Sergey V. Zelenov

Test Generation for Compilers and Other Formal Text Processors

A concept of automated test suites generation for testing compilers and other formal text processors is suggested. An approach based on the generation of tests from models is used.

Automatic Test Generation for Model-Based Code Generators

This paper presents a novel testing approach for model-based design tools, termed GraphOTK, and applies it to the optimizing component of a code generator for Simulink and Stateflow models.

Automated generation of positive and negative tests for parsers

In this paper we describe a specification-based approach to automated generation of both positive and negative test sets for parsers. We propose coverage criteria definitions for such test sets and algorithms for generation of the test sets with respect to proposed coverage criteria. We also present practical results of the technique application to testing syntax analyzers of several languages including C and Java.

Model-based testing of optimizing compilers

We describe a test development method, named OTK1, that is aimed at optimizing compiler testing. The OTK method is based on constructing a model of optimizers input data. The method allows developing tests targeted to testing a chosen optimizer. A formal data model is constructed on the basis of an abstract informal description of an algorithm of the optimizer under test. In the paper, we consider in detail the process of analyzing an optimization algorithm and building a formal model. We also consider in outline the other part of the method, test selection and test running. The OTK method has been successfully applied in several case studies, including test development for several different optimizing compilers for modern architectures.

Generation of Positive and Negative Tests for Parsers

A methodology for automatic positive and negative test set generation for testing parsers is described. Coverage criteria for such test sets based on the model approach to testing are proposed. Methods for the generation of test sets satisfying these criteria are discussed. Results of the application of the proposed methodology for testing parsers for various languages including C and Java are presented.

Using abstract models for the generation of test data with a complex structure

An approach to the automatic generation of test data having a complex structure (such as XML documents, programs in various programming languages, and the like) is presented. It is based on abstract models that represent various views of the structure of the desired data. The approach enables one to generate small sets of test data for testing the functionality of the target system. The use of abstract models makes configuration of the generation procedure easy and clear; it also facilitates the maintenance and reuse of existing configurations of the test data generator. The approach was implemented in the test data generator called Pinery, which was successfully used in a number of projects including testing commercial C/C++ compilers.

Schedulability Analysis for Strictly Periodic Tasks in RTOS

A new look at the problem of constructing a scheduler in the case of a group of strictly periodic tasks is proposed. The structure of the system of periods is represented in terms of graph theory. A criterion for the existence of a conflict-free schedule based on this representation is obtained, and generic schemes of algorithms for constructing such a schedule are described. The proposed approach is illustrated by building schedules for a number of strictly periodic tasks.

Targeted data generation for application testing on databases

Applications that process large datasets are a widely used type of software. Particularly, these applications address data integration problems in the field of enterprise application integration. These problems are solved using special tool environments supporting the development, implementation, and monitoring of applications that implement the Extract, Transform, and Load (ETL) model. The specific feature of functional testing of such applications is that there are many possible combinations of input data. The approaches implemented in the data generation tools for functional testing of applications on databases, in the best case, are based on database and SQL-query schemes that are used in applications to be tested. These approaches ensure the coverage of functionality of an application under testing only by conducting an exhaustive search of possible combinations of input data. The approach to data generation proposed in this paper makes it possible to find a coverage of application functionality with a nearly optimal amount of data (one dataset for one functional branch of the application).

Directed Generation of Test Data for Static Semantics Checker

We present an automatic method, named SemaTESK, for generation of test sets for a translator front end. We focus on the validation and verification of static semantics checker. Most the know methods for semantics test generation produce test suites by filtering a pre-generated set of random texts in the target language. In contrast, SemaTESK allows to generate tests for context conditions directly. It significantly reduces generation time and allows reaching completeness criteria defined in the paper. The presented method to specify static semantics allows to formalize informal requirements described in normative documents (e.g. standard). The method includes SRL notation for compact formal specification of context conditions and STG tool for efficient generation of test suite from SRL specification.