现代IDL编程指南

5星 · 超过95%的资源需积分: 10 184 浏览量更新于2024-07-23 5 收藏 20MB PDF 举报

"Modern IDL: A Guide to IDL Programming by Michael Galloy" "Modern IDL" 是一本关于IDL（Interface Definition Language）高级编程的指南，由Michael Galloy撰写。这本书面向那些希望深入理解和掌握IDL语言的读者，提供了一系列全面且实用的编程技巧。通过本书，读者可以了解到现代IDL的最新特性和最佳实践，从而提升其在科研、数据处理和可视化等领域的应用能力。 IDL是一种广泛用于科学计算和数据处理的编程语言，尤其在天文学、地球科学和医学成像等领域有广泛应用。"Modern IDL" 涵盖了从基础语法到高级特性的全面内容，包括数据类型、控制结构、函数和模块的使用、错误处理、优化技术以及与其他编程语言的接口等。书中详细介绍了如何有效地利用IDL的强大功能，如高效的数据操作、自定义数据类型、面向对象编程、图形用户界面开发以及并发和并行编程等。此外，作者还分享了在实际项目中可能遇到的问题及其解决方案，帮助读者解决在编程过程中可能碰到的难题。在版权方面，作者强调未经出版商书面许可，不得复制或以任何形式传播书中的内容。书中可能提及的制造商和销售商的产品名称可能被用作商标，但作者和出版商均不对商标准确性承担责任。尽管作者和出版商在编写过程中已经尽可能地确保了信息的准确性和完整性，但他们对因使用书中信息而导致的任何错误、遗漏或损害不承担直接责任。这表明，虽然本书提供了丰富的知识，但读者在实际应用时仍需结合具体情况进行判断。此书特别献给Usenet新闻组comp.lang.idl-pvwave上的网友们，无论他们提出问题还是给出解答，都为作者提供了宝贵的学习资源。 "Modern IDL" 是一本为IDL程序员准备的详尽指南，旨在提升他们在现代科学计算环境中的编程技能，无论是初学者还是经验丰富的开发者，都能从中受益。通过阅读和实践，读者将能够更好地运用IDL来解决复杂的数据处理和分析问题。

Preface — How this book is organized

xiv

3. How this book is organized

Chapters 1–6 form an introduction to IDL while chapters 7–13 discuss more advanced topics. The appendices provide

additional reference information.

Chapter 1, Introduction to IDL [p. 1]

This chapter introduces IDL through simple examples. These examples are not intended to completely explain

the details of IDL, but to demonstrate the broad capabilities. If you are new to IDL, read at least the first

section.

Chapter 2, The IDL environment [p. 11]

This chapter gives details about the IDL environment, both the command line and Workbench. Topics include

executive commands, the preference system, the online help, and the contents of the IDL distribution.

Chapter 3, Variables [p. 25]

This chapter discusses IDL variables and basic operations on them. Strings, including regular expressions, are

discussed in detail. Other forms of memory such as pointers, objects, common blocks, and shared memory are

also discussed.

Chapter 4, File Input/Output [p. 75]

This chapter discusses reading and writing files as well as networking topics. The scientific data file formats

supported by IDL, particularly HDF 5, are discussed as well as parsing XML files.

Chapter 5, Programming [p. 107]

This chapter introduces techniques for putting code into files as routines, main-level programs, or batch files.

Error handling, debugging, and distributing an application are discussed.

Chapter 6, Direct graphics [p. 131]

The direct graphics chapter begins with a comparison of direct graphics versus object graphics. It also includes

basic concepts such as color and graphics windows. Examples of various types of visualizations are created.

Sending output to other graphics devices such as PostScript files is also discussed.

Chapter 7, Objects [p. 185]

This chapter introduces object-oriented programming in IDL though simple examples and an explanation of

IDL’s syntax, while more practical uses of objects are covered in later chapters.

Chapter 8, Widgets [p. 203]

This chapter describes how to create a GUI application in IDL. Reusable components called compound widgets

are also discussed, as well as more advanced object-oriented techniques.

Chapter 9, Object graphics [p. 237]

This chapter describes the object graphics system as well as providing examples of common visualization

types.

Chapter 10, Using the iTools [p. 309]

This chapter provides an introduction to using the iTools as well as an introduction to programmatically

controlling them.

Chapter 11, Performance [p. 323]

This chapter discusses how to measure and improve the performance of IDL code in terms of speed and

memory usage.

Preface — Example code

xvi

A few paths through this book intended to bring the reader up to speed to tackle particular problems are listed below:

1. Those new to IDL and looking to do ad hoc analysis on data and produce simple visualizations should read

Chapter 2, The IDL environment [p. 11], Chapter 3, Variables [p. 25], Chapter 4, File Input/Output [p.

75], and Chapter 6, Direct graphics [p. 131]. Chapter 10, Using the iTools [p. 309] might be useful for

the those who want to interactively analyze and visualize their data with a provided GUI application.

2. Those looking to do more involved processing should read Chapter 5, Programming [p. 107] in addition to

those listed above.

3. Those looking to create interactive applications should also read Chapter 8, Widgets [p. 203], in addition to

all of the above.

4. Those looking to create (possibly interactive) sophisticated 3-dimensional graphics should then read Chapter 7,

Objects [p. 185] and Chapter 9, Object graphics [p. 237]. Alternatively, it might be possible that

Chapter 10, Using the iTools [p. 309] provides a simpler, though difficult to customize, solution.

Those new to IDL should start with Section 1.1, “Getting starting with IDL” [p. 1].

4. Example code

The example programs and batch files containing all the code displayed at the command line from the book can be

downloaded at modernidl.idldev.com.

The example code was modified in some cases to produce the figures shown in this book, e.g., to produce PostScript

output approriate for the book. Style and formatting, such as white-on-black changed to black-on-white, was also

modified in some cases for presentation in book format.

5. Developers from other languages

This section describes IDL’s relation to several popular programming languages. For those familiar with C/C++, Java,

Fortran, or Python, it may be helpful to have a brief comparison of IDL with a known language.

Statically typed languages like C or C++ do not have anything like IDL’s interactive, interpreted environment with

dynamically typed variables. While IDL does allow techniques such as pointers and object-oriented programming, it has

a simplified syntax for them, particularly pointers. Techniques like looping over the elements of an array can be done

(slowly) in IDL, but vectorized techniques which operate on entire array of elements at a time are much more efficient

and approach the speed of well-crafted C code. IDL is written in C, so its default extension and embedding APIs are

provided in C. The CALL_EXTERNAL routine and Dynamically Loadable Modules (DLMs) provide techniques for

extending IDL, i.e., writing C code that can be called from within IDL. In the opposite direction, Callable IDL is a

technique for writing C programs which execute IDL code.

Java developers should note that while IDL allows object-oriented programming, it also allows standard procedures and

functions and that objects in IDL are not automatically garbage collected (although IDL 8.0 does have automatic garbage

collection, the HEAP_GC routine is not a replacement in previous versions of IDL). The Java bridges allow both IDL

code to instantiate Java objects and Java code to instantiate IDL objects.

Fortran code can be called from within IDL using the CALL_EXTERNAL function.

Preface — How to contact me

xvii

Python provides the most similar environment to IDL because of the ability to interact with the language at a command

line as well as write scripts. IDL lacks namespaces and important data structures such as lists and dicts (IDL 8.0 adds

basic list and hash support, but more advanced features like list comprehensions are not provided). Python named

arguments are similar to keywords in IDL. Both languages support object-oriented programming with a simple syntax

and rely on “duck typing”. The Python package NumPy (or the older numeric or numarray packages) provides many of

the same vectorized array processing capabilities provided in IDL. For those with existing Python code, the commercial

product Slither can instantiate Python objects from within IDL, while the open source pyIDL can call IDL code from

within Python. While Python advertises itself as coming with “batteries included,” this is not true for the scientific

programmer. A Python installation with capabilities equivalent to the IDL distribution would require many packages like

NumPy, SciPy, PyTables, matplotlib, VTK, etc.

Developers from other languages often ask whether IDL is column or row major. Memory layout of multi-dimensional

IDL arrays is contiguous by varying the first dimension, i.e., the indices of the first elements of a 3 by 4 array are:

[0, 0], [1, 0], [2, 0], [0, 1], ...

This issue is complicated by the fact that IDL uses image processing notation for labeling the dimensions of an array,

calling the first dimension columns and the second dimension rows. Since the order of the terms row and column are

switched from how they are used in mathematical notation or languages like C, it is clearer to just indicate that the

memory in IDL is contiguous by varying the first dimension. Using IDL’s terms, rows are stored in contiguous memory.

6. How to contact me

Errata and example code updates for the book will appear on the modernidl.idldev.com website. Please submit tickets

for possible errors in the book and accompanying code on this website as well.

Contact me directly at mgalloy@idldev.com. Follow my blog at michaelgalloy.com for notices about new editions and

other projects.

7. Acknowledgments

Thanks to the reviewers who read each one of the chapters as I finished them and made many useful suggestions

without which this book would have been impossible: Ron Kneusel, Dawn Lenz, Jim Pendleton, Mark Piper, and Paulo

Penteado.

Finally, thanks to the comp.lang.idl-pvwave newsgroup; several examples in this book were created originally to

answer questions posted on the newsgroup. I was constantly learning new things whether asking a question, answering a

question, or just reading.

Michael Galloy

Boulder, CO

April 2011

剩余463页未读，继续阅读

feixiang1983929

粉丝: 0
资源: 4

现代IDL编程指南

IDL使用演示

IDLtraining

董彦卿 idl电子书

envi找不到idl

怎么安装idl2cpp Ubuntu

java idl 使用场景

IDL转换工具是什么意思

idl8.4 安装教程

TAO_IDL 编译器

idl select

最新资源