diff --git a/common/whiteList/whitelist.go b/common/whiteList/whitelist.go new file mode 100644 index 0000000..119d807 --- /dev/null +++ b/common/whiteList/whitelist.go @@ -0,0 +1,74 @@ +package whiteList + +import ( + "fmt" + "os" + "regexp" + "strings" + + "gemini-grc/config" + "gemini-grc/logging" + "git.antanst.com/antanst/xerrors" +) + +var whitelist []regexp.Regexp //nolint:gochecknoglobals + +func Initialize() error { + var err error + + // Initialize whitelist + if config.CONFIG.WhitelistPath != "" { + if err = loadWhitelist(config.CONFIG.WhitelistPath); err != nil { + return err + } + } + + return nil +} + +func loadWhitelist(filePath string) error { + if whitelist != nil { + return nil + } + + data, err := os.ReadFile(filePath) + if err != nil { + whitelist = []regexp.Regexp{} + return xerrors.NewError(fmt.Errorf("could not load whitelist file: %w", err), 0, "", true) + } + + lines := strings.Split(string(data), "\n") + whitelist = []regexp.Regexp{} + + for _, line := range lines { + line = strings.TrimSpace(line) + if line == "" || strings.HasPrefix(line, "#") { + continue + } + regex, err := regexp.Compile(line) + if err != nil { + return xerrors.NewError(fmt.Errorf("could not compile whitelist line %s: %w", line, err), 0, "", true) + } + whitelist = append(whitelist, *regex) + } + + if len(whitelist) > 0 { + logging.LogInfo("Loaded %d whitelist entries", len(whitelist)) + } + + return nil +} + +func Shutdown() error { + return nil +} + +// IsWhitelisted checks if the URL matches any whitelist pattern +func IsWhitelisted(u string) bool { + for _, v := range whitelist { + if v.MatchString(u) { + return true + } + } + return false +} diff --git a/common/whiteList/whitelist_test.go b/common/whiteList/whitelist_test.go new file mode 100644 index 0000000..a55aa1d --- /dev/null +++ b/common/whiteList/whitelist_test.go @@ -0,0 +1,87 @@ +package whiteList + +import ( + "os" + "regexp" + "testing" + + "gemini-grc/config" +) + +func TestIsWhitelisted(t *testing.T) { + // Set up a test whitelist + whitelist = []regexp.Regexp{ + *regexp.MustCompile(`^gemini://example\.com`), + *regexp.MustCompile(`^gemini://test\.org/path`), + } + + testCases := []struct { + url string + expected bool + }{ + {"gemini://example.com", true}, + {"gemini://example.com/path", true}, + {"gemini://test.org", false}, + {"gemini://test.org/path", true}, + {"gemini://test.org/path/subpath", true}, + {"gemini://other.site", false}, + } + + for _, tc := range testCases { + result := IsWhitelisted(tc.url) + if result != tc.expected { + t.Errorf("IsWhitelisted(%s) = %v, want %v", tc.url, result, tc.expected) + } + } +} + +func TestLoadWhitelist(t *testing.T) { + // Create a temporary whitelist file + content := `# This is a test whitelist +^gemini://example\.com +^gemini://test\.org/path +` + tmpfile, err := os.CreateTemp("", "whitelist") + if err != nil { + t.Fatal(err) + } + defer os.Remove(tmpfile.Name()) + + if _, err := tmpfile.Write([]byte(content)); err != nil { + t.Fatal(err) + } + if err := tmpfile.Close(); err != nil { + t.Fatal(err) + } + + // Reset whitelist + whitelist = nil + + // Set up configuration to use the temporary file + oldPath := config.CONFIG.WhitelistPath + config.CONFIG.WhitelistPath = tmpfile.Name() + defer func() { + config.CONFIG.WhitelistPath = oldPath + }() + + // Load whitelist from the file + err = loadWhitelist(tmpfile.Name()) + if err != nil { + t.Fatalf("loadWhitelist() error = %v", err) + } + + // Check if whitelist was loaded correctly + if len(whitelist) != 2 { + t.Errorf("loadWhitelist() loaded %d entries, want 2", len(whitelist)) + } + + // Test a whitelisted URL + if !IsWhitelisted("gemini://example.com") { + t.Error("IsWhitelisted(\"gemini://example.com\") = false, want true") + } + + // Test a URL in a whitelisted path + if !IsWhitelisted("gemini://test.org/path/subpage.gmi") { + t.Error("IsWhitelisted(\"gemini://test.org/path/subpage.gmi\") = false, want true") + } +}