Rework parsing structure
This commit is contained in:
parent
a4ec29f084
commit
b9f419a59d
GPG key ID:
38297DE43F75FFE2
17 changed files with 246 additions and 278 deletions
|
|
@ -1,17 +0,0 @@
|
|||
package better_parser
|
||||
|
||||
import com.github.h0tk3y.betterParse.grammar.Grammar
|
||||
import com.github.h0tk3y.betterParse.grammar.parseToEnd
|
||||
import prolog.Program
|
||||
import prolog.ast.logic.Clause
|
||||
import prolog.ast.terms.Atom
|
||||
|
||||
class PrologParser {
|
||||
private val parser: Grammar<List<Clause>> = SimpleSourceParser() as Grammar<List<Clause>>
|
||||
|
||||
public fun parse(input: String) {
|
||||
val clauses: List<Clause> = parser.parseToEnd(input)
|
||||
|
||||
Program.load(clauses)
|
||||
}
|
||||
}
|
||||
|
|
@ -1,70 +0,0 @@
|
|||
package better_parser
|
||||
|
||||
import com.github.h0tk3y.betterParse.combinators.*
|
||||
import com.github.h0tk3y.betterParse.grammar.Grammar
|
||||
import com.github.h0tk3y.betterParse.lexer.literalToken
|
||||
import com.github.h0tk3y.betterParse.lexer.regexToken
|
||||
import com.github.h0tk3y.betterParse.parser.Parser
|
||||
import prolog.ast.logic.Fact
|
||||
import prolog.ast.arithmetic.Integer
|
||||
import prolog.ast.arithmetic.Float
|
||||
import prolog.ast.logic.Clause
|
||||
import prolog.ast.logic.LogicOperand
|
||||
import prolog.ast.logic.Rule
|
||||
import prolog.ast.terms.*
|
||||
import prolog.builtins.Conjunction
|
||||
import prolog.builtins.Disjunction
|
||||
|
||||
class PrologSourceParser : Grammar<List<Clause>>() {
|
||||
// Define the tokens
|
||||
private val atom by regexToken("[a-z][a-zA-Z0-9_]*")
|
||||
private val variable by regexToken("[A-Z][a-zA-Z0-9_]*")
|
||||
private val number by regexToken("-?[0-9]+(\\.[0-9]+)?")
|
||||
private val whitespace by regexToken("\\s+", ignore = true)
|
||||
|
||||
private val comma by literalToken(",")
|
||||
private val semicolon by literalToken(";")
|
||||
private val neck by literalToken(":-")
|
||||
private val lparen by literalToken("(")
|
||||
private val rparen by literalToken(")")
|
||||
private val dot by literalToken(".")
|
||||
|
||||
private val atomParser by atom use { Atom(text) }
|
||||
private val variableParser by variable use { Variable(text) }
|
||||
private val intParser by number use { Integer(text.toInt()) }
|
||||
private val floatParser by number use { Float(text.toFloat()) }
|
||||
private val numberParser by (intParser or floatParser)
|
||||
private val compoundTermParser by (atomParser and skip(lparen) and separated(
|
||||
atomParser or variableParser,
|
||||
comma
|
||||
) and skip(rparen)) use {
|
||||
CompoundTerm(t1, t2.terms)
|
||||
}
|
||||
|
||||
private val termParser: Parser<Term> by (numberParser or variableParser or compoundTermParser or atomParser)
|
||||
|
||||
private val logicOperandParser: Parser<LogicOperand> by (termParser or compoundTermParser or atomParser) map {
|
||||
it as LogicOperand
|
||||
}
|
||||
|
||||
private val conjunctionParser: Parser<Conjunction> by (logicOperandParser and comma and logicOperandParser) use {
|
||||
Conjunction(t1, t3)
|
||||
}
|
||||
private val disjunctionParser: Parser<Disjunction> by (logicOperandParser and semicolon and logicOperandParser) use {
|
||||
Disjunction(t1, t3)
|
||||
}
|
||||
|
||||
private val operatorParser: Parser<Operator> by (conjunctionParser or disjunctionParser)
|
||||
|
||||
private val headParser by (compoundTermParser or atomParser)
|
||||
private val bodyParser by (operatorParser or compoundTermParser or atomParser)
|
||||
|
||||
private val factParser by (headParser and dot) use { Fact(t1 as Head) }
|
||||
private val ruleParser by (headParser and neck and bodyParser and dot) use {
|
||||
Rule(t1 as Head, t3 as Body)
|
||||
}
|
||||
|
||||
private val clauseParser: Parser<Clause> by (factParser or ruleParser)
|
||||
|
||||
override val rootParser: Parser<List<Clause>> by zeroOrMore(clauseParser)
|
||||
}
|
||||
|
|
@ -1,67 +0,0 @@
|
|||
package better_parser
|
||||
|
||||
import com.github.h0tk3y.betterParse.combinators.*
|
||||
import com.github.h0tk3y.betterParse.grammar.Grammar
|
||||
import com.github.h0tk3y.betterParse.grammar.parser
|
||||
import com.github.h0tk3y.betterParse.lexer.Token
|
||||
import com.github.h0tk3y.betterParse.lexer.literalToken
|
||||
import com.github.h0tk3y.betterParse.lexer.regexToken
|
||||
import com.github.h0tk3y.betterParse.lexer.token
|
||||
import com.github.h0tk3y.betterParse.parser.Parser
|
||||
import prolog.ast.arithmetic.Float
|
||||
import prolog.ast.arithmetic.Integer
|
||||
import prolog.ast.terms.Atom
|
||||
import prolog.ast.terms.Structure
|
||||
import prolog.ast.terms.Term
|
||||
import prolog.ast.terms.Variable
|
||||
|
||||
open class SimplePrologParser : Grammar<Any>() {
|
||||
// Prolog tokens
|
||||
protected val nameToken: Token by regexToken("[a-z][a-zA-Z0-9_]*")
|
||||
protected val variableToken: Token by regexToken("[A-Z][a-zA-Z0-9_]*")
|
||||
|
||||
// Arithmetic tokens
|
||||
private val floatToken: Token by regexToken("-?[1-9][0-9]*\\.[0-9]+")
|
||||
private val integerToken: Token by regexToken("-?([1-9][0-9]*|0)")
|
||||
|
||||
// Special tokens
|
||||
protected val neck by literalToken(":-")
|
||||
protected val comma: Token by literalToken(",")
|
||||
protected val leftParenthesis: Token by literalToken("(")
|
||||
protected val rightParenthesis: Token by literalToken(")")
|
||||
protected val dot by literalToken(".")
|
||||
|
||||
// Ignored tokens
|
||||
protected val whitespace: Token by regexToken("\\s+", ignore = true)
|
||||
protected val singleLineComment: Token by regexToken("%[^\\n]*", ignore = true)
|
||||
protected val multiLineComment: Token by regexToken("/\\*.*?\\*/", ignore = true)
|
||||
|
||||
protected val dummy by token { _, _ -> -1 } use { throw IllegalStateException("This parser should not be used") }
|
||||
|
||||
// Prolog parsers
|
||||
protected val variable: Parser<Variable> by variableToken use { Variable(text) }
|
||||
protected val atom: Parser<Atom> by nameToken use { Atom(text) }
|
||||
protected val compound: Parser<Structure> by (atom and skip(leftParenthesis) and separated(
|
||||
parser(::term),
|
||||
comma,
|
||||
acceptZero = true
|
||||
) and skip(rightParenthesis)) use {
|
||||
Structure(t1, t2.terms)
|
||||
}
|
||||
|
||||
// Arithmetic parsers
|
||||
private val int: Parser<Integer> by integerToken use { Integer(text.toInt()) }
|
||||
private val float: Parser<Float> by floatToken use {
|
||||
Float(text.toFloat())
|
||||
}
|
||||
|
||||
protected val term: Parser<Term> by (dummy
|
||||
or float
|
||||
or int
|
||||
or variable
|
||||
or compound
|
||||
or atom
|
||||
) map { it }
|
||||
|
||||
override val rootParser: Parser<Any> = term
|
||||
}
|
||||
|
|
@ -1,27 +0,0 @@
|
|||
package better_parser
|
||||
|
||||
import com.github.h0tk3y.betterParse.combinators.times
|
||||
import com.github.h0tk3y.betterParse.combinators.unaryMinus
|
||||
import com.github.h0tk3y.betterParse.combinators.use
|
||||
import com.github.h0tk3y.betterParse.grammar.parseToEnd
|
||||
import com.github.h0tk3y.betterParse.parser.Parser
|
||||
import prolog.ast.logic.LogicOperand
|
||||
import prolog.builtins.Query
|
||||
|
||||
class SimpleReplParser(val debug: Boolean = false) : SimpleSourceParser() {
|
||||
override val rootParser: Parser<Query> by (body * -dot) use { Query(this as LogicOperand) }
|
||||
|
||||
fun parse(input: String): Query {
|
||||
if (debug) {
|
||||
println("Parsing input: $input")
|
||||
}
|
||||
|
||||
val query = parseToEnd(input) as Query
|
||||
|
||||
if (debug) {
|
||||
println("Parsed query: $query")
|
||||
}
|
||||
|
||||
return query
|
||||
}
|
||||
}
|
||||
|
|
@ -1,48 +0,0 @@
|
|||
package better_parser
|
||||
|
||||
import com.github.h0tk3y.betterParse.combinators.*
|
||||
import com.github.h0tk3y.betterParse.grammar.parser
|
||||
import com.github.h0tk3y.betterParse.lexer.literalToken
|
||||
import com.github.h0tk3y.betterParse.parser.Parser
|
||||
import prolog.ast.arithmetic.ArithmeticOperator
|
||||
import prolog.ast.logic.*
|
||||
import prolog.ast.terms.*
|
||||
import prolog.builtins.Conjunction
|
||||
import prolog.builtins.Disjunction
|
||||
|
||||
open class SimpleSourceParser : SimplePrologParser() {
|
||||
protected val simpleLogicOperand: Parser<LogicOperand> by (dummy
|
||||
or compound
|
||||
or atom
|
||||
)
|
||||
protected val logicOperand: Parser<LogicOperand> by (dummy
|
||||
or parser(::operator)
|
||||
or simpleLogicOperand
|
||||
)
|
||||
|
||||
protected val arithmeticOperator: Parser<ArithmeticOperator> by dummy
|
||||
protected val logicOperator: Parser<LogicOperator> by (simpleLogicOperand * comma * logicOperand) use {
|
||||
Conjunction(t1, t3)
|
||||
}
|
||||
|
||||
protected val operator: Parser<Operator> by (arithmeticOperator or logicOperator) use { this as Operator }
|
||||
|
||||
protected val head: Parser<Head> by (dummy
|
||||
or compound
|
||||
or atom
|
||||
)
|
||||
protected val body: Parser<Body> by (dummy
|
||||
or operator
|
||||
or head
|
||||
) use { this as Body }
|
||||
|
||||
// ----
|
||||
|
||||
private val rule: Parser<Rule> by (head * -neck * body) use { Rule(t1, t2) }
|
||||
private val fact: Parser<Fact> by head use { Fact(this) }
|
||||
|
||||
private val clause: Parser<Clause> by ((rule or fact) * -dot)
|
||||
private val clauses: Parser<List<Clause>> by zeroOrMore(clause)
|
||||
|
||||
override val rootParser: Parser<Any> by clauses
|
||||
}
|
||||
11
src/parser/Parser.kt
Normal file
11
src/parser/Parser.kt
Normal file
|
|
@ -0,0 +1,11 @@
|
|||
package parser
|
||||
|
||||
interface Parser {
|
||||
/**
|
||||
* Parses the input string and returns the parsed result.
|
||||
*
|
||||
* @param input The input string to parse.
|
||||
* @return The parsed result, which is the AST of the input.
|
||||
*/
|
||||
fun parse(input: String): Any
|
||||
}
|
||||
9
src/parser/ReplParser.kt
Normal file
9
src/parser/ReplParser.kt
Normal file
|
|
@ -0,0 +1,9 @@
|
|||
package parser
|
||||
|
||||
import prolog.builtins.Query
|
||||
|
||||
class ReplParser: Parser {
|
||||
override fun parse(input: String): Query {
|
||||
TODO("Not yet implemented")
|
||||
}
|
||||
}
|
||||
12
src/parser/ScriptParser.kt
Normal file
12
src/parser/ScriptParser.kt
Normal file
|
|
@ -0,0 +1,12 @@
|
|||
package parser
|
||||
|
||||
import com.github.h0tk3y.betterParse.grammar.Grammar
|
||||
import com.github.h0tk3y.betterParse.grammar.parseToEnd
|
||||
import parser.grammars.LogicGrammar
|
||||
import prolog.ast.logic.Clause
|
||||
|
||||
class ScriptParser: Parser {
|
||||
private val grammar: Grammar<List<Clause>> = LogicGrammar() as Grammar<List<Clause>>
|
||||
|
||||
override fun parse(input: String): List<Clause> = grammar.parseToEnd(input)
|
||||
}
|
||||
22
src/parser/grammars/LogicGrammar.kt
Normal file
22
src/parser/grammars/LogicGrammar.kt
Normal file
|
|
@ -0,0 +1,22 @@
|
|||
package parser.grammars
|
||||
|
||||
import com.github.h0tk3y.betterParse.combinators.oneOrMore
|
||||
import com.github.h0tk3y.betterParse.combinators.or
|
||||
import com.github.h0tk3y.betterParse.combinators.separated
|
||||
import com.github.h0tk3y.betterParse.combinators.times
|
||||
import com.github.h0tk3y.betterParse.combinators.unaryMinus
|
||||
import com.github.h0tk3y.betterParse.combinators.use
|
||||
import com.github.h0tk3y.betterParse.parser.Parser
|
||||
import prolog.ast.logic.Clause
|
||||
import prolog.ast.logic.Fact
|
||||
import prolog.ast.logic.Rule
|
||||
|
||||
class LogicGrammar : TermsGrammar() {
|
||||
protected val rule: Parser<Rule> by (head * -neck * body) use { Rule(t1, t2) }
|
||||
protected val fact: Parser<Fact> by head use { Fact(this) }
|
||||
|
||||
protected val clause: Parser<Clause> by ((rule or fact) * -dot)
|
||||
protected val clauses: Parser<List<Clause>> by oneOrMore(clause)
|
||||
|
||||
override val rootParser: Parser<Any> by clauses
|
||||
}
|
||||
74
src/parser/grammars/TermsGrammar.kt
Normal file
74
src/parser/grammars/TermsGrammar.kt
Normal file
|
|
@ -0,0 +1,74 @@
|
|||
package parser.grammars
|
||||
|
||||
import com.github.h0tk3y.betterParse.combinators.or
|
||||
import com.github.h0tk3y.betterParse.combinators.separated
|
||||
import com.github.h0tk3y.betterParse.combinators.times
|
||||
import com.github.h0tk3y.betterParse.combinators.unaryMinus
|
||||
import com.github.h0tk3y.betterParse.combinators.use
|
||||
import com.github.h0tk3y.betterParse.grammar.parser
|
||||
import com.github.h0tk3y.betterParse.parser.Parser
|
||||
import prolog.ast.arithmetic.Float
|
||||
import prolog.ast.arithmetic.Integer
|
||||
import prolog.ast.logic.LogicOperand
|
||||
import prolog.ast.logic.LogicOperator
|
||||
import prolog.ast.terms.Atom
|
||||
import prolog.ast.terms.Body
|
||||
import prolog.ast.terms.Head
|
||||
import prolog.ast.terms.Operator
|
||||
import prolog.ast.terms.Structure
|
||||
import prolog.ast.terms.Term
|
||||
import prolog.ast.terms.Variable
|
||||
import prolog.builtins.Conjunction
|
||||
|
||||
open class TermsGrammar : Tokens() {
|
||||
// Basic named terms
|
||||
protected val variable: Parser<Variable> by variableToken use { Variable(text) }
|
||||
protected val atom: Parser<Atom> by nameToken use { Atom(text) }
|
||||
protected val compound: Parser<Structure> by (atom * -leftParenthesis * separated(
|
||||
parser(::term),
|
||||
comma,
|
||||
acceptZero = true
|
||||
) * -rightParenthesis) use {
|
||||
Structure(t1, t2.terms)
|
||||
}
|
||||
|
||||
// Basic arithmetic
|
||||
protected val int: Parser<Integer> by integerToken use { Integer(text.toInt()) }
|
||||
protected val float: Parser<Float> by floatToken use { Float(text.toFloat()) }
|
||||
|
||||
// Operators
|
||||
protected val simpleLogicOperand: Parser<LogicOperand> by (dummy
|
||||
or compound
|
||||
or atom
|
||||
)
|
||||
protected val logicOperand: Parser<LogicOperand> by (dummy
|
||||
or parser(::operator)
|
||||
or simpleLogicOperand
|
||||
)
|
||||
protected val logicOperator: Parser<LogicOperator> by (simpleLogicOperand * -comma * logicOperand) use {
|
||||
Conjunction(t1, t2)
|
||||
}
|
||||
protected val operator: Parser<Operator> by (dummy
|
||||
or logicOperator
|
||||
)
|
||||
|
||||
// Parts
|
||||
protected val head: Parser<Head> by (dummy
|
||||
or compound
|
||||
or atom
|
||||
)
|
||||
protected val body: Parser<Body> by (dummy
|
||||
or operator
|
||||
or head
|
||||
) use { this as Body }
|
||||
|
||||
protected val term: Parser<Term> by (dummy
|
||||
or float
|
||||
or int
|
||||
or variable
|
||||
or compound
|
||||
or atom
|
||||
)
|
||||
|
||||
override val rootParser: Parser<Any> by term
|
||||
}
|
||||
32
src/parser/grammars/Tokens.kt
Normal file
32
src/parser/grammars/Tokens.kt
Normal file
|
|
@ -0,0 +1,32 @@
|
|||
package parser.grammars
|
||||
|
||||
import com.github.h0tk3y.betterParse.combinators.use
|
||||
import com.github.h0tk3y.betterParse.grammar.Grammar
|
||||
import com.github.h0tk3y.betterParse.lexer.Token
|
||||
import com.github.h0tk3y.betterParse.lexer.literalToken
|
||||
import com.github.h0tk3y.betterParse.lexer.regexToken
|
||||
import com.github.h0tk3y.betterParse.lexer.token
|
||||
|
||||
abstract class Tokens : Grammar<Any>() {
|
||||
// Prolog tokens
|
||||
protected val nameToken: Token by regexToken("[a-z][a-zA-Z0-9_]*")
|
||||
protected val variableToken: Token by regexToken("[A-Z][a-zA-Z0-9_]*")
|
||||
|
||||
// Arithmetic tokens
|
||||
protected val floatToken: Token by regexToken("-?[1-9][0-9]*\\.[0-9]+")
|
||||
protected val integerToken: Token by regexToken("-?([1-9][0-9]*|0)")
|
||||
|
||||
// Special tokens
|
||||
protected val neck by literalToken(":-")
|
||||
protected val comma: Token by literalToken(",")
|
||||
protected val leftParenthesis: Token by literalToken("(")
|
||||
protected val rightParenthesis: Token by literalToken(")")
|
||||
protected val dot by literalToken(".")
|
||||
|
||||
// Ignored tokens
|
||||
protected val whitespace: Token by regexToken("\\s+", ignore = true)
|
||||
protected val singleLineComment: Token by regexToken("%[^\\n]*", ignore = true)
|
||||
protected val multiLineComment: Token by regexToken("/\\*.*?\\*/", ignore = true)
|
||||
|
||||
protected val dummy by token { _, _ -> -1 } use { throw IllegalStateException("This parser should not be used") }
|
||||
}
|
||||
Reference in a new issue