please help
This commit is contained in:
@@ -3,16 +3,17 @@ package main
|
||||
import (
|
||||
"fmt"
|
||||
|
||||
"github.com/wyst-lang/wyst/token"
|
||||
"github.com/wyst-lang/wyst/parser"
|
||||
)
|
||||
|
||||
func main() {
|
||||
tokens, err := Tokenize("test 123", token.RULES_TOP)
|
||||
code := "hello"
|
||||
ast, state, err := parser.Parse(code, parser.TOP)
|
||||
if err != nil {
|
||||
fmt.Printf("Syntax error at %s \n", err)
|
||||
fmt.Printf("SyntaxErr at %d:%d: %s\n", state.Line, state.Column, err)
|
||||
return
|
||||
}
|
||||
for i := 0; i < len(tokens); i++ {
|
||||
fmt.Printf("%s\n", tokens[i])
|
||||
for i := 0; i < len(ast); i++ {
|
||||
fmt.Printf("%s\n", ast[i])
|
||||
}
|
||||
}
|
||||
|
||||
@@ -0,0 +1,60 @@
|
||||
package parser
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
|
||||
"github.com/wyst-lang/wyst/tokenizer"
|
||||
)
|
||||
|
||||
type inner interface {
|
||||
isInner()
|
||||
}
|
||||
|
||||
type Rule struct {
|
||||
Name string
|
||||
Inner inner
|
||||
}
|
||||
|
||||
func (t Rule) String() string {
|
||||
return fmt.Sprintf("AstRule(%s)", t.Name)
|
||||
}
|
||||
|
||||
type Rules struct {
|
||||
Value []Rule
|
||||
}
|
||||
|
||||
func (v Rules) isInner() {}
|
||||
|
||||
type Token struct {
|
||||
Value tokenizer.Rule
|
||||
}
|
||||
|
||||
func (v Token) isInner() {}
|
||||
|
||||
type String struct {
|
||||
Value string
|
||||
}
|
||||
|
||||
func (v String) isInner() {}
|
||||
|
||||
type AstNode struct {
|
||||
Rule Rule
|
||||
Value string
|
||||
Inner []AstNode
|
||||
}
|
||||
|
||||
func (t AstNode) String() string {
|
||||
return fmt.Sprintf("AstNode(\n rule=%s,\n value=%s,\n inner=%v\n)", t.Rule, t.Value, t.Inner)
|
||||
}
|
||||
|
||||
// Ported rules from the tokenizer
|
||||
var (
|
||||
IDENTIFIER = Rule{"IDENTIFIER", Token{tokenizer.IDENTIFIER}}
|
||||
)
|
||||
|
||||
var (
|
||||
TOP = Rule{"TOP", Rules{
|
||||
[]Rule{
|
||||
IDENTIFIER,
|
||||
}}}
|
||||
)
|
||||
@@ -0,0 +1,17 @@
|
||||
package parser
|
||||
|
||||
import (
|
||||
"github.com/wyst-lang/wyst/tokenizer"
|
||||
)
|
||||
|
||||
func Parse(code string, rule Rule) ([]AstNode, tokenizer.State, error) {
|
||||
tokens, state, _ := tokenizer.Tokenize(code)
|
||||
var ast = []AstNode{}
|
||||
for i := 0; i < len(tokens); i++ {
|
||||
// token := tokens[i]
|
||||
if rule.Inner == (String{""}) {
|
||||
|
||||
}
|
||||
}
|
||||
return ast, state, nil
|
||||
}
|
||||
@@ -1,26 +0,0 @@
|
||||
package token
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"regexp"
|
||||
)
|
||||
|
||||
type Rule struct {
|
||||
Pattern regexp.Regexp
|
||||
Name string
|
||||
}
|
||||
|
||||
func (t Rule) String() string {
|
||||
return fmt.Sprintf("TokenRule(%s)", t.Name)
|
||||
}
|
||||
|
||||
func NewToken(name string, pattern regexp.Regexp) Rule {
|
||||
return Rule{Name: name, Pattern: pattern}
|
||||
}
|
||||
|
||||
var (
|
||||
NUMBER = NewToken("NUMBER", *regexp.MustCompile("^[0-9]*"))
|
||||
IDENTIFIER = NewToken("IDENTIFIER", *regexp.MustCompile("^[A-Za-z_][A-Za-z_0-9]*"))
|
||||
)
|
||||
|
||||
var RULES_TOP = []Rule{IDENTIFIER, NUMBER}
|
||||
@@ -0,0 +1,26 @@
|
||||
package tokenizer
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"regexp"
|
||||
)
|
||||
|
||||
type Rule struct {
|
||||
Pattern regexp.Regexp
|
||||
Name string
|
||||
}
|
||||
|
||||
func (t Rule) String() string {
|
||||
return fmt.Sprintf("TokenRule(%s)", t.Name)
|
||||
}
|
||||
|
||||
func NewRule(name string, pattern regexp.Regexp) Rule {
|
||||
return Rule{Name: name, Pattern: pattern}
|
||||
}
|
||||
|
||||
var (
|
||||
NUMBER = NewRule("NUMBER", *regexp.MustCompile(`^\d+`))
|
||||
IDENTIFIER = NewRule("IDENTIFIER", *regexp.MustCompile("^[A-Za-z_][A-Za-z_0-9]*"))
|
||||
)
|
||||
|
||||
var RULES = []Rule{IDENTIFIER, NUMBER}
|
||||
@@ -1,10 +1,8 @@
|
||||
package main
|
||||
package tokenizer
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"strings"
|
||||
|
||||
"github.com/wyst-lang/wyst/token"
|
||||
)
|
||||
|
||||
type State struct {
|
||||
@@ -13,7 +11,7 @@ type State struct {
|
||||
}
|
||||
|
||||
type Token struct {
|
||||
Rule token.Rule
|
||||
Rule Rule
|
||||
Value string
|
||||
Position State
|
||||
}
|
||||
@@ -22,17 +20,17 @@ func (t Token) String() string {
|
||||
return fmt.Sprintf("Token(\n rule=%s,\n value=%s\n)", t.Rule, t.Value)
|
||||
}
|
||||
|
||||
func Tokenize(code string, rules []token.Rule) ([]Token, error) {
|
||||
func Tokenize(code string) ([]Token, State, error) {
|
||||
var tokens = []Token{}
|
||||
var state = State{1, 0}
|
||||
for code != "" {
|
||||
matched := false
|
||||
for i := 0; i < len(rules); i++ {
|
||||
if rules[i].Pattern.MatchString(code) {
|
||||
match := rules[i].Pattern.FindString(code)
|
||||
for i := 0; i < len(RULES); i++ {
|
||||
if RULES[i].Pattern.MatchString(code) {
|
||||
match := RULES[i].Pattern.FindString(code)
|
||||
matched = true
|
||||
code = code[len(match):]
|
||||
tokens = append(tokens, Token{Rule: rules[i], Value: match, Position: state})
|
||||
tokens = append(tokens, Token{Rule: RULES[i], Value: match, Position: state})
|
||||
state.Column += len(match)
|
||||
break
|
||||
} else if strings.HasPrefix(code, "\n") {
|
||||
@@ -47,8 +45,8 @@ func Tokenize(code string, rules []token.Rule) ([]Token, error) {
|
||||
}
|
||||
}
|
||||
if !matched {
|
||||
return tokens, fmt.Errorf("%d:%d", state.Line, state.Column)
|
||||
return tokens, state, fmt.Errorf("invalid character or token")
|
||||
}
|
||||
}
|
||||
return tokens, nil
|
||||
return tokens, state, nil
|
||||
}
|
||||
Reference in New Issue
Block a user