Book Image

Hands-On Design Patterns with Delphi

By : Primož Gabrijelčič
Book Image

Hands-On Design Patterns with Delphi

By: Primož Gabrijelčič

Overview of this book

Design patterns have proven to be the go-to solution for many common programming scenarios. This book focuses on design patterns applied to the Delphi language. The book will provide you with insights into the language and its capabilities of a runtime library. You'll start by exploring a variety of design patterns and understanding them through real-world examples. This will entail a short explanation of the concept of design patterns and the original set of the 'Gang of Four' patterns, which will help you in structuring your designs efficiently. Next, you'll cover the most important 'anti-patterns' (essentially bad software development practices) to aid you in steering clear of problems during programming. You'll then learn about the eight most important patterns for each creational, structural, and behavioral type. After this, you'll be introduced to the concept of 'concurrency' patterns, which are design patterns specifically related to multithreading and parallel computation. These will enable you to develop and improve an interface between items and harmonize shared memories within threads. Toward the concluding chapters, you'll explore design patterns specific to program design and other categories of patterns that do not fall under the 'design' umbrella. By the end of this book, you'll be able to address common design problems encountered while developing applications and feel confident while building scalable projects.
Table of Contents (18 chapters)
Title Page
Copyright and Credits
About Packt

Dependency injection

The Dependency injection (DI) pattern  is the first of many design patterns in this book that were not described in the Gang of Four book. There's a good reason for that. The DI pattern has sprung from the unit-testing movement, and in 1994, unit testing was not yet a thing. In that era, applications were treated as monolithic blocks of code, and not as collections of loosely connected units.

DI works by changing responsibility for object creation. In classical OOP, each object creates new objects that it needs for functioning. This makes the program very rigid and hard to test. DI turns this on its head. If an object needs other objects to functions, the owner (the code which creates the first objects) should also create these other objects and pass them to the first one. In short, a dependency injection says this: Don't create things yourself; your owner should provide them. This decouples objects (makes interconnections between them less tight) and simplifies testing...