* fix parsing token components with brackets without spaces * handle more edge cases * do not use force unwrap * use first/last instead of hasPrefix/hasSuffix * update CHANGELOG