Transcript CS2403 Programming Language Class Sildes

Implementation Issues
Cs2403 Programming Language
Spring 2005
Kun-Yuan Hsieh
PLLab, NTHU
Cs2403 Programming Languages
Why studying?
• Increased capacity to express ideas
– Language feature limits the form of algorithm you
construct
– Simulating other language’s feature to break this limit
• Improved ability of choosing right PL
– To give you the ability to make informed language
choices
• Increase ability to learn new PL
– Programming language is in a state of continuous
evolution
– Vocabulary and fundamental concepts helps you learn
PLLab, NTHU
Cs2403 Programming Languages
Why studying? (cont.)
• Better understanding of implementation
– To use a language in the way it was designed to
be used
• Overall advancement of computing
– Most popular languages are the best available?
– Better educated programmer leads to better
language population?
PLLab, NTHU
Cs2403 Programming Languages
Evaluation criteria
• Readability
– Important because maintenance is cost
– Overall simplicity since we tend to learn subset of it
• Feature multiplicity: more than one way to accomplish one
operation
– Count = count +1
– Count += 1
– Count ++
• operation overloading
– Reducing # of operations
• Less readable if carried too far
– Assembly language is simple but less readable
• Orthogonality
PLLab, NTHU
Cs2403 Programming Languages
Evaluation criteria
• Readability (cont.)
– Control statement
• Read from top to bottom is easier to understand than jump to
follow the execution order
– Data type and structure makes your meaning clear
• timeOut = 1 v.s. timeOut = true!!
– Syntax
– Identifier forms: using connotative names
– Special words: program appearance
– Form and meaning: statement indicates the purpose
PLLab, NTHU
Cs2403 Programming Languages
Evaluation criteria
• Writability
– What affects readability affect writability
– Simplicity and orthogonality
• Simplicity avoid misusing but may leads to undetected error
– Support for abstraction to define then use complicated
structures in a simple way
• Subprogram
• Tree (using point or array!!)
– Expressivity
PLLab, NTHU
Cs2403 Programming Languages
Evaluation criteria
• Reliability
– Performs to its specification under all
conditions
– Type checking
– Exception handling
– Aliasing
– Readability and writability
PLLab, NTHU
Cs2403 Programming Languages
Evaluation criteria
• Cost
– Training programmers to use language
– Writing programs
– Compiling programs, optimization
– Executing programs
– Language implementation system
• Java interpreter is free
– Reliability
– Maintaining programs
PLLab, NTHU
Cs2403 Programming Languages
Influences on language design
• Computer architecture
– Software executes on hardware
• Programming methodologies
– 1950s and early 1960s: machine efficiency
– Late 1960s: programming efficiency, readability,
better control structures
– Late 1970s: data abstraction
– Middle 1980s: object-oriented programming
PLLab, NTHU
Cs2403 Programming Languages
The von Neumann Arch (cont.)
Memory
(stores both instructions and data)
Results of
operations
Arithmetic and
logic unit
Instructions and data
Control unit
PLLab, NTHU
Cs2403 Programming Languages
Input and
output devices
A grid-computer arch
(non-Von Neumann)
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
CPU
PLLab, NTHU
Cs2403 Programming Languages
Categories & trade-off
• PL categories:
–
–
–
–
Imperative (C, Pascal, FORTRAN)
Functional (LISP, Scheme)
Logic (Prolog)
Object-Oriented (C++, C Sharp, Java)
• Design trade-off
– Reliability v.s. cost of execution
– Writability v.s. readability
– Flexibility v.s. safety
PLLab, NTHU
Cs2403 Programming Languages
Implementation methods
• Compilation (Pascal, C, C++, FORTRAN)
– Translate high-level program to machine code
– Slow translation
– Fast execution
• Interpretation (LISP, Scheme, scripts)
– No translation, interpreted by interpreter
– Slow execution (translated on execution)
– Easy to debug, run-time error refers to source level units
• Hybrid method (Java, CSharp, Perl)
– Translated to intermediate language (Java byte code, MSIL)
– Medium execution speed
PLLab, NTHU
Cs2403 Programming Languages
Compiler examples
PLLab, NTHU
Cs2403 Programming Languages
Interpreter v.s. Compiler
Input data
Source Program
Source Program
Interpreter
Compiler
PLLab, NTHU
Cs2403 Programming Languages
Results
Executable
(sample.exe)
Hybrid method
Java Source Code
Java Compiler
(private property)
Java Interpreter
on Windows
Java Byte Code
Java Interpreter
on Unix
(Stored in the disk &
Publicly available)
PLLab, NTHU
Cs2403 Programming Languages
Hybrid method example- .Net
C#
Visual
J# .NET
VB .NET
Compile
into
MSIL
MSIL
CLR do
this
Linux native
code
Windows
native code
Will Support
soon
Support
now
Mac OS
native code
Will Support
soon
PLLab, NTHU
Cs2403 Programming Languages
Helloworld.il
PLLab, NTHU
Cs2403 Programming Languages
Virtual computer concept
Virtual C++
Computer
C++
compiler
LISP interpreter
Operating system
FORTRAN
compiler
Macroinstruction
interpreter
Operating
System
Command
interpreter
Bare
machine
C compiler
Assembler
Virtual C
Computer
Ada compiler
…
PLLab, NTHU
Cs2403 Programming Languages
Levels of Programming Languages
High-level program
class Triangle {
...
float surface()
return b*h/2;
}
Low-level
program
ldloc.i4 b
ldloc.i4 h
mul
ldc.i4 2
div
ret
Executable Machine
code
0001001001000101001
0010011101100101011
01001...
PLLab, NTHU
Cs2403 Programming Languages
Compiler
phases
Source program
Lexical analyzer
Syntax analyzer
Semantic analyser
Symbol-table
manager
Error handler
Intermediate code
generator
Code optimizer
Code generator
PLLab,
NTHU
Target
program
Cs2403 Programming Languages
Lexical analysis
• Source program is read from left-to-right and group
into tokens
• Tokens: sequences of characters have a meaning
• Ex:
– a= b + c * 10
•
•
•
•
•
•
•
identifier a
assignment symbol =
identifier b
plus sign +
identifier c
multiplication sign *
number 10
PLLab, NTHU
Cs2403 Programming Languages
Syntax analysis
• Grouping the tokens into
grammatical phrases
(hierarchical)
• Usually presented as a parse
tree
• Expressed by recursive rules
– Any identifier is an
expression
– Any number is an
expression
– If exp1 and exp2 are
expressions, then so are
Assignment
statement
identifier
=
expression
expression
+
expression
expression
*
a
identifier
b
identifier
c
• exp1 + exp2
• exp1 * exp3
PLLab, NTHU
Cs2403 Programming Languages
expression
number
10
Semantics
• Specification of semantics is concerned with
specifying the “meaning” of well-formed
programs.
• Terminology:
– Expressions are evaluated and yield values (and may or
may not perform side effects)
– Commands are executed and perform side effects.
– Declarations are elaborated to produce bindings
• Side effects:
– change the values of variables
– perform input/output
PLLab, NTHU
Cs2403 Programming Languages
Syntax Analysis
Dataflow chart
Source Program
Stream of Characters
Scanner
Error Reports
Stream of “Tokens”
Parser
Error Reports
Abstract Syntax Tree
PLLab, NTHU
Cs2403 Programming Languages
Type Checking
• For most statically typed programming languages,
a bottom up algorithm over the AST:
• Types of expression AST leaves are known
immediately:
– literals => obvious
– variables => from the ID table
– named constants => from the ID table
• Types of internal nodes are inferred from the type
of the children and the type rule for that kind of
expression
PLLab, NTHU
Cs2403 Programming Languages