Pollak D. Beginning Scala
Подождите немного. Документ загружается.
■CONTENTS
ix
■CHAPTER 4 Fun with Functions, and Never Having to Close That
JDBC Connection
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
A Function Is an Instance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Passing Functions As Parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Partial Application and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Functions and Type Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Functions Bound to Variables in Scope . . . . . . . . . . . . . . . . . . . . . . 100
Putting Functions in Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Functions and Interactive Applications. . . . . . . . . . . . . . . . . . . . . . . 104
Building New Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Call-by-Name, Call-by-Value, and General Laziness. . . . . . . . . . . . . . . . 107
Build Your Own Control Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
JDBC Looping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
■CHAPTER 5 Pattern Matching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Basic Pattern Matching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Matching Any Type. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
More Pattern Matching. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Testing Data Types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Case Classes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Pattern Matching in Lists. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Pattern Matching and Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Nested Pattern Matching in Case Classes . . . . . . . . . . . . . . . . . . . . 126
Examining the Internals of Pattern Matching. . . . . . . . . . . . . . . . . . 128
Pattern Matching As Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Object-Oriented and Functional Tensions. . . . . . . . . . . . . . . . . . . . . . . . . 131
Shape Abstractions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
■CHAPTER 6 Actors and Concurrency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
A Different Approach to Concurrency: Look Ma, No Locks . . . . . . . . . . . 137
Defining an Actor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Defining an Actor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Would You Like State With That? . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Instantiating an Actor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Implementing a Listener . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
As an Actor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
19897FM.fm Page ix Monday, April 27, 2009 11:51 AM
x
■CONTENTS
Actors: More Than Write-Only . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
Beyond the Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Protocol Handler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Actors, Atomic Updates, and Transactions . . . . . . . . . . . . . . . . . . . 153
Composing Actors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
■CHAPTER 7 Traits and Types and Gnarly Stuff for Architects. . . . . . . . 171
Show Me Some Bad Java Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
Library Pimping, Implicit Conversions, and Vampires . . . . . . . . . . . . . . . 173
Library Pimping. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Inviting Vampires Into Your Code . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
Traits: Interfaces on Steroids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Traits and Class Hierarchies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Types—It’s Beyond Generic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
Table Code. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
The Column . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
The Table. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Concrete Columns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
Building a Query . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
Making Queries Type-Safe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Boilerplate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
Vending JDBC Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Making a Concrete Class That Implements Table. . . . . . . . . . . . . . 205
Type-Safe Fun. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Variance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Invariant Parameter Types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Covariant Parameter Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Contravariant Parameter Types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
Rules of Variance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Poignantly Killing Bunnies: Dwemthy’s Stairs . . . . . . . . . . . . . . . . . . . . . 214
The Immutable World. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
The Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Walking Through the Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
19897FM.fm Page x Monday, April 27, 2009 11:51 AM
■CONTENTS
xi
■CHAPTER 8 Parsers—Because BNF Is Not Just for Academics
Anymore
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
Higher-Order Functions and Combinators. . . . . . . . . . . . . . . . . . . . . . . . . 233
Higher-Order Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
Combinators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
The Calculator Parser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
JSON Parsing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
Twitter JSON Parsing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
Parsing the JSON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
■CHAPTER 9 Scaling Your Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Testing: Behavior-Driven Development. . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Specs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
Scala Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
Build Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Maven . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Ant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
Buildr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
Other Build Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
Team Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
Bringing New Team Members on Board . . . . . . . . . . . . . . . . . . . . . 272
Organizing Teams Around Scala. . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Best Practices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
Options Instead of Null Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
Focus on Immutability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
Keep Methods Short. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Refactor Mercilessly. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Compose Functions and Compose Classes . . . . . . . . . . . . . . . . . . . 278
19897FM.fm Page xi Monday, April 27, 2009 11:51 AM
xii
■CONTENTS
Selling Scala in Your Organization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
The Super-Soft Sell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
The Back-Office Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
The Big Sell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
My Direct Experience (YMMV) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
Selecting Team Members . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
Yeah, But How Compatible? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
A Tad Bit of History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Present-Day Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Popping the Stack to the CEO or Board . . . . . . . . . . . . . . . . . . . . . . 285
Why Is Scala Different? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
Anything You Can Do, I Can Do Better (in Scala). . . . . . . . . . . . . . . 287
Giving Scala a Lift. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
How Lift’s Comet Support Works . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
■INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
19897FM.fm Page xii Monday, April 27, 2009 11:51 AM
xiii
Foreword
A curious attribute of the Scala programming language is that it expends a lot of expressive
power in order to make the life of casual programmers simpler. For instance, Scala provides
sophisticated type parametrization and traits so that more advanced programmers can design
libraries which are at the same time general and easy to use.
Casual users can profit from these libraries long before they discover the full power of
Scala’s abstraction constructs. The language design avoids the temptation of simply defining
some handy primitives in the syntax, even if these primitives would be useful many times to
many users. Instead of fixed primitives, the language design tries very hard to provide general
construction principles, with which users can then define their own fundamental constructs, at
no loss in syntactic convenience. This idea ranges from simple things, such as being able to
define your own numeric data types, to full-blown domain-specific APIs such as Scala’s support
for concurrent actors.
There is hardly a better example of this approach than David Pollak’s Lift Web Framework
for Scala. It leverages the full expressive power of Scala to provide a framework that’s at the
same time simple to use, powerful, and secure. David has been one of the earliest adopters of
Scala. He has become a crystallization point for many important developments in the Scala
community. He fostered the vibrant Lift community, organized the first Scala Lift Off conference
in 2008, and promoted the adoption of Scala in many important industrial applications.
His experience puts him in a unique position to explain the language thoroughly and
competently. At the same time, David is a great communicator, and he has always the needs of
beginners to the language in mind. This book shows these two traits of his very well.
Written by an expert user of the language, it puts the focus on what’s important for a
beginner.
It moves you swiftly from simple examples to more complete applications, explaining both
language and core libraries. It manages to show that Scala is, fundamentally, a pretty simple
language to pick up, and at the same time a language with few limits if you want to progress.
I hope you’ll enjoy reading it as much as I did.
Martin Odersky
EPFL
Designer of Scala
19897FM.fm Page xiii Monday, April 27, 2009 11:51 AM
19897FM.fm Page xiv Monday, April 27, 2009 11:51 AM
xv
About the Author
■DAVID POLLAK has been writing commercial software since 1977. He
wrote the first real-time spreadsheet, Mesa, and the world’s highest-
performance spreadsheet engine, Integer. Since 1996, David has been
using and devising web development tools. As CTO of CMP Media,
David oversaw the first large-scale deployment of WebLogic. David was
CTO and VPE at Cenzic, a web application security company. David has
also developed numerous commercial projects in Ruby on Rails.
In 2007, David founded the Lift Web Framework open source project.
Lift is an expressive and elegant framework for writing web applications. Lift stresses the
importance of security, maintainability, scalability, and performance, while allowing for high
levels of developer productivity. Lift open source software is licensed under an Apache 2.0
license.
David is a consultant in San Francisco and works on Lift-based projects including Innova-
tion Games Online and ESME.
19897FM.fm Page xv Monday, April 27, 2009 11:51 AM
19897FM.fm Page xvi Monday, April 27, 2009 11:51 AM
xvii
About the Technical Reviewer
■PAUL SNIVELY is a 30-year veteran of the software industry with a passion
for programming languages. Studying computer science at Indiana
University hooked him on Scheme and functional programming.
The launch of the Macintosh in 1984 introduced him to objects via
Smalltalk, Object Pascal, and MacApp. He later discovered object-
functional programming in OCaml, and is thrilled to find a language as
powerful as Scala for the JVM. Paul is a former contributing editor for
MacTutor Magazine, served as a technical reviewer for Paradigms of
Artificial Intelligence Programming: Case Studies in Common Lisp by Peter Norvig (Morgan
Kaufmann, 1991) and is an editor for http://lambda-the-ultimate.org, the programming
languages weblog.
19897FM.fm Page xvii Monday, April 27, 2009 11:51 AM
19897FM.fm Page xviii Monday, April 27, 2009 11:51 AM