---

Introduction

Contents:

Organization of This Document

Archives and serializations are two ways in which you can create architecture-independent byte streams of hierarchical data. Byte streams can then be written to a file or transmitted to another process, perhaps over a network. When the byte stream is decoded, the hierarchy is regenerated. Archives provide a detailed record of a collection of interrelated objects and values. Serializations record only the simple hierarchy of property-list values.

You should read this document to learn how to create and extract archived representations of object graphs.

Organization of This Document

This programming topic contains the following articles:

• “Object Graphs” introduces the concept of an object graph and discusses the two techniques for turning objects into byte streams: archives and serializations.

• “Archives” describes the different types of archive and archiver classes.

• “Creating an Archive” describes how to create an archive.

• “Decoding an Archive” describes how to decode an archive.

• “Encoding and Decoding Objects” describes how to implement the methods that allow an object to be encoded in and decoded from archives.

• “Encoding and Decoding C Data Types” describes how to encode and decode C data types that do not have convenience methods defined in the archive classes.

• “Forward and Backward Compatibility for Keyed Archives” provides some tips on how to make your classes more compatible with previous and future versions of your classes in keyed archives.

• “Legacy Support for Non-Keyed Archives” describes how to convert older classes to use keyed coding instead of non-keyed coding.

• “Subclassing NSCoder” provides some tips on how to create your own coder classes.

• “Serializations” discusses the behavior of serializations.

• “Serializing Objects” describes how to create and read serializations.

---