Course Overview
Overall Compiler Structure

(授業の概要; コンパイラ全体の仕組み)

Language Theory and Compilers

(言語理論とコンパイラ)

1st lecture, April 7, 2017

http://www.sw.it.aoyama.ac.jp/2017/Compiler/lecture1.html

Martin J. Dürst

duerst@it.aoyama.ac.jp, Building O, Room 529

Today's Schedule

Self-introduction
About this course
Difficulty of processing input
Compiler structure
Lexical analysis and parsing
Example of an automaton
Example of a formal language grammar

Self-Introduction

Born in Zürich, Switzerland
lic.oec.publ. from University of Zurich (修士相当、情報科学専攻、経営学と日本学副専攻)
文部省奨学金で来日、東京大学で理学博士 (情報科学専攻)
チューリッヒ大学情報科学科マルチメディア研究室主任助手
慶應義塾大学政策・メディア研究科特別研究助教授
Internationalization Activity Lead at W3C
Visiting Scientist at MIT
At Aoyama Gakuin University since 2005

TA: Sōichirō Yashiki (矢舗宗一郎、M1)

授業の位置付け

3 年前期
第二科目群 (選択必修、JE 科目)
ベースとなる科目:
- 計算機概論
- 情報数学
- 計算機実習 (プログラミング)

成績評価方法

(目安)

授業中のミニテスト: 20%
演習課題: 30%
期末試験: 50%

授業の目標

(言語) 理論と応用の関係の理解
ツールによる応用
- コンパイラ
- 文書処理
- 入力の解析
- データ形式の設計
- 小さい言語の設計と実装
  (ドメイン特化言語など)

The Importance of Compilers

Blog by Steve Yegge

Executive Summary:

If you don't know how compilers work, then you don't know how computers work.
If you're not 100% sure whether you know how compilers work, then you don't know how they work.

授業の進め方

前方着席
資料配付とウェブ公開
書き込みが重要
一部ノートパソコンで演習
宿題、レポート、ミニテスト、期末試験
出席しないと大損

Course Schedule

Schedule

参考書

(授業は言語理論とコンパイラ両方をカバーするが、参考書はそれぞれ片方に集中)

Course Contents

	Theory	Compilers	Other applications
Front end	language theory, automata	lexical analysis, parsing	regular expressions, text/data formats
Back end		optimization, code generation

Concentrate on input, not output
Use various representations from theory to application

Example of Difference between Input and Output

(Computer Practice I, problems 04A1 and 04C1)

Character itself (internal representation)	Escaping in HTML/XML (external representation)
`'`	`'`
`"`	`"`
`<`	`<`
`>`	`>`
`&`	`&`

→: output
←: input
Which direction is more difficult?

Difficulties for Input

Not structured (just a sequence of bytes/characters)
Anything goes (including errors)

Deciding whether some input is correct or not is a model for computation in general
Deciding whether input is correct is equivalent to recognition

The Function of a Compiler

Bridge between software and hardware

Input: Program that can be understood by humans
- Language: High-level program language
- Medium: source (file/program)
Output: Program that can be understood by a machine
- Language: assembly language, machine language
- Medium: object code、machine code

Example Compiler Input/Output

Input fragment:

sum += price * 25;

Output (assembly language):

LOAD   R1, price  ; load from price into R1 (register 1)
CONST  R2, 25     ; set constant 25 into R2 (register 2)
MUL    R1, R1, R2 ; set the multiple of R1 and R2 into R1
LOAD   R2, sum    ; load from sum into R2
ADD    R2, R1, R2 ; set the sum of R1 and R2 into R2
STORE  sum, R2    ; store the contents of R2 into sum

Logical Structure of a Compiler

[preprocessor]
Lexical analysis
Parsing (syntax analysis)
Semantic analysis
Optimization (or 5)
Code generation (or 4)
[assembler]
[linker, loader]

Compiler Types and Related Software

One pass compiler
X-pass compiler (x between 1 and 70 (IBM's PL/1 compiler in the 1970ies)
Cross-compiler (e.g. compiling on PC for smartphone)
Dynamic/just-in-time (JIT) compiler
Preprocessor (runs before the compiler)
Interpreter (e.g. for Ruby)

Example of Lexical Analysis

Fragment of input program (sequence of characters):

;

Output (sequence of tokens):

id("sum"), plusequal, id("price"), asterisk, int(25), semicolon

Overview of Lexical Analysis

Convert a sequence of characters to a sequence of tokens
Several characters become one token
Similar to finding words in a natural language text
Tokens can have attributes
Examples: id (identifier, with name)、int (integer, with value)
Many token names are derived from the symbol shape
(example: * is asterisk, not mult, because it can also be used for pointers)
White space is used during lexical analysis, but then discarded

Example of Parsing

Input (sequence of tokens):

id("sum"), plusequal, id("price"), asterisk, int(25), semicolon

Output (syntax tree):

Details of Syntax Trees

Tokens become nodes in the syntax tree
For expressions:
- Operators become parents (and represent the operation)
- Operands (including subexpressions) become children
- The structure of the tree shows priority and associativity of operations
  (operations at the bottom are evaluated first)
Abstract constructs (e.g. statement,...) are added as parents
Parentheses and other separators (,, ;,...) are used during parsing, but then discarded

More Examples

price = pretax / 100 * (108 - discount);

if (a > 5)
    b = 15;

Example of an Automaton

Very simple automatic vending machine:

Input: 50 Yen coins
Output: Water bottle (price: 150 Yen)

State transition diagram:

Language and Grammar

Language is defined on two layers:
- Structure (syntax)
- Meaning (semantics)
Grammar defines (restricts) the structure of a language
Example (simple imperative sentences):
eat bread、read books、play music
Grammar for imperative sentences:
ImperativeSentence → Verb Noun
The grammar allows sentences such as
"read bread" or "play books",
but semantically, they are problematic

Grammar of Imperative Sentences

ImperativeSentence → Verb Noun
Verb → eat
Verb → read
Verb → play
Noun → bread
Noun → music
Noun → books

Homework / 宿題

Deadline: April 13, 2017 (Thursday), 19:00

Where to submit: Box in front of room O-529 (building O, 5th floor)

Format: A4 single page (using both sides is okay; NO cover page), easily readable handwriting (NO printouts), name (kanji and kana) and student number at the top right

Problem: For the one-line C program fragment below, and based on the examples given in this lecture, write down:

the result of lexical analysis
the result of parsing
the output of the compiler (in assumbly language; comments are not needed; use SUB for substraction, and DIV for division)

grade = english + math/2 - absent*5;

Glossary

lexical analysis: 字句解析
parsing, syntax analysis: 構文解析
automaton: オートマトン
formal language: 形式言語
grammar: 文法
executive summary: 役員 (時間がない人) のための要約
front end: フロントエンド
back end: バックエンド
optimization: 最適化
code generation: コード生成
regular expression: 正規表現
text format: 文書形式
data format: データ形式
internal representation: 内部表現
external representation: 外部表現
(e.g. face) recognition: (顔) 認識
high-level program language: 高級プログラム言語
source (file/program): ソース (ファイル・プログラム)、原始プログラム
object code: 目的プログラム
machine code: 実行プログラム
assembly language: アセンブリ言語
register: レジスタ
preprocessor: プリプロセッサ
semantic analysis: 意味解析
assembler: アセンブラ (アセンブリ言語を処理するソフト)
linker: リンカ
loader: ローダ
one pass compiler: ワンパス・コンパイラ
x-pass compiler: x-パス・コンパイラ
cross-compiler: クロスコンパイラ
dynamic/just-in-time (JIT) compiler: 動的コンパイラ
preprocessor: プリプロセッサ
interpreter: インタプリタ、通訳系
token: トークン、記号、符
natural language: 自然言語
attribute: 属性
identifier (発音: アイデンティファイア): 識別子
syntax tree: 構文木
operator: 演算子
operand: 被演算子
expression: 式
subexpression: 部分式
statement (of a program): 文
separators: 区切り記号
automatic vending machine: 自動販売機
state transition diagram: 状態遷移図
structure: 構造
syntax: 構文
semantics: 意味 (論)
imperative sentence: 命令文
verb: 動詞
noun: 名詞

Course Overview Overall Compiler Structure