Implemented unary operator parsing

This commit is contained in:
Eduard Urbach 2024-07-27 12:49:39 +02:00
parent d001e4e55f
commit 944bacf4e1
Signed by: akyoto
GPG Key ID: C874F672B1AF20C0
8 changed files with 142 additions and 16 deletions

View File

@ -157,6 +157,7 @@ This is what generates expressions from tokens.
- [x] `<<=`, `>>=`
- [x] `==`, `!=`, `<`, `<=`, `>`, `>=`
- [x] `&&`, `||`
- [ ] `!`, `-`
### Architecture

View File

@ -17,17 +17,22 @@ func TestParse(t *testing.T) {
}{
{"Identity", "1", "1"},
{"Basic calculation", "1+2", "(+ 1 2)"},
{"Same operator", "1+2+3", "(+ (+ 1 2) 3)"},
{"Same operator 2", "1+2+3+4", "(+ (+ (+ 1 2) 3) 4)"},
{"Different operator", "1+2-3", "(- (+ 1 2) 3)"},
{"Different operator 2", "1+2-3+4", "(+ (- (+ 1 2) 3) 4)"},
{"Different operator 3", "1+2-3+4-5", "(- (+ (- (+ 1 2) 3) 4) 5)"},
{"Grouped identity", "(1)", "1"},
{"Grouped identity 2", "((1))", "1"},
{"Grouped identity 3", "(((1)))", "1"},
{"Adding identity", "(1)+(2)", "(+ 1 2)"},
{"Adding identity 2", "(1)+(2)+(3)", "(+ (+ 1 2) 3)"},
{"Adding identity 3", "(1)+(2)+(3)+(4)", "(+ (+ (+ 1 2) 3) 4)"},
{"Grouping", "(1+2)", "(+ 1 2)"},
{"Grouping 2", "(1+2+3)", "(+ (+ 1 2) 3)"},
{"Grouping 3", "((1)+(2)+(3))", "(+ (+ 1 2) 3)"},
@ -35,9 +40,11 @@ func TestParse(t *testing.T) {
{"Grouping right", "1*(2+3)", "(* 1 (+ 2 3))"},
{"Grouping same operator", "1+(2+3)", "(+ 1 (+ 2 3))"},
{"Grouping same operator 2", "1+(2+3)+(4+5)", "(+ (+ 1 (+ 2 3)) (+ 4 5))"},
{"Two groups", "(1+2)*(3+4)", "(* (+ 1 2) (+ 3 4))"},
{"Two groups 2", "(1+2-3)*(3+4-5)", "(* (- (+ 1 2) 3) (- (+ 3 4) 5))"},
{"Two groups 3", "(1+2)*(3+4-5)", "(* (+ 1 2) (- (+ 3 4) 5))"},
{"Operator priority", "1+2*3", "(+ 1 (* 2 3))"},
{"Operator priority 2", "1*2+3", "(+ (* 1 2) 3)"},
{"Operator priority 3", "1+2*3+4", "(+ (+ 1 (* 2 3)) 4)"},
@ -46,10 +53,25 @@ func TestParse(t *testing.T) {
{"Operator priority 6", "1+2*3+4*5", "(+ (+ 1 (* 2 3)) (* 4 5))"},
{"Operator priority 7", "1+2*3*4*5*6", "(+ 1 (* (* (* (* 2 3) 4) 5) 6))"},
{"Operator priority 8", "1*2*3+4*5*6", "(+ (* (* 1 2) 3) (* (* 4 5) 6))"},
{"Complex", "(1+2-3*4)*(5+6-7*8)", "(* (- (+ 1 2) (* 3 4)) (- (+ 5 6) (* 7 8)))"},
{"Complex 2", "(1+2*3-4)*(5+6*7-8)", "(* (- (+ 1 (* 2 3)) 4) (- (+ 5 (* 6 7)) 8))"},
{"Complex 3", "(1+2*3-4)*(5+6*7-8)+9-10*11", "(- (+ (* (- (+ 1 (* 2 3)) 4) (- (+ 5 (* 6 7)) 8)) 9) (* 10 11))"},
{"Unary", "!", "!"},
{"Unary not", "!", "!"},
{"Unary not 2", "!a", "(! a)"},
{"Unary not 3", "!(!a)", "(! (! a))"},
{"Unary not 4", "!(a||b)", "(! (|| a b))"},
{"Unary not 5", "a || !b", "(|| a (! b))"},
{"Unary minus", "-", "-"},
{"Unary minus 2", "-a", "(- a)"},
{"Unary minus 3", "-(-a)", "(- (- a))"},
{"Unary minus 4", "-a+b", "(+ (- a) b)"},
{"Unary minus 5", "-(a+b)", "(- (+ a b))"},
{"Unary minus 6", "a + -b", "(+ a (- b))"},
{"Unary minus 7", "-a + -b", "(+ (- a) (- b))"},
{"Function calls", "a()", "(λ a)"},
{"Function calls 2", "a(1)", "(λ a 1)"},
{"Function calls 3", "a(1)+1", "(+ (λ a 1) 1)"},
@ -77,8 +99,10 @@ func TestParse(t *testing.T) {
{"Function calls 25", "a(1-2*3)", "(λ a (- 1 (* 2 3)))"},
{"Function calls 26", "1+2*a()+4", "(+ (+ 1 (* 2 (λ a))) 4)"},
{"Function calls 27", "sum(a,b)*2+15*4", "(+ (* (λ sum a b) 2) (* 15 4))"},
{"Package function calls", "math.sum(a,b)", "(λ (. math sum) a b)"},
{"Package function calls 2", "generic.math.sum(a,b)", "(λ (. (. generic math) sum) a b)"},
{"Array access", "a[0]", "(@ a 0)"},
{"Array access 2", "a[b+c]", "(@ a (+ b c))"},
{"Array access 3", "a.b[c]", "(@ (. a b) c)"},

View File

@ -10,7 +10,7 @@ import (
type Operator struct {
Symbol string
Precedence int8
Operands int
Operands int8
}
// Operators defines the operators used in the language.
@ -19,6 +19,7 @@ var Operators = map[token.Kind]*Operator{
token.Period: {".", 13, 2},
token.Call: {"λ", 12, 1},
token.Array: {"@", 12, 2},
token.Negate: {"-", 11, 1},
token.Not: {"!", 11, 1},
token.Mul: {"*", 10, 2},
token.Div: {"/", 10, 2},
@ -73,7 +74,7 @@ func numOperands(symbol token.Kind) int {
return -1
}
return operator.Operands
return int(operator.Operands)
}
func precedence(symbol token.Kind) int8 {

View File

@ -117,7 +117,11 @@ func Parse(tokens []token.Token) *Expression {
newPrecedence := node.Precedence
if newPrecedence > oldPrecedence {
cursor.LastChild().Replace(node)
if len(cursor.Children) == numOperands(cursor.Token.Kind) {
cursor.LastChild().Replace(node)
} else {
cursor.AddChild(node)
}
} else {
start := cursor

View File

@ -38,10 +38,11 @@ const (
Shr // >>
LogicalAnd // &&
LogicalOr // ||
Not // ! (unary)
Negate // - (unary)
Equal // ==
Less // <
Greater // >
Not // !
NotEqual // !=
LessEqual // <=
GreaterEqual // >=

View File

@ -28,6 +28,11 @@ func (t Token) IsAssignment() bool {
return t.Kind > _assignments && t.Kind < _assignmentsEnd
}
// IsExpressionStart returns true if the token starts an expression.
func (t Token) IsExpressionStart() bool {
return t.Kind == GroupStart || t.Kind == ArrayStart || t.Kind == BlockStart
}
// IsKeyword returns true if the token is a keyword.
func (t Token) IsKeyword() bool {
return t.Kind > _keywords && t.Kind < _keywordsEnd

View File

@ -26,8 +26,28 @@ func Tokenize(buffer []byte) List {
tokens = append(tokens, Token{Kind: ArrayEnd, Position: i, Length: 1})
case '\n':
tokens = append(tokens, Token{Kind: NewLine, Position: i, Length: 1})
case '-':
if len(tokens) == 0 || tokens[len(tokens)-1].IsOperator() || tokens[len(tokens)-1].IsExpressionStart() {
tokens = append(tokens, Token{Kind: Negate, Position: i, Length: 1})
} else {
if i+1 < Position(len(buffer)) && buffer[i+1] == '=' {
tokens = append(tokens, Token{Kind: SubAssign, Position: i, Length: 2})
i++
} else {
tokens = append(tokens, Token{Kind: Sub, Position: i, Length: 1})
}
}
case '/':
if i+1 >= Position(len(buffer)) || buffer[i+1] != '/' {
if i+1 < Position(len(buffer)) && buffer[i+1] == '/' {
position := i
for i < Position(len(buffer)) && buffer[i] != '\n' {
i++
}
tokens = append(tokens, Token{Kind: Comment, Position: position, Length: Length(i - position)})
} else {
position := i
i++
@ -45,14 +65,6 @@ func Tokenize(buffer []byte) List {
}
tokens = append(tokens, Token{Kind: kind, Position: position, Length: Length(i - position)})
} else {
position := i
for i < Position(len(buffer)) && buffer[i] != '\n' {
i++
}
tokens = append(tokens, Token{Kind: Comment, Position: position, Length: Length(i - position)})
}
continue

View File

@ -83,13 +83,81 @@ func TestNumber(t *testing.T) {
}
func TestOperator(t *testing.T) {
tokens := token.Tokenize([]byte(`+ - * /`))
tokens := token.Tokenize([]byte(`a + b - c * d / e`))
expected := []token.Kind{
token.Identifier,
token.Add,
token.Identifier,
token.Sub,
token.Identifier,
token.Mul,
token.Identifier,
token.Div,
token.Identifier,
token.EOF,
}
for i, kind := range expected {
assert.Equal(t, tokens[i].Kind, kind)
}
}
func TestNegateFirstToken(t *testing.T) {
tokens := token.Tokenize([]byte(`-a`))
expected := []token.Kind{
token.Negate,
token.Identifier,
token.EOF,
}
for i, kind := range expected {
assert.Equal(t, tokens[i].Kind, kind)
}
}
func TestNegateAfterGroupStart(t *testing.T) {
tokens := token.Tokenize([]byte(`(-a)`))
expected := []token.Kind{
token.GroupStart,
token.Negate,
token.Identifier,
token.GroupEnd,
token.EOF,
}
for i, kind := range expected {
assert.Equal(t, tokens[i].Kind, kind)
}
}
func TestNegateSub(t *testing.T) {
tokens := token.Tokenize([]byte(`-a-b`))
expected := []token.Kind{
token.Negate,
token.Identifier,
token.Sub,
token.Identifier,
token.EOF,
}
for i, kind := range expected {
assert.Equal(t, tokens[i].Kind, kind)
}
}
func TestNegateAfterOperator(t *testing.T) {
tokens := token.Tokenize([]byte(`-a + -b`))
expected := []token.Kind{
token.Negate,
token.Identifier,
token.Add,
token.Negate,
token.Identifier,
token.EOF,
}
@ -99,18 +167,28 @@ func TestOperator(t *testing.T) {
}
func TestOperatorAssign(t *testing.T) {
tokens := token.Tokenize([]byte(`+= -= *= /= &= |= ^= <<= >>=`))
tokens := token.Tokenize([]byte(`a += b -= c *= d /= e &= f |= g ^= h <<= i >>= j`))
expected := []token.Kind{
token.Identifier,
token.AddAssign,
token.Identifier,
token.SubAssign,
token.Identifier,
token.MulAssign,
token.Identifier,
token.DivAssign,
token.Identifier,
token.AndAssign,
token.Identifier,
token.OrAssign,
token.Identifier,
token.XorAssign,
token.Identifier,
token.ShlAssign,
token.Identifier,
token.ShrAssign,
token.Identifier,
token.EOF,
}