz"))
 	// A few random keys to verify
 	keysToCheck := map[string]string{
 		"systemd.user_unit": "log-service.service",
@@ -74,7 +79,7 @@ func TestInputFieldsTranslation(t *testing.T) {
 			env := newInputTestingEnvironment(t)
 
 			inp := env.mustCreateInput(mapstr.M{
-				"paths":                 []string{path.Join("testdata", "input-multiline-parser.journal")},
+				"paths":                 []string{out},
 				"include_matches.match": []string{"_SYSTEMD_USER_UNIT=log-service.service"},
 				"save_remote_hostname":  tc.saveRemoteHostname,
 			})
@@ -118,9 +123,10 @@ func TestInputFieldsTranslation(t *testing.T) {
 // __CURSOR - it is added to the registry and there are other tests for it
 // __MONOTONIC_TIMESTAMP - it is part of the cursor
 func TestCompareGoSystemdWithJournalctl(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "input-multiline-parser.journal.gz"))
 	env := newInputTestingEnvironment(t)
 	inp := env.mustCreateInput(mapstr.M{
-		"paths": []string{path.Join("testdata", "input-multiline-parser.journal")},
+		"paths": []string{out},
 		"seek":  "head",
 	})
 
@@ -185,6 +191,7 @@ func TestCompareGoSystemdWithJournalctl(t *testing.T) {
 }
 
 func TestMatchers(t *testing.T) {
+	out := decompress(t, filepath.Join("testdata", "matchers.journal.gz"))
 	// If this test fails, uncomment the following line to see the debug logs
 	// logp.DevelopmentSetup()
 	testCases := []struct {
@@ -274,7 +281,7 @@ func TestMatchers(t *testing.T) {
 		t.Run(tc.name, func(t *testing.T) {
 			env := newInputTestingEnvironment(t)
 			cfg := mapstr.M{
-				"paths":           []string{path.Join("testdata", "matchers.journal")},
+				"paths":           []string{out},
 				"include_matches": tc.matchers,
 			}
 			cfg.Update(mapstr.M(tc.confiFields))
@@ -322,3 +329,27 @@ func TestReaderAdapterCanHandleNonStringFields(t *testing.T) {
 		})
 	}
 }
+
+func decompress(t *testing.T, namegz string) string {
+	t.Helper()
+
+	ingz, err := os.Open(namegz)
+	require.NoError(t, err)
+	defer ingz.Close()
+
+	out := filepath.Join(t.TempDir(), strings.TrimSuffix(filepath.Base(namegz), ".gz"))
+
+	dst, err := os.Create(out)
+	require.NoError(t, err)
+	defer dst.Close()
+
+	gr, err := gzip.NewReader(ingz)
+	require.NoError(t, err)
+	defer gr.Close()
+
+	//nolint:gosec // this is used in tests
+	_, err = io.Copy(dst, gr)
+	require.NoError(t, err)
+
+	return out
+}