[extension/filestorage] replace path-unsafe characters in component n…

…ames

Fixes open-telemetry#3148

.chloggen/replace-slashes-in-component-names.yaml

@@ -0,0 +1,16 @@
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: breaking
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: extension/filestorage
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Replace path-unsafe characters in component names
+
+# One or more tracking issues related to the change
+issues: [3148]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:

extension/storage/filestorage/extension.go

@@ -7,6 +7,7 @@ import (
 	"context"
 	"fmt"
 	"path/filepath"
+	"regexp"
 
 	"go.opentelemetry.io/collector/component"
 	"go.opentelemetry.io/collector/extension"
@@ -49,7 +50,7 @@ func (lfs *localFileStorage) GetClient(_ context.Context, kind component.Kind, e
 	} else {
 		rawName = fmt.Sprintf("%s_%s_%s_%s", kindString(kind), ent.Type(), ent.Name(), name)
 	}
-	// TODO sanitize rawName
+	rawName = sanitize(rawName)
 	absoluteName := filepath.Join(lfs.cfg.Directory, rawName)
 	client, err := newClient(lfs.logger, absoluteName, lfs.cfg.Timeout, lfs.cfg.Compaction)
 
@@ -84,3 +85,18 @@ func kindString(k component.Kind) string {
 		return "other" // not expected
 	}
 }
+
+// sanitize replaces characters in name that are not safe in a file path
+func sanitize(name string) string {
+	// Specify unsafe characters by a negation of allowed characters:
+	// - uppercase and lowercase letters A-Z, a-z
+	// - digits 0-9
+	// - dot `.`
+	// - hyphen `-`
+	// - underscore `_`
+	// - tilde `~`
+	unsafeCharactersRegex := regexp.MustCompile(`[^A-Za-z0-9.\-_~]`)
+
+	// Replace all characters other than the above with the tilde `~`
+	return unsafeCharactersRegex.ReplaceAllString(name, "~")
+}

extension/storage/filestorage/extension_test.go

@@ -11,6 +11,7 @@ import (
 	"sync"
 	"testing"
 
+	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 	"go.opentelemetry.io/collector/component"
 	"go.opentelemetry.io/collector/extension/experimental/storage"
@@ -185,6 +186,57 @@ func TestTwoClientsWithDifferentNames(t *testing.T) {
 	require.Equal(t, myBytes2, data)
 }
 
+func TestSanitize(t *testing.T) {
+	testCases := []struct {
+		name          string
+		componentName string
+		sanitizedName string
+	}{
+		{
+			name:          "safe characters",
+			componentName: `.UPPERCASE-lowercase_1234567890~`,
+			sanitizedName: `.UPPERCASE-lowercase_1234567890~`,
+		},
+		{
+			name:          "unsafe characters",
+			componentName: `slash/backslash\colon:asterisk*questionmark?quote'doublequote"angle<>pipe|exclamationmark!percent%space `,
+			sanitizedName: `slash~backslash~colon~asterisk~questionmark~quote~doublequote~angle~~pipe~exclamationmark~percent~space~`,
+		},
+	}
+
+	for _, testCase := range testCases {
+		t.Run(testCase.name, func(t *testing.T) {
+			assert.Equal(t, testCase.sanitizedName, sanitize(testCase.componentName))
+		})
+	}
+}
+
+func TestComponentNameWithUnsafeCharacters(t *testing.T) {
+	ctx := context.Background()
+
+	tempDir := t.TempDir()
+
+	f := NewFactory()
+	cfg := f.CreateDefaultConfig().(*Config)
+	cfg.Directory = tempDir
+
+	extension, err := f.CreateExtension(context.Background(), extensiontest.NewNopCreateSettings(), cfg)
+	require.NoError(t, err)
+
+	se, ok := extension.(storage.Extension)
+	require.True(t, ok)
+
+	client, err := se.GetClient(
+		ctx,
+		component.KindReceiver,
+		newTestEntity("my/slashed/omponent*"),
+		"",
+	)
+
+	require.NoError(t, err)
+	require.NotNil(t, client)
+}
+
 func TestGetClientErrorsOnDeletedDirectory(t *testing.T) {
 	ctx := context.Background()