No description available.

No description available.

No description available.

Per pastaruosius metus, lankstaus programavimo metodikos sulaukė daug dėmesio. Programų kodo pertvarkymas tapo viena iš labiausiai naudojamų praktikų, ypač ekstremaliame programavime. Todėl poreikis turėti programos kodo pertvarkymų įrankį išaugo iki tokio lygio, kad pertvarkymo įrankiai tapo reikalaujama priemone šiuolaikinėse programų kūrimo aplinkose. Tyrimo sritis – abstrakčios sintaksės medžių pertvarkymo algoritmai. Tyrimo objektas – abstrakčios sintaksės medžių pertvarkymo procesai skirtingose programavimo kalbose (C++ ir Java). Šio darbo tikslas – ištirti, kaip efektyviai būtų galima vykdyti pertvarkymus bendriniam AST, tam sukuriant pertvarkymų biblioteką ir atliekant eksperimentus įvairių kalbų atvejais. / Over the past years, agile development methodologies have attracted a lot of attention. Refactoring has become one of the most heavily used practices, especially in Extreme Programming. Therefore the need to have powerful refactoring tools has grown to such an extent, that there is a required feature for modern-day IDEs to have implemented refactoring tools. The aim of this work is to build a refactoring library for the generic abstract syntax tree and test the algorithms for speed. Generic abstract syntax tree (GAST) is a tree structure that can store elements of several programming language. Achieving this goal required doing certain tasks that you find in this document: analysis of technologies and existing software; several algorithms for refactoring; user need and specification of requirements; library model, expressed in UML diagrams; test plan and test procedure. Experiments have shown that it is necessary to improve the refactoring algorithms. But the library and the AST structure is worth further development.

Medžių tipo duomenų struktūras transformuojantys įrankiai naudojami išspręsti gausybę įvairaus tipo problemų: kompiliatorių, kodo optimizatorių, dokumentacijos ir kodo generavimo įrankių kūrimo ir pan. Bet nei vienas iš šių egzistuojančių įrankių nėra specialiai skirtas abstrakčios sintaksės medžių transformacijoms į nuo kalbos nepriklausomą formą, ar iš jo. Šis magistro darbas aprašo sistemą, kuri pateiktą konkretų abstrakčios sintaksės medį (abstrakčios sintaksės medį, kuris išreiškia realios programavimo kalbos – C#, Java, Ruby ar pan. – programinio kodo struktūrą) transformuoja į nuo kalbos nepriklausomą abstrakčios sintaksės medį, bei atvirkščiai. Taisyklės, aprašančios vieno medžio transformavimą į kitą, yra aprašytos XML formatu paremtuose failuose, kurių struktūra aiški ir paprasta. Sistemą galima naudoti kaip nepriklausomą komponentą arba kaip įrankį trečiųjų šalių programinėje įrangoje. Tiek sistemos pagrindinės bibliotekos, tiek jas naudojantys įrankiai yra parašyti Java programavimo kalba. Sistemos tyrimu siekta įrodyti du dalykus: kad sistema gali teisingai transformuoti pateiktąjį konkretų abstrakčios sintaksės medį į jį atitinkantį bendrinį medį, ir atvirkščiai; ir kad transformavimo vykdymo laikas yra leistinose ribose. Abiejų tikslų buvo pasiekta. Remiantis šiais rezultatais prieita išvados, jog sistema yra tinkamas kandidatas tapti pagrindu trečiųjų šalių įrankiams, atliekantiems programinio kodo migravimą, pertvarkymus (angl. „refactoring“), modelių... [toliau žr. visą tekstą] / Tools for transforming tree-like data structures are used for solving a range of different problems, such as creation of compilers, code optimizers, documentation and code generation tools, etc. But none of these existing tools seem to be specialized in transforming abstract syntax trees to a language-independent format or from it. This master‘s thesis describes a system that takes concrete abstract syntax trees (ASTs that represent the structure of code written in a real programming language, such a C#, Java, Ruby, etc.) and transforms them into language-independent abstract syntax trees, and vice versa. Rules for mapping one AST to the other are defined in XML-based configuration files that have a simple and clear structure. The system can be used as a stand-alone tool or used as component in third party software. Both the system’s core libraries and tools are written in Java. The purpose of the system’s investigation was to prove two things: that the system can correctly transform a given specific AST to an equivalent generic AST, and vise versa; and that the transformation execution time is within an acceptable range. Both goals were met. This led to a conclusion that the system is a suitable candidate for use by third party tools that handle code migration, refactoring, model transformation and other types of similar tasks.

Operatorless Prolog text is LL(1) in nature and any standard LL parser generator tool can be used to parse it. However, the Prolog text that conforms to the ISO Prolog standard allows the definition of dynamic operators. Since Prolog operators can be defined at run-time, operator symbols are not present in the grammar rules of the language. Unless the parser generator allows for some flexibility in the specification of the grammar rules, it is very difficult to generate a parser for such text. In this thesis we discuss the existing parsing methods and their modified versions to parse languages with dynamic operator capabilities. Implementation details of a parser using Javacc as a parser generator tool to parse standard Prolog text is provided. The output of the parser is an “Abstract Syntax Tree” that reflects the correct precedence and associativity rules among the various operators (static and dynamic) of the language. Empirical results are provided that show that a Prolog parser that is generated by the parser generator like Javacc is comparable in efficiency to a hand-coded parser.

This thesis describes the Drag language. Drag is a general purpose, gradually typed, lexically scoped, and multi-paradigm pro- gramming language. The essence of Drag is to build the abstract syntax trees of the programs directly and interactively. Our work includes the language specification and a prototype program. The language specification focuses on the syntax, the semantic model, and the type system. The prototype consists of an interactive editor and a compiler that targets several plat- forms, among which we focus on the LLVM platform in this thesis.

<p>A complete Java interface to OpenModelica has been created, supporting both standard Modelica and the metamodeling extensions in MetaModelica. It is bidirectional, and capable of passing both standard Modelica data types, as well as abstract syntax trees and list structures to and from Java and process them in either Java or the OpenModelica Compiler.It currently uses the existing CORBA interface as well as JNI for standard Modelica. It is also capable of automatically generating the Java classes corresponding to MetaModelica code.This interface opens up increased possibilities for tool integration between OpenModelica and Java-based tools, since for example models or model fragments can be extracted from OpenModelica, processed in a Java tool, and put back into the main model representation in OpenModelica.</p><p>A first version text generation template language for MetaModelica is also presented. The goal for such a language is the ability to create a more concise and readablecode when translating an abstract syntax tree (AST) to text.</p>

<p>An unavoidable amount of money is spent on maintaining existing software systems today. Software maintenance cost generally higher than development cost of the system therefore lowering maintenance cost is highly appreciated in software industry.</p><p>A significant part of maintenance activities is related to repeating the investigation of problems and applying repeated solutions several times. A software system may contain a common bug in several different places and it might take extra effort and time to fix all existences of this bug. This operation commonly increases the cost of Software Maintenance Activities.</p><p>Detecting duplicate code fragments can significantly decrease the time and effort therefore the maintenance cost. Clone code detection can be achieved via analyzing the source code of given software system. An abstract syntax tree based clone detector for java systems is designed and implemented through this study.</p><p>This master thesis examines a software engineering process to create an abstract syntax tree based clone detector for the projects implemented in Java programming language.</p>

A complete Java interface to OpenModelica has been created, supporting both standard Modelica and the metamodeling extensions in MetaModelica. It is bidirectional, and capable of passing both standard Modelica data types, as well as abstract syntax trees and list structures to and from Java and process them in either Java or the OpenModelica Compiler.It currently uses the existing CORBA interface as well as JNI for standard Modelica. It is also capable of automatically generating the Java classes corresponding to MetaModelica code.This interface opens up increased possibilities for tool integration between OpenModelica and Java-based tools, since for example models or model fragments can be extracted from OpenModelica, processed in a Java tool, and put back into the main model representation in OpenModelica. A first version text generation template language for MetaModelica is also presented. The goal for such a language is the ability to create a more concise and readablecode when translating an abstract syntax tree (AST) to text.