maride/php-parser
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[refactoring] remove scanner token

Browse Source
This commit is contained in:
Vadym Slizov 2020-08-17 20:31:04 +03:00
parent 394092269a
commit 97747c5ac0
36 changed files with 6591 additions and 9813 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
internal/php5/parser.go
View File

@ -12,7 +12,7 @@ import (
// Parser structure
type Parser struct {
Lexer *scanner.Lexer
currentToken *scanner.Token
currentToken *token.Token
rootNode ast.Vertex
errHandlerFunc func(*errors.Error)
}
@ -40,8 +40,7 @@ func (p *Parser) Error(msg string) {
return
}
var pos = p.currentToken.Position
p.errHandlerFunc(errors.NewError(msg, &pos))
p.errHandlerFunc(errors.NewError(msg, p.currentToken.Position))
}
// Parse the php7 Parser entrypoint
@ -82,7 +81,7 @@ func (p *Parser) MoveFreeFloating(src ast.Vertex, dst ast.Vertex) {
delete(src.GetNode().Tokens, token.Start)
}
func (p *Parser) setFreeFloating(dst ast.Vertex, pos token.Position, tokens []token.Token) {
func (p *Parser) setFreeFloating(dst ast.Vertex, pos token.Position, tokens []*token.Token) {
if len(tokens) == 0 {
return
}
@ -98,7 +97,7 @@ func (p *Parser) setFreeFloating(dst ast.Vertex, pos token.Position, tokens []to
}
}
func (p *Parser) setFreeFloatingTokens(dst ast.Vertex, pos token.Position, tokens []token.Token) {
func (p *Parser) setFreeFloatingTokens(dst ast.Vertex, pos token.Position, tokens []*token.Token) {
if len(tokens) == 0 {
return
}
@ -108,14 +107,14 @@ func (p *Parser) setFreeFloatingTokens(dst ast.Vertex, pos token.Position, token
*dstCollection = make(token.Collection)
}
(*dstCollection)[pos] = make([]token.Token, 0)
(*dstCollection)[pos] = make([]*token.Token, 0)
for _, v := range tokens {
(*dstCollection)[pos] = append((*dstCollection)[pos], v)
}
}
func (p *Parser) setToken(dst ast.Vertex, pos token.Position, tokens []token.Token) {
func (p *Parser) setToken(dst ast.Vertex, pos token.Position, tokens []*token.Token) {
if len(tokens) == 0 {
return
}
@ -141,7 +140,7 @@ func (p *Parser) splitSemiColonAndPhpCloseTag(htmlNode ast.Vertex, prevNode ast.
}
if semiColon[0].Value[0] == ';' {
p.setFreeFloatingTokens(prevNode, token.SemiColon, []token.Token{
p.setFreeFloatingTokens(prevNode, token.SemiColon, []*token.Token{
{
ID: token.ID(';'),
Value: semiColon[0].Value[0:1],
@ -155,28 +154,18 @@ func (p *Parser) splitSemiColonAndPhpCloseTag(htmlNode ast.Vertex, prevNode ast.
tlen = 3
}
phpCloseTag := []token.Token{}
phpCloseTag := []*token.Token{}
if vlen-tlen > 1 {
phpCloseTag = append(phpCloseTag, token.Token{
phpCloseTag = append(phpCloseTag, &token.Token{
ID: token.T_WHITESPACE,
Value: semiColon[0].Value[1 : vlen-tlen],
})
}
phpCloseTag = append(phpCloseTag, token.Token{
phpCloseTag = append(phpCloseTag, &token.Token{
ID: T_CLOSE_TAG,
Value: semiColon[0].Value[vlen-tlen:],
})
p.setFreeFloatingTokens(htmlNode, token.Start, append(phpCloseTag, htmlNode.GetNode().Tokens[token.Start]...))
}
func (p *Parser) returnTokenToPool(yyDollar []yySymType, yyVAL *yySymType) {
for i := 1; i < len(yyDollar); i++ {
if yyDollar[i].token != nil {
p.Lexer.ReturnTokenToPool(yyDollar[i].token)
}
yyDollar[i].token = nil
}
yyVAL.token = nil
}

774
internal/php5/parser_test.go
View File

File diff suppressed because it is too large Load Diff

3351
internal/php5/php5.go
View File

File diff suppressed because it is too large Load Diff

2314
internal/php5/php5.y
View File

File diff suppressed because it is too large Load Diff

2
internal/php5/php5_bench_test.go
View File

@ -414,7 +414,7 @@ CAD;
`
for n := 0; n < b.N; n++ {
lexer := scanner.NewLexer([]byte(src), "5.6", false, nil)
lexer := scanner.NewLexer([]byte(src), "5.6", nil)
php5parser := php5.NewParser(lexer, nil)
php5parser.Parse()
}

11
internal/php5/php5_test.go
View File

@ -22212,11 +22212,12 @@ func TestPhp5(t *testing.T) {
},
}
lexer := scanner.NewLexer([]byte(src), "5.6", false, nil)
lexer := scanner.NewLexer([]byte(src), "5.6", nil)
php5parser := php5.NewParser(lexer, nil)
php5parser.Parse()
actual := php5parser.GetRootNode()
traverser.NewDFS(new(visitor.FilterParserNodes)).Traverse(actual)
traverser.NewDFS(new(visitor.FilterTokens)).Traverse(actual)
assert.DeepEqual(t, expected, actual)
}
@ -22351,11 +22352,12 @@ func TestPhp5Strings(t *testing.T) {
},
}
lexer := scanner.NewLexer([]byte(src), "5.6", false, nil)
lexer := scanner.NewLexer([]byte(src), "5.6", nil)
php5parser := php5.NewParser(lexer, nil)
php5parser.Parse()
actual := php5parser.GetRootNode()
traverser.NewDFS(new(visitor.FilterParserNodes)).Traverse(actual)
traverser.NewDFS(new(visitor.FilterTokens)).Traverse(actual)
assert.DeepEqual(t, expected, actual)
}
@ -22579,11 +22581,12 @@ CAD;
},
}
lexer := scanner.NewLexer([]byte(src), "5.6", false, nil)
lexer := scanner.NewLexer([]byte(src), "5.6", nil)
php5parser := php5.NewParser(lexer, nil)
php5parser.Parse()
actual := php5parser.GetRootNode()
traverser.NewDFS(new(visitor.FilterParserNodes)).Traverse(actual)
traverser.NewDFS(new(visitor.FilterTokens)).Traverse(actual)
assert.DeepEqual(t, expected, actual)
}
@ -22606,7 +22609,7 @@ func TestPhp5ControlCharsErrors(t *testing.T) {
parserErrors = append(parserErrors, e)
}
lexer := scanner.NewLexer([]byte(src), "5.6", false, errorHandlerFunc)
lexer := scanner.NewLexer([]byte(src), "5.6", errorHandlerFunc)
php5parser := php5.NewParser(lexer, errorHandlerFunc)
php5parser.Parse()
assert.DeepEqual(t, expected, parserErrors)

31
internal/php7/parser.go
View File

@ -12,7 +12,7 @@ import (
// Parser structure
type Parser struct {
Lexer *scanner.Lexer
currentToken *scanner.Token
currentToken *token.Token
rootNode ast.Vertex
errHandlerFunc func(*errors.Error)
}
@ -39,8 +39,7 @@ func (p *Parser) Error(msg string) {
return
}
var pos = p.currentToken.Position
p.errHandlerFunc(errors.NewError(msg, &pos))
p.errHandlerFunc(errors.NewError(msg, p.currentToken.Position))
}
// Parse the php7 Parser entrypoint
@ -82,7 +81,7 @@ func (p *Parser) MoveFreeFloating(src ast.Vertex, dst ast.Vertex) {
delete(src.GetNode().Tokens, token.Start)
}
func (p *Parser) setFreeFloating(dst ast.Vertex, pos token.Position, tokens []token.Token) {
func (p *Parser) setFreeFloating(dst ast.Vertex, pos token.Position, tokens []*token.Token) {
if len(tokens) == 0 {
return
}
@ -98,7 +97,7 @@ func (p *Parser) setFreeFloating(dst ast.Vertex, pos token.Position, tokens []to
}
}
func (p *Parser) setFreeFloatingTokens(dst ast.Vertex, pos token.Position, tokens []token.Token) {
func (p *Parser) setFreeFloatingTokens(dst ast.Vertex, pos token.Position, tokens []*token.Token) {
if len(tokens) == 0 {
return
}
@ -108,14 +107,14 @@ func (p *Parser) setFreeFloatingTokens(dst ast.Vertex, pos token.Position, token
*dstCollection = make(token.Collection)
}
(*dstCollection)[pos] = make([]token.Token, 0)
(*dstCollection)[pos] = make([]*token.Token, 0)
for _, v := range tokens {
(*dstCollection)[pos] = append((*dstCollection)[pos], v)
}
}
func (p *Parser) setToken(dst ast.Vertex, pos token.Position, tokens []token.Token) {
func (p *Parser) setToken(dst ast.Vertex, pos token.Position, tokens []*token.Token) {
if len(tokens) == 0 {
return
}
@ -141,7 +140,7 @@ func (p *Parser) splitSemiColonAndPhpCloseTag(htmlNode ast.Vertex, prevNode ast.
}
if semiColon[0].Value[0] == ';' {
p.setFreeFloatingTokens(prevNode, token.SemiColon, []token.Token{
p.setFreeFloatingTokens(prevNode, token.SemiColon, []*token.Token{
{
ID: token.ID(';'),
Value: semiColon[0].Value[0:1],
@ -155,28 +154,18 @@ func (p *Parser) splitSemiColonAndPhpCloseTag(htmlNode ast.Vertex, prevNode ast.
tlen = 3
}
phpCloseTag := []token.Token{}
phpCloseTag := []*token.Token{}
if vlen-tlen > 1 {
phpCloseTag = append(phpCloseTag, token.Token{
phpCloseTag = append(phpCloseTag, &token.Token{
ID: token.T_WHITESPACE,
Value: semiColon[0].Value[1 : vlen-tlen],
})
}
phpCloseTag = append(phpCloseTag, token.Token{
phpCloseTag = append(phpCloseTag, &token.Token{
ID: T_CLOSE_TAG,
Value: semiColon[0].Value[vlen-tlen:],
})
p.setFreeFloatingTokens(htmlNode, token.Start, append(phpCloseTag, htmlNode.GetNode().Tokens[token.Start]...))
}
func (p *Parser) returnTokenToPool(yyDollar []yySymType, yyVAL *yySymType) {
for i := 1; i < len(yyDollar); i++ {
if yyDollar[i].token != nil {
p.Lexer.ReturnTokenToPool(yyDollar[i].token)
}
yyDollar[i].token = nil
}
yyVAL.token = nil
}

846
internal/php7/parser_test.go
View File

File diff suppressed because it is too large Load Diff

2890
internal/php7/php7.go
View File

File diff suppressed because it is too large Load Diff

1909
internal/php7/php7.y
View File

File diff suppressed because it is too large Load Diff

2
internal/php7/php7_bench_test.go
View File

@ -382,7 +382,7 @@ CAD;
`
for n := 0; n < b.N; n++ {
lexer := scanner.NewLexer([]byte(src), "7.4", false, nil)
lexer := scanner.NewLexer([]byte(src), "7.4", nil)
php7parser := php7.NewParser(lexer, nil)
php7parser.Parse()
}

11
internal/php7/php7_test.go
View File

@ -19595,11 +19595,12 @@ func TestPhp7(t *testing.T) {
},
}
lexer := scanner.NewLexer(src, "7.4", false, nil)
lexer := scanner.NewLexer(src, "7.4", nil)
php7parser := php7.NewParser(lexer, nil)
php7parser.Parse()
actual := php7parser.GetRootNode()
traverser.NewDFS(new(visitor.FilterParserNodes)).Traverse(actual)
traverser.NewDFS(new(visitor.FilterTokens)).Traverse(actual)
assert.DeepEqual(t, expected, actual)
}
@ -19734,11 +19735,12 @@ func TestPhp5Strings(t *testing.T) {
},
}
lexer := scanner.NewLexer([]byte(src), "7.4", false, nil)
lexer := scanner.NewLexer([]byte(src), "7.4", nil)
php7parser := php7.NewParser(lexer, nil)
php7parser.Parse()
actual := php7parser.GetRootNode()
traverser.NewDFS(new(visitor.FilterParserNodes)).Traverse(actual)
traverser.NewDFS(new(visitor.FilterTokens)).Traverse(actual)
assert.DeepEqual(t, expected, actual)
}
@ -19962,11 +19964,12 @@ CAD;
},
}
lexer := scanner.NewLexer([]byte(src), "7.4", false, nil)
lexer := scanner.NewLexer([]byte(src), "7.4", nil)
php7parser := php7.NewParser(lexer, nil)
php7parser.Parse()
actual := php7parser.GetRootNode()
traverser.NewDFS(new(visitor.FilterParserNodes)).Traverse(actual)
traverser.NewDFS(new(visitor.FilterTokens)).Traverse(actual)
assert.DeepEqual(t, expected, actual)
}
@ -19989,7 +19992,7 @@ func TestPhp7ControlCharsErrors(t *testing.T) {
parserErrors = append(parserErrors, e)
}
lexer := scanner.NewLexer([]byte(src), "7.4", false, errorHandlerFunc)
lexer := scanner.NewLexer([]byte(src), "7.4", errorHandlerFunc)
php7parser := php7.NewParser(lexer, errorHandlerFunc)
php7parser.Parse()
assert.DeepEqual(t, expected, parserErrors)

16
internal/position/position.go
View File

@ -1,9 +1,9 @@
package position
import (
"github.com/z7zmey/php-parser/internal/scanner"
"github.com/z7zmey/php-parser/pkg/ast"
"github.com/z7zmey/php-parser/pkg/position"
"github.com/z7zmey/php-parser/pkg/token"
)
type startPos struct {
@ -95,7 +95,7 @@ func NewNodePosition(n ast.Vertex) *position.Position {
}
// NewTokenPosition returns new Position
func NewTokenPosition(t *scanner.Token) *position.Position {
func NewTokenPosition(t *token.Token) *position.Position {
return &position.Position{
StartLine: t.Position.StartLine,
EndLine: t.Position.EndLine,
@ -105,7 +105,7 @@ func NewTokenPosition(t *scanner.Token) *position.Position {
}
// NewTokensPosition returns new Position
func NewTokensPosition(startToken *scanner.Token, endToken *scanner.Token) *position.Position {
func NewTokensPosition(startToken *token.Token, endToken *token.Token) *position.Position {
return &position.Position{
StartLine: startToken.Position.StartLine,
EndLine: endToken.Position.EndLine,
@ -115,7 +115,7 @@ func NewTokensPosition(startToken *scanner.Token, endToken *scanner.Token) *posi
}
// NewTokenNodePosition returns new Position
func NewTokenNodePosition(t *scanner.Token, n ast.Vertex) *position.Position {
func NewTokenNodePosition(t *token.Token, n ast.Vertex) *position.Position {
return &position.Position{
StartLine: t.Position.StartLine,
EndLine: getNodeEndPos(n).endLine,
@ -125,7 +125,7 @@ func NewTokenNodePosition(t *scanner.Token, n ast.Vertex) *position.Position {
}
// NewNodeTokenPosition returns new Position
func NewNodeTokenPosition(n ast.Vertex, t *scanner.Token) *position.Position {
func NewNodeTokenPosition(n ast.Vertex, t *token.Token) *position.Position {
return &position.Position{
StartLine: getNodeStartPos(n).startLine,
EndLine: t.Position.EndLine,
@ -145,7 +145,7 @@ func NewNodesPosition(startNode ast.Vertex, endNode ast.Vertex) *position.Positi
}
// NewNodeListTokenPosition returns new Position
func NewNodeListTokenPosition(list []ast.Vertex, t *scanner.Token) *position.Position {
func NewNodeListTokenPosition(list []ast.Vertex, t *token.Token) *position.Position {
return &position.Position{
StartLine: getListStartPos(list).startLine,
EndLine: t.Position.EndLine,
@ -155,7 +155,7 @@ func NewNodeListTokenPosition(list []ast.Vertex, t *scanner.Token) *position.Pos
}
// NewTokenNodeListPosition returns new Position
func NewTokenNodeListPosition(t *scanner.Token, list []ast.Vertex) *position.Position {
func NewTokenNodeListPosition(t *token.Token, list []ast.Vertex) *position.Position {
return &position.Position{
StartLine: t.Position.StartLine,
EndLine: getListEndPos(list).endLine,
@ -185,7 +185,7 @@ func NewNodeListNodePosition(list []ast.Vertex, n ast.Vertex) *position.Position
}
// NewOptionalListTokensPosition returns new Position
func NewOptionalListTokensPosition(list []ast.Vertex, t *scanner.Token, endToken *scanner.Token) *position.Position {
func NewOptionalListTokensPosition(list []ast.Vertex, t *token.Token, endToken *token.Token) *position.Position {
if list == nil {
return &position.Position{
StartLine: t.Position.StartLine,

54
internal/position/position_test.go
View File

@ -5,15 +5,15 @@ import (
"testing"
builder "github.com/z7zmey/php-parser/internal/position"
"github.com/z7zmey/php-parser/internal/scanner"
"github.com/z7zmey/php-parser/pkg/ast"
"github.com/z7zmey/php-parser/pkg/position"
"github.com/z7zmey/php-parser/pkg/token"
)
func TestNewTokenPosition(t *testing.T) {
tkn := &scanner.Token{
tkn := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 1,
EndLine: 1,
StartPos: 0,
@ -29,18 +29,18 @@ func TestNewTokenPosition(t *testing.T) {
}
func TestNewTokensPosition(t *testing.T) {
token1 := &scanner.Token{
token1 := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 1,
EndLine: 1,
StartPos: 0,
EndPos: 3,
},
}
token2 := &scanner.Token{
token2 := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 2,
EndLine: 2,
StartPos: 4,
@ -71,9 +71,9 @@ func TestNewNodePosition(t *testing.T) {
}
func TestNewTokenNodePosition(t *testing.T) {
tkn := &scanner.Token{
tkn := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 1,
EndLine: 1,
StartPos: 0,
@ -108,9 +108,9 @@ func TestNewNodeTokenPosition(t *testing.T) {
},
}
tkn := &scanner.Token{
tkn := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 2,
EndLine: 2,
StartPos: 10,
@ -202,9 +202,9 @@ func TestNewNodeListTokenPosition(t *testing.T) {
},
}
tkn := &scanner.Token{
tkn := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 3,
EndLine: 3,
StartPos: 20,
@ -218,9 +218,9 @@ func TestNewNodeListTokenPosition(t *testing.T) {
}
func TestNewTokenNodeListPosition(t *testing.T) {
tkn := &scanner.Token{
tkn := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 1,
EndLine: 1,
StartPos: 0,
@ -332,18 +332,18 @@ func TestNewNodeListNodePosition(t *testing.T) {
}
func TestNewOptionalListTokensPosition(t *testing.T) {
token1 := &scanner.Token{
token1 := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 1,
EndLine: 1,
StartPos: 0,
EndPos: 3,
},
}
token2 := &scanner.Token{
token2 := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 2,
EndLine: 2,
StartPos: 4,
@ -378,18 +378,18 @@ func TestNewOptionalListTokensPosition2(t *testing.T) {
},
}
token1 := &scanner.Token{
token1 := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 4,
EndLine: 4,
StartPos: 27,
EndPos: 29,
},
}
token2 := &scanner.Token{
token2 := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 5,
EndLine: 5,
StartPos: 30,
@ -426,9 +426,9 @@ func TestNilNodeListPos(t *testing.T) {
}
func TestNilNodeListTokenPos(t *testing.T) {
token := &scanner.Token{
token := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 1,
EndLine: 1,
StartPos: 0,
@ -459,9 +459,9 @@ func TestEmptyNodeListPos(t *testing.T) {
}
func TestEmptyNodeListTokenPos(t *testing.T) {
token := &scanner.Token{
token := &token.Token{
Value: []byte(`foo`),
Position: position.Position{
Position: &position.Position{
StartLine: 1,
EndLine: 1,
StartPos: 0,

56
internal/scanner/lexer.go
View File

@ -13,31 +13,32 @@ import (
type Lexer struct {
data []byte
phpVersion string
withTokens bool
errHandlerFunc func(*errors.Error)
sts, ste int
p, pe, cs int
ts, te, act int
stack []int
top int
heredocLabel []byte
tokenPool *TokenPool
tokenPool *token.Pool
positionPool *position.Pool
newLines NewLines
}
func NewLexer(data []byte, phpVersion string, withTokens bool, errHandlerFunc func(*errors.Error)) *Lexer {
func NewLexer(data []byte, phpVersion string, errHandlerFunc func(*errors.Error)) *Lexer {
lex := &Lexer{
data: data,
phpVersion: phpVersion,
withTokens: withTokens,
errHandlerFunc: errHandlerFunc,
pe: len(data),
stack: make([]int, 0),
tokenPool: &TokenPool{},
newLines: NewLines{make([]int, 0, 128)},
tokenPool: token.NewPool(position.DefaultBlockSize),
positionPool: position.NewPool(position.DefaultBlockSize),
newLines: NewLines{make([]int, 0, 128)},
}
initLexer(lex)
@ -45,26 +46,37 @@ func NewLexer(data []byte, phpVersion string, withTokens bool, errHandlerFunc fu
return lex
}
func (lex *Lexer) ReturnTokenToPool(t *Token) {
lex.tokenPool.Put(t)
func (lex *Lexer) setTokenPosition(token *token.Token) {
pos := lex.positionPool.Get()
pos.StartLine = lex.newLines.GetLine(lex.ts)
pos.EndLine = lex.newLines.GetLine(lex.te - 1)
pos.StartPos = lex.ts
pos.EndPos = lex.te
token.Position = pos
}
func (lex *Lexer) setTokenPosition(token *Token) {
token.Position.StartLine = lex.newLines.GetLine(lex.ts)
token.Position.EndLine = lex.newLines.GetLine(lex.te - 1)
token.Position.StartPos = lex.ts
token.Position.EndPos = lex.te
}
func (lex *Lexer) addHiddenToken(t *Token, id TokenID, ps, pe int) {
if !lex.withTokens {
return
func (lex *Lexer) addSkippedToken(t *token.Token, id token.ID, ps, pe int) {
if lex.sts == 0 {
lex.sts = lex.ts
}
t.Tokens = append(t.Tokens, token.Token{
ID: token.ID(id),
Value: lex.data[ps:pe],
})
lex.ste = lex.te
// TODO remove after parser refactoring
skippedTkn := lex.tokenPool.Get()
skippedTkn.ID = id
skippedTkn.Value = lex.data[ps:pe]
lex.setTokenPosition(skippedTkn)
if t.SkippedTokens == nil {
t.SkippedTokens = make([]*token.Token, 0, 2)
}
t.SkippedTokens = append(t.SkippedTokens, skippedTkn)
}
func (lex *Lexer) isNotStringVar() bool {

145
internal/scanner/lexer_tokens.go
View File

@ -1,145 +0,0 @@
package scanner
type TokenID int
//go:generate stringer -type=TokenID -output ./tokenid_string.go
const (
T_INCLUDE TokenID = iota + 57346
T_INCLUDE_ONCE
T_EXIT
T_IF
T_LNUMBER
T_DNUMBER
T_STRING
T_STRING_VARNAME
T_VARIABLE
T_NUM_STRING
T_INLINE_HTML
T_CHARACTER
T_BAD_CHARACTER
T_ENCAPSED_AND_WHITESPACE
T_CONSTANT_ENCAPSED_STRING
T_ECHO
T_DO
T_WHILE
T_ENDWHILE
T_FOR
T_ENDFOR
T_FOREACH
T_ENDFOREACH
T_DECLARE
T_ENDDECLARE
T_AS
T_SWITCH
T_ENDSWITCH
T_CASE
T_DEFAULT
T_BREAK
T_CONTINUE
T_GOTO
T_FUNCTION
T_FN
T_CONST
T_RETURN
T_TRY
T_CATCH
T_FINALLY
T_THROW
T_USE
T_INSTEADOF
T_GLOBAL
T_VAR
T_UNSET
T_ISSET
T_EMPTY
T_HALT_COMPILER
T_CLASS
T_TRAIT
T_INTERFACE
T_EXTENDS
T_IMPLEMENTS
T_OBJECT_OPERATOR
T_DOUBLE_ARROW
T_LIST
T_ARRAY
T_CALLABLE
T_CLASS_C
T_TRAIT_C
T_METHOD_C
T_FUNC_C
T_LINE
T_FILE
T_COMMENT
T_DOC_COMMENT
T_OPEN_TAG
T_OPEN_TAG_WITH_ECHO
T_CLOSE_TAG
T_WHITESPACE
T_START_HEREDOC
T_END_HEREDOC
T_DOLLAR_OPEN_CURLY_BRACES
T_CURLY_OPEN
T_PAAMAYIM_NEKUDOTAYIM
T_NAMESPACE
T_NS_C
T_DIR
T_NS_SEPARATOR
T_ELLIPSIS
T_EVAL
T_REQUIRE
T_REQUIRE_ONCE
T_LOGICAL_OR
T_LOGICAL_XOR
T_LOGICAL_AND
T_INSTANCEOF
T_NEW
T_CLONE
T_ELSEIF
T_ELSE
T_ENDIF
T_PRINT
T_YIELD
T_STATIC
T_ABSTRACT
T_FINAL
T_PRIVATE
T_PROTECTED
T_PUBLIC
T_INC
T_DEC
T_YIELD_FROM
T_INT_CAST
T_DOUBLE_CAST
T_STRING_CAST
T_ARRAY_CAST
T_OBJECT_CAST
T_BOOL_CAST
T_UNSET_CAST
T_COALESCE
T_SPACESHIP
T_NOELSE
T_PLUS_EQUAL
T_MINUS_EQUAL
T_MUL_EQUAL
T_POW_EQUAL
T_DIV_EQUAL
T_CONCAT_EQUAL
T_MOD_EQUAL
T_AND_EQUAL
T_OR_EQUAL
T_XOR_EQUAL
T_SL_EQUAL
T_SR_EQUAL
T_COALESCE_EQUAL
T_BOOLEAN_OR
T_BOOLEAN_AND
T_POW
T_SL
T_SR
T_IS_IDENTICAL
T_IS_NOT_IDENTICAL
T_IS_EQUAL
T_IS_NOT_EQUAL
T_IS_SMALLER_OR_EQUAL
T_IS_GREATER_OR_EQUAL
)

1984
internal/scanner/scanner.go
View File

File diff suppressed because it is too large Load Diff

402
internal/scanner/scanner.rl
View File

@ -4,6 +4,8 @@ import (
"fmt"
"strconv"
"strings"
"github.com/z7zmey/php-parser/pkg/token"
)
%%{
@ -18,13 +20,14 @@ func initLexer(lex *Lexer) {
%% write init;
}
func (lex *Lexer) Lex() *Token {
func (lex *Lexer) Lex() *token.Token {
eof := lex.pe
var tok TokenID
var tok token.ID
token := lex.tokenPool.Get()
token.Tokens = token.Tokens[:0]
token.Value = lex.data[0:0]
tkn := lex.tokenPool.Get()
lex.sts = 0
lex.ste = 0
lblStart := 0
lblEnd := 0
@ -124,7 +127,7 @@ func (lex *Lexer) Lex() *Token {
main := |*
"#!" any* :>> newline => {
lex.addHiddenToken(token, T_COMMENT, lex.ts, lex.te)
lex.addSkippedToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
any => {
fnext html;
@ -135,42 +138,42 @@ func (lex *Lexer) Lex() *Token {
html := |*
any_line+ -- '<?' => {
lex.ungetStr("<")
lex.setTokenPosition(token)
tok = T_INLINE_HTML;
lex.setTokenPosition(tkn)
tok = token.T_INLINE_HTML;
fbreak;
};
'<?' => {
lex.addHiddenToken(token, T_OPEN_TAG, lex.ts, lex.te)
lex.addSkippedToken(tkn, token.T_OPEN_TAG, lex.ts, lex.te)
fnext php;
};
'<?php'i ( [ \t] | newline ) => {
lex.ungetCnt(lex.te - lex.ts - 5)
lex.addHiddenToken(token, T_OPEN_TAG, lex.ts, lex.ts+5)
lex.addSkippedToken(tkn, token.T_OPEN_TAG, lex.ts, lex.ts+5)
fnext php;
};
'<?='i => {
lex.setTokenPosition(token);
tok = T_ECHO;
lex.setTokenPosition(tkn);
tok = token.T_ECHO;
fnext php;
fbreak;
};
*|;
php := |*
whitespace_line* => {lex.addHiddenToken(token, T_WHITESPACE, lex.ts, lex.te)};
'?>' newline? => {lex.setTokenPosition(token); tok = TokenID(int(';')); fnext html; fbreak;};
';' whitespace_line* '?>' newline? => {lex.setTokenPosition(token); tok = TokenID(int(';')); fnext html; fbreak;};
whitespace_line* => {lex.addSkippedToken(tkn, token.T_WHITESPACE, lex.ts, lex.te)};
'?>' newline? => {lex.setTokenPosition(tkn); tok = token.ID(int(';')); fnext html; fbreak;};
';' whitespace_line* '?>' newline? => {lex.setTokenPosition(tkn); tok = token.ID(int(';')); fnext html; fbreak;};
(dnum | exponent_dnum) => {lex.setTokenPosition(token); tok = T_DNUMBER; fbreak;};
(dnum | exponent_dnum) => {lex.setTokenPosition(tkn); tok = token.T_DNUMBER; fbreak;};
bnum => {
s := strings.Replace(string(lex.data[lex.ts+2:lex.te]), "_", "", -1)
_, err := strconv.ParseInt(s, 2, 0)
if err == nil {
lex.setTokenPosition(token); tok = T_LNUMBER; fbreak;
lex.setTokenPosition(tkn); tok = token.T_LNUMBER; fbreak;
}
lex.setTokenPosition(token); tok = T_DNUMBER; fbreak;
lex.setTokenPosition(tkn); tok = token.T_DNUMBER; fbreak;
};
lnum => {
base := 10
@ -182,142 +185,142 @@ func (lex *Lexer) Lex() *Token {
_, err := strconv.ParseInt(s, base, 0)
if err == nil {
lex.setTokenPosition(token); tok = T_LNUMBER; fbreak;
lex.setTokenPosition(tkn); tok = token.T_LNUMBER; fbreak;
}
lex.setTokenPosition(token); tok = T_DNUMBER; fbreak;
lex.setTokenPosition(tkn); tok = token.T_DNUMBER; fbreak;
};
hnum => {
s := strings.Replace(string(lex.data[lex.ts+2:lex.te]), "_", "", -1)
_, err := strconv.ParseInt(s, 16, 0)
if err == nil {
lex.setTokenPosition(token); tok = T_LNUMBER; fbreak;
lex.setTokenPosition(tkn); tok = token.T_LNUMBER; fbreak;
}
lex.setTokenPosition(token); tok = T_DNUMBER; fbreak;
lex.setTokenPosition(tkn); tok = token.T_DNUMBER; fbreak;
};
'abstract'i => {lex.setTokenPosition(token); tok = T_ABSTRACT; fbreak;};
'array'i => {lex.setTokenPosition(token); tok = T_ARRAY; fbreak;};
'as'i => {lex.setTokenPosition(token); tok = T_AS; fbreak;};
'break'i => {lex.setTokenPosition(token); tok = T_BREAK; fbreak;};
'callable'i => {lex.setTokenPosition(token); tok = T_CALLABLE; fbreak;};
'case'i => {lex.setTokenPosition(token); tok = T_CASE; fbreak;};
'catch'i => {lex.setTokenPosition(token); tok = T_CATCH; fbreak;};
'class'i => {lex.setTokenPosition(token); tok = T_CLASS; fbreak;};
'clone'i => {lex.setTokenPosition(token); tok = T_CLONE; fbreak;};
'const'i => {lex.setTokenPosition(token); tok = T_CONST; fbreak;};
'continue'i => {lex.setTokenPosition(token); tok = T_CONTINUE; fbreak;};
'declare'i => {lex.setTokenPosition(token); tok = T_DECLARE; fbreak;};
'default'i => {lex.setTokenPosition(token); tok = T_DEFAULT; fbreak;};
'do'i => {lex.setTokenPosition(token); tok = T_DO; fbreak;};
'echo'i => {lex.setTokenPosition(token); tok = T_ECHO; fbreak;};
'else'i => {lex.setTokenPosition(token); tok = T_ELSE; fbreak;};
'elseif'i => {lex.setTokenPosition(token); tok = T_ELSEIF; fbreak;};
'empty'i => {lex.setTokenPosition(token); tok = T_EMPTY; fbreak;};
'enddeclare'i => {lex.setTokenPosition(token); tok = T_ENDDECLARE; fbreak;};
'endfor'i => {lex.setTokenPosition(token); tok = T_ENDFOR; fbreak;};
'endforeach'i => {lex.setTokenPosition(token); tok = T_ENDFOREACH; fbreak;};
'endif'i => {lex.setTokenPosition(token); tok = T_ENDIF; fbreak;};
'endswitch'i => {lex.setTokenPosition(token); tok = T_ENDSWITCH; fbreak;};
'endwhile'i => {lex.setTokenPosition(token); tok = T_ENDWHILE; fbreak;};
'eval'i => {lex.setTokenPosition(token); tok = T_EVAL; fbreak;};
'exit'i | 'die'i => {lex.setTokenPosition(token); tok = T_EXIT; fbreak;};
'extends'i => {lex.setTokenPosition(token); tok = T_EXTENDS; fbreak;};
'final'i => {lex.setTokenPosition(token); tok = T_FINAL; fbreak;};
'finally'i => {lex.setTokenPosition(token); tok = T_FINALLY; fbreak;};
'for'i => {lex.setTokenPosition(token); tok = T_FOR; fbreak;};
'foreach'i => {lex.setTokenPosition(token); tok = T_FOREACH; fbreak;};
'function'i | 'cfunction'i => {lex.setTokenPosition(token); tok = T_FUNCTION; fbreak;};
'fn'i => {lex.setTokenPosition(token); tok = T_FN; fbreak;};
'global'i => {lex.setTokenPosition(token); tok = T_GLOBAL; fbreak;};
'goto'i => {lex.setTokenPosition(token); tok = T_GOTO; fbreak;};
'if'i => {lex.setTokenPosition(token); tok = T_IF; fbreak;};
'isset'i => {lex.setTokenPosition(token); tok = T_ISSET; fbreak;};
'implements'i => {lex.setTokenPosition(token); tok = T_IMPLEMENTS; fbreak;};
'instanceof'i => {lex.setTokenPosition(token); tok = T_INSTANCEOF; fbreak;};
'insteadof'i => {lex.setTokenPosition(token); tok = T_INSTEADOF; fbreak;};
'interface'i => {lex.setTokenPosition(token); tok = T_INTERFACE; fbreak;};
'list'i => {lex.setTokenPosition(token); tok = T_LIST; fbreak;};
'namespace'i => {lex.setTokenPosition(token); tok = T_NAMESPACE; fbreak;};
'private'i => {lex.setTokenPosition(token); tok = T_PRIVATE; fbreak;};
'public'i => {lex.setTokenPosition(token); tok = T_PUBLIC; fbreak;};
'print'i => {lex.setTokenPosition(token); tok = T_PRINT; fbreak;};
'protected'i => {lex.setTokenPosition(token); tok = T_PROTECTED; fbreak;};
'return'i => {lex.setTokenPosition(token); tok = T_RETURN; fbreak;};
'static'i => {lex.setTokenPosition(token); tok = T_STATIC; fbreak;};
'switch'i => {lex.setTokenPosition(token); tok = T_SWITCH; fbreak;};
'throw'i => {lex.setTokenPosition(token); tok = T_THROW; fbreak;};
'trait'i => {lex.setTokenPosition(token); tok = T_TRAIT; fbreak;};
'try'i => {lex.setTokenPosition(token); tok = T_TRY; fbreak;};
'unset'i => {lex.setTokenPosition(token); tok = T_UNSET; fbreak;};
'use'i => {lex.setTokenPosition(token); tok = T_USE; fbreak;};
'var'i => {lex.setTokenPosition(token); tok = T_VAR; fbreak;};
'while'i => {lex.setTokenPosition(token); tok = T_WHILE; fbreak;};
'yield'i whitespace_line* 'from'i => {lex.setTokenPosition(token); tok = T_YIELD_FROM; fbreak;};
'yield'i => {lex.setTokenPosition(token); tok = T_YIELD; fbreak;};
'include'i => {lex.setTokenPosition(token); tok = T_INCLUDE; fbreak;};
'include_once'i => {lex.setTokenPosition(token); tok = T_INCLUDE_ONCE; fbreak;};
'require'i => {lex.setTokenPosition(token); tok = T_REQUIRE; fbreak;};
'require_once'i => {lex.setTokenPosition(token); tok = T_REQUIRE_ONCE; fbreak;};
'__CLASS__'i => {lex.setTokenPosition(token); tok = T_CLASS_C; fbreak;};
'__DIR__'i => {lex.setTokenPosition(token); tok = T_DIR; fbreak;};
'__FILE__'i => {lex.setTokenPosition(token); tok = T_FILE; fbreak;};
'__FUNCTION__'i => {lex.setTokenPosition(token); tok = T_FUNC_C; fbreak;};
'__LINE__'i => {lex.setTokenPosition(token); tok = T_LINE; fbreak;};
'__NAMESPACE__'i => {lex.setTokenPosition(token); tok = T_NS_C; fbreak;};
'__METHOD__'i => {lex.setTokenPosition(token); tok = T_METHOD_C; fbreak;};
'__TRAIT__'i => {lex.setTokenPosition(token); tok = T_TRAIT_C; fbreak;};
'__halt_compiler'i => {lex.setTokenPosition(token); tok = T_HALT_COMPILER; fnext halt_compiller_open_parenthesis; fbreak;};
'new'i => {lex.setTokenPosition(token); tok = T_NEW; fbreak;};
'and'i => {lex.setTokenPosition(token); tok = T_LOGICAL_AND; fbreak;};
'or'i => {lex.setTokenPosition(token); tok = T_LOGICAL_OR; fbreak;};
'xor'i => {lex.setTokenPosition(token); tok = T_LOGICAL_XOR; fbreak;};
'\\' => {lex.setTokenPosition(token); tok = T_NS_SEPARATOR; fbreak;};
'...' => {lex.setTokenPosition(token); tok = T_ELLIPSIS; fbreak;};
'::' => {lex.setTokenPosition(token); tok = T_PAAMAYIM_NEKUDOTAYIM; fbreak;};
'&&' => {lex.setTokenPosition(token); tok = T_BOOLEAN_AND; fbreak;};
'||' => {lex.setTokenPosition(token); tok = T_BOOLEAN_OR; fbreak;};
'&=' => {lex.setTokenPosition(token); tok = T_AND_EQUAL; fbreak;};
'|=' => {lex.setTokenPosition(token); tok = T_OR_EQUAL; fbreak;};
'.=' => {lex.setTokenPosition(token); tok = T_CONCAT_EQUAL; fbreak;};
'*=' => {lex.setTokenPosition(token); tok = T_MUL_EQUAL; fbreak;};
'**=' => {lex.setTokenPosition(token); tok = T_POW_EQUAL; fbreak;};
'/=' => {lex.setTokenPosition(token); tok = T_DIV_EQUAL; fbreak;};
'+=' => {lex.setTokenPosition(token); tok = T_PLUS_EQUAL; fbreak;};
'-=' => {lex.setTokenPosition(token); tok = T_MINUS_EQUAL; fbreak;};
'^=' => {lex.setTokenPosition(token); tok = T_XOR_EQUAL; fbreak;};
'%=' => {lex.setTokenPosition(token); tok = T_MOD_EQUAL; fbreak;};
'--' => {lex.setTokenPosition(token); tok = T_DEC; fbreak;};
'++' => {lex.setTokenPosition(token); tok = T_INC; fbreak;};
'=>' => {lex.setTokenPosition(token); tok = T_DOUBLE_ARROW; fbreak;};
'<=>' => {lex.setTokenPosition(token); tok = T_SPACESHIP; fbreak;};
'!=' | '<>' => {lex.setTokenPosition(token); tok = T_IS_NOT_EQUAL; fbreak;};
'!==' => {lex.setTokenPosition(token); tok = T_IS_NOT_IDENTICAL; fbreak;};
'==' => {lex.setTokenPosition(token); tok = T_IS_EQUAL; fbreak;};
'===' => {lex.setTokenPosition(token); tok = T_IS_IDENTICAL; fbreak;};
'<<=' => {lex.setTokenPosition(token); tok = T_SL_EQUAL; fbreak;};
'>>=' => {lex.setTokenPosition(token); tok = T_SR_EQUAL; fbreak;};
'>=' => {lex.setTokenPosition(token); tok = T_IS_GREATER_OR_EQUAL; fbreak;};
'<=' => {lex.setTokenPosition(token); tok = T_IS_SMALLER_OR_EQUAL; fbreak;};
'**' => {lex.setTokenPosition(token); tok = T_POW; fbreak;};
'<<' => {lex.setTokenPosition(token); tok = T_SL; fbreak;};
'>>' => {lex.setTokenPosition(token); tok = T_SR; fbreak;};
'??' => {lex.setTokenPosition(token); tok = T_COALESCE; fbreak;};
'??=' => {lex.setTokenPosition(token); tok = T_COALESCE_EQUAL; fbreak;};
'abstract'i => {lex.setTokenPosition(tkn); tok = token.T_ABSTRACT; fbreak;};
'array'i => {lex.setTokenPosition(tkn); tok = token.T_ARRAY; fbreak;};
'as'i => {lex.setTokenPosition(tkn); tok = token.T_AS; fbreak;};
'break'i => {lex.setTokenPosition(tkn); tok = token.T_BREAK; fbreak;};
'callable'i => {lex.setTokenPosition(tkn); tok = token.T_CALLABLE; fbreak;};
'case'i => {lex.setTokenPosition(tkn); tok = token.T_CASE; fbreak;};
'catch'i => {lex.setTokenPosition(tkn); tok = token.T_CATCH; fbreak;};
'class'i => {lex.setTokenPosition(tkn); tok = token.T_CLASS; fbreak;};
'clone'i => {lex.setTokenPosition(tkn); tok = token.T_CLONE; fbreak;};
'const'i => {lex.setTokenPosition(tkn); tok = token.T_CONST; fbreak;};
'continue'i => {lex.setTokenPosition(tkn); tok = token.T_CONTINUE; fbreak;};
'declare'i => {lex.setTokenPosition(tkn); tok = token.T_DECLARE; fbreak;};
'default'i => {lex.setTokenPosition(tkn); tok = token.T_DEFAULT; fbreak;};
'do'i => {lex.setTokenPosition(tkn); tok = token.T_DO; fbreak;};
'echo'i => {lex.setTokenPosition(tkn); tok = token.T_ECHO; fbreak;};
'else'i => {lex.setTokenPosition(tkn); tok = token.T_ELSE; fbreak;};
'elseif'i => {lex.setTokenPosition(tkn); tok = token.T_ELSEIF; fbreak;};
'empty'i => {lex.setTokenPosition(tkn); tok = token.T_EMPTY; fbreak;};
'enddeclare'i => {lex.setTokenPosition(tkn); tok = token.T_ENDDECLARE; fbreak;};
'endfor'i => {lex.setTokenPosition(tkn); tok = token.T_ENDFOR; fbreak;};
'endforeach'i => {lex.setTokenPosition(tkn); tok = token.T_ENDFOREACH; fbreak;};
'endif'i => {lex.setTokenPosition(tkn); tok = token.T_ENDIF; fbreak;};
'endswitch'i => {lex.setTokenPosition(tkn); tok = token.T_ENDSWITCH; fbreak;};
'endwhile'i => {lex.setTokenPosition(tkn); tok = token.T_ENDWHILE; fbreak;};
'eval'i => {lex.setTokenPosition(tkn); tok = token.T_EVAL; fbreak;};
'exit'i | 'die'i => {lex.setTokenPosition(tkn); tok = token.T_EXIT; fbreak;};
'extends'i => {lex.setTokenPosition(tkn); tok = token.T_EXTENDS; fbreak;};
'final'i => {lex.setTokenPosition(tkn); tok = token.T_FINAL; fbreak;};
'finally'i => {lex.setTokenPosition(tkn); tok = token.T_FINALLY; fbreak;};
'for'i => {lex.setTokenPosition(tkn); tok = token.T_FOR; fbreak;};
'foreach'i => {lex.setTokenPosition(tkn); tok = token.T_FOREACH; fbreak;};
'function'i | 'cfunction'i => {lex.setTokenPosition(tkn); tok = token.T_FUNCTION; fbreak;};
'fn'i => {lex.setTokenPosition(tkn); tok = token.T_FN; fbreak;};
'global'i => {lex.setTokenPosition(tkn); tok = token.T_GLOBAL; fbreak;};
'goto'i => {lex.setTokenPosition(tkn); tok = token.T_GOTO; fbreak;};
'if'i => {lex.setTokenPosition(tkn); tok = token.T_IF; fbreak;};
'isset'i => {lex.setTokenPosition(tkn); tok = token.T_ISSET; fbreak;};
'implements'i => {lex.setTokenPosition(tkn); tok = token.T_IMPLEMENTS; fbreak;};
'instanceof'i => {lex.setTokenPosition(tkn); tok = token.T_INSTANCEOF; fbreak;};
'insteadof'i => {lex.setTokenPosition(tkn); tok = token.T_INSTEADOF; fbreak;};
'interface'i => {lex.setTokenPosition(tkn); tok = token.T_INTERFACE; fbreak;};
'list'i => {lex.setTokenPosition(tkn); tok = token.T_LIST; fbreak;};
'namespace'i => {lex.setTokenPosition(tkn); tok = token.T_NAMESPACE; fbreak;};
'private'i => {lex.setTokenPosition(tkn); tok = token.T_PRIVATE; fbreak;};
'public'i => {lex.setTokenPosition(tkn); tok = token.T_PUBLIC; fbreak;};
'print'i => {lex.setTokenPosition(tkn); tok = token.T_PRINT; fbreak;};
'protected'i => {lex.setTokenPosition(tkn); tok = token.T_PROTECTED; fbreak;};
'return'i => {lex.setTokenPosition(tkn); tok = token.T_RETURN; fbreak;};
'static'i => {lex.setTokenPosition(tkn); tok = token.T_STATIC; fbreak;};
'switch'i => {lex.setTokenPosition(tkn); tok = token.T_SWITCH; fbreak;};
'throw'i => {lex.setTokenPosition(tkn); tok = token.T_THROW; fbreak;};
'trait'i => {lex.setTokenPosition(tkn); tok = token.T_TRAIT; fbreak;};
'try'i => {lex.setTokenPosition(tkn); tok = token.T_TRY; fbreak;};
'unset'i => {lex.setTokenPosition(tkn); tok = token.T_UNSET; fbreak;};
'use'i => {lex.setTokenPosition(tkn); tok = token.T_USE; fbreak;};
'var'i => {lex.setTokenPosition(tkn); tok = token.T_VAR; fbreak;};
'while'i => {lex.setTokenPosition(tkn); tok = token.T_WHILE; fbreak;};
'yield'i whitespace_line* 'from'i => {lex.setTokenPosition(tkn); tok = token.T_YIELD_FROM; fbreak;};
'yield'i => {lex.setTokenPosition(tkn); tok = token.T_YIELD; fbreak;};
'include'i => {lex.setTokenPosition(tkn); tok = token.T_INCLUDE; fbreak;};
'include_once'i => {lex.setTokenPosition(tkn); tok = token.T_INCLUDE_ONCE; fbreak;};
'require'i => {lex.setTokenPosition(tkn); tok = token.T_REQUIRE; fbreak;};
'require_once'i => {lex.setTokenPosition(tkn); tok = token.T_REQUIRE_ONCE; fbreak;};
'__CLASS__'i => {lex.setTokenPosition(tkn); tok = token.T_CLASS_C; fbreak;};
'__DIR__'i => {lex.setTokenPosition(tkn); tok = token.T_DIR; fbreak;};
'__FILE__'i => {lex.setTokenPosition(tkn); tok = token.T_FILE; fbreak;};
'__FUNCTION__'i => {lex.setTokenPosition(tkn); tok = token.T_FUNC_C; fbreak;};
'__LINE__'i => {lex.setTokenPosition(tkn); tok = token.T_LINE; fbreak;};
'__NAMESPACE__'i => {lex.setTokenPosition(tkn); tok = token.T_NS_C; fbreak;};
'__METHOD__'i => {lex.setTokenPosition(tkn); tok = token.T_METHOD_C; fbreak;};
'__TRAIT__'i => {lex.setTokenPosition(tkn); tok = token.T_TRAIT_C; fbreak;};
'__halt_compiler'i => {lex.setTokenPosition(tkn); tok = token.T_HALT_COMPILER; fnext halt_compiller_open_parenthesis; fbreak;};
'new'i => {lex.setTokenPosition(tkn); tok = token.T_NEW; fbreak;};
'and'i => {lex.setTokenPosition(tkn); tok = token.T_LOGICAL_AND; fbreak;};
'or'i => {lex.setTokenPosition(tkn); tok = token.T_LOGICAL_OR; fbreak;};
'xor'i => {lex.setTokenPosition(tkn); tok = token.T_LOGICAL_XOR; fbreak;};
'\\' => {lex.setTokenPosition(tkn); tok = token.T_NS_SEPARATOR; fbreak;};
'...' => {lex.setTokenPosition(tkn); tok = token.T_ELLIPSIS; fbreak;};
'::' => {lex.setTokenPosition(tkn); tok = token.T_PAAMAYIM_NEKUDOTAYIM; fbreak;};
'&&' => {lex.setTokenPosition(tkn); tok = token.T_BOOLEAN_AND; fbreak;};
'||' => {lex.setTokenPosition(tkn); tok = token.T_BOOLEAN_OR; fbreak;};
'&=' => {lex.setTokenPosition(tkn); tok = token.T_AND_EQUAL; fbreak;};
'|=' => {lex.setTokenPosition(tkn); tok = token.T_OR_EQUAL; fbreak;};
'.=' => {lex.setTokenPosition(tkn); tok = token.T_CONCAT_EQUAL; fbreak;};
'*=' => {lex.setTokenPosition(tkn); tok = token.T_MUL_EQUAL; fbreak;};
'**=' => {lex.setTokenPosition(tkn); tok = token.T_POW_EQUAL; fbreak;};
'/=' => {lex.setTokenPosition(tkn); tok = token.T_DIV_EQUAL; fbreak;};
'+=' => {lex.setTokenPosition(tkn); tok = token.T_PLUS_EQUAL; fbreak;};
'-=' => {lex.setTokenPosition(tkn); tok = token.T_MINUS_EQUAL; fbreak;};
'^=' => {lex.setTokenPosition(tkn); tok = token.T_XOR_EQUAL; fbreak;};
'%=' => {lex.setTokenPosition(tkn); tok = token.T_MOD_EQUAL; fbreak;};
'--' => {lex.setTokenPosition(tkn); tok = token.T_DEC; fbreak;};
'++' => {lex.setTokenPosition(tkn); tok = token.T_INC; fbreak;};
'=>' => {lex.setTokenPosition(tkn); tok = token.T_DOUBLE_ARROW; fbreak;};
'<=>' => {lex.setTokenPosition(tkn); tok = token.T_SPACESHIP; fbreak;};
'!=' | '<>' => {lex.setTokenPosition(tkn); tok = token.T_IS_NOT_EQUAL; fbreak;};
'!==' => {lex.setTokenPosition(tkn); tok = token.T_IS_NOT_IDENTICAL; fbreak;};
'==' => {lex.setTokenPosition(tkn); tok = token.T_IS_EQUAL; fbreak;};
'===' => {lex.setTokenPosition(tkn); tok = token.T_IS_IDENTICAL; fbreak;};
'<<=' => {lex.setTokenPosition(tkn); tok = token.T_SL_EQUAL; fbreak;};
'>>=' => {lex.setTokenPosition(tkn); tok = token.T_SR_EQUAL; fbreak;};
'>=' => {lex.setTokenPosition(tkn); tok = token.T_IS_GREATER_OR_EQUAL; fbreak;};
'<=' => {lex.setTokenPosition(tkn); tok = token.T_IS_SMALLER_OR_EQUAL; fbreak;};
'**' => {lex.setTokenPosition(tkn); tok = token.T_POW; fbreak;};
'<<' => {lex.setTokenPosition(tkn); tok = token.T_SL; fbreak;};
'>>' => {lex.setTokenPosition(tkn); tok = token.T_SR; fbreak;};
'??' => {lex.setTokenPosition(tkn); tok = token.T_COALESCE; fbreak;};
'??=' => {lex.setTokenPosition(tkn); tok = token.T_COALESCE_EQUAL; fbreak;};
'(' whitespace* 'array'i whitespace* ')' => {lex.setTokenPosition(token); tok = T_ARRAY_CAST; fbreak;};
'(' whitespace* ('bool'i|'boolean'i) whitespace* ')' => {lex.setTokenPosition(token); tok = T_BOOL_CAST; fbreak;};
'(' whitespace* ('real'i|'double'i|'float'i) whitespace* ')' => {lex.setTokenPosition(token); tok = T_DOUBLE_CAST; fbreak;};
'(' whitespace* ('int'i|'integer'i) whitespace* ')' => {lex.setTokenPosition(token); tok = T_INT_CAST; fbreak;};
'(' whitespace* 'object'i whitespace* ')' => {lex.setTokenPosition(token); tok = T_OBJECT_CAST; fbreak;};
'(' whitespace* ('string'i|'binary'i) whitespace* ')' => {lex.setTokenPosition(token); tok = T_STRING_CAST; fbreak;};
'(' whitespace* 'unset'i whitespace* ')' => {lex.setTokenPosition(token); tok = T_UNSET_CAST; fbreak;};
'(' whitespace* 'array'i whitespace* ')' => {lex.setTokenPosition(tkn); tok = token.T_ARRAY_CAST; fbreak;};
'(' whitespace* ('bool'i|'boolean'i) whitespace* ')' => {lex.setTokenPosition(tkn); tok = token.T_BOOL_CAST; fbreak;};
'(' whitespace* ('real'i|'double'i|'float'i) whitespace* ')' => {lex.setTokenPosition(tkn); tok = token.T_DOUBLE_CAST; fbreak;};
'(' whitespace* ('int'i|'integer'i) whitespace* ')' => {lex.setTokenPosition(tkn); tok = token.T_INT_CAST; fbreak;};
'(' whitespace* 'object'i whitespace* ')' => {lex.setTokenPosition(tkn); tok = token.T_OBJECT_CAST; fbreak;};
'(' whitespace* ('string'i|'binary'i) whitespace* ')' => {lex.setTokenPosition(tkn); tok = token.T_STRING_CAST; fbreak;};
'(' whitespace* 'unset'i whitespace* ')' => {lex.setTokenPosition(tkn); tok = token.T_UNSET_CAST; fbreak;};
('#' | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addHiddenToken(token, T_COMMENT, lex.ts, lex.te)
lex.addSkippedToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
'/*' any_line* :>> '*/' {
isDocComment := false;
@ -326,37 +329,35 @@ func (lex *Lexer) Lex() *Token {
}
if isDocComment {
lex.addHiddenToken(token, T_DOC_COMMENT, lex.ts, lex.te)
lex.addSkippedToken(tkn, token.T_DOC_COMMENT, lex.ts, lex.te)
} else {
lex.addHiddenToken(token, T_COMMENT, lex.ts, lex.te)
lex.addSkippedToken(tkn, token.T_COMMENT, lex.ts, lex.te)
}
};
operators => {
// rune, _ := utf8.DecodeRune(lex.data[lex.ts:lex.te]);
// tok = TokenID(Rune2Class(rune));
lex.setTokenPosition(token);
tok = TokenID(int(lex.data[lex.ts]));
lex.setTokenPosition(tkn);
tok = token.ID(int(lex.data[lex.ts]));
fbreak;
};
"{" => { lex.setTokenPosition(token); tok = TokenID(int('{')); lex.call(ftargs, fentry(php)); goto _out; };
"}" => { lex.setTokenPosition(token); tok = TokenID(int('}')); lex.ret(1); goto _out;};
"$" varname => { lex.setTokenPosition(token); tok = T_VARIABLE; fbreak; };
varname => { lex.setTokenPosition(token); tok = T_STRING; fbreak; };
"{" => { lex.setTokenPosition(tkn); tok = token.ID(int('{')); lex.call(ftargs, fentry(php)); goto _out; };
"}" => { lex.setTokenPosition(tkn); tok = token.ID(int('}')); lex.ret(1); goto _out;};
"$" varname => { lex.setTokenPosition(tkn); tok = token.T_VARIABLE; fbreak; };
varname => { lex.setTokenPosition(tkn); tok = token.T_STRING; fbreak; };
"->" => { lex.setTokenPosition(token); tok = T_OBJECT_OPERATOR; fnext property; fbreak; };
"->" => { lex.setTokenPosition(tkn); tok = token.T_OBJECT_OPERATOR; fnext property; fbreak; };
constant_string => {
lex.setTokenPosition(token);
tok = T_CONSTANT_ENCAPSED_STRING;
lex.setTokenPosition(tkn);
tok = token.T_CONSTANT_ENCAPSED_STRING;
fbreak;
};
"b"i? "<<<" [ \t]* ( heredoc_label | ("'" heredoc_label "'") | ('"' heredoc_label '"') ) newline => {
lex.heredocLabel = lex.data[lblStart:lblEnd]
lex.setTokenPosition(token);
tok = T_START_HEREDOC;
lex.setTokenPosition(tkn);
tok = token.T_START_HEREDOC;
if lex.isHeredocEnd(lex.p+1) {
fnext heredoc_end;
@ -367,8 +368,8 @@ func (lex *Lexer) Lex() *Token {
}
fbreak;
};
"`" => {lex.setTokenPosition(token); tok = TokenID(int('`')); fnext backqote; fbreak;};
'"' => {lex.setTokenPosition(token); tok = TokenID(int('"')); fnext template_string; fbreak;};
"`" => {lex.setTokenPosition(tkn); tok = token.ID(int('`')); fnext backqote; fbreak;};
'"' => {lex.setTokenPosition(tkn); tok = token.ID(int('"')); fnext template_string; fbreak;};
any_line => {
c := lex.data[lex.p]
@ -377,28 +378,28 @@ func (lex *Lexer) Lex() *Token {
*|;
property := |*
whitespace_line* => {lex.addHiddenToken(token, T_WHITESPACE, lex.ts, lex.te)};
"->" => {lex.setTokenPosition(token); tok = T_OBJECT_OPERATOR; fbreak;};
varname => {lex.setTokenPosition(token); tok = T_STRING; fnext php; fbreak;};
whitespace_line* => {lex.addSkippedToken(tkn, token.T_WHITESPACE, lex.ts, lex.te)};
"->" => {lex.setTokenPosition(tkn); tok = token.T_OBJECT_OPERATOR; fbreak;};
varname => {lex.setTokenPosition(tkn); tok = token.T_STRING; fnext php; fbreak;};
any => {lex.ungetCnt(1); fgoto php;};
*|;
nowdoc := |*
any_line* when is_not_heredoc_end => {
lex.setTokenPosition(token);
tok = T_ENCAPSED_AND_WHITESPACE;
lex.setTokenPosition(tkn);
tok = token.T_ENCAPSED_AND_WHITESPACE;
fnext heredoc_end;
fbreak;
};
*|;
heredoc := |*
"{$" => {lex.ungetCnt(1); lex.setTokenPosition(token); tok = T_CURLY_OPEN; lex.call(ftargs, fentry(php)); goto _out;};
"${" => {lex.setTokenPosition(token); tok = T_DOLLAR_OPEN_CURLY_BRACES; lex.call(ftargs, fentry(string_var_name)); goto _out;};
"{$" => {lex.ungetCnt(1); lex.setTokenPosition(tkn); tok = token.T_CURLY_OPEN; lex.call(ftargs, fentry(php)); goto _out;};
"${" => {lex.setTokenPosition(tkn); tok = token.T_DOLLAR_OPEN_CURLY_BRACES; lex.call(ftargs, fentry(string_var_name)); goto _out;};
"$" => {lex.ungetCnt(1); fcall string_var;};
any_line* when is_not_heredoc_end_or_var => {
lex.setTokenPosition(token);
tok = T_ENCAPSED_AND_WHITESPACE;
lex.setTokenPosition(tkn);
tok = token.T_ENCAPSED_AND_WHITESPACE;
if len(lex.data) > lex.p+1 && lex.data[lex.p+1] != '$' && lex.data[lex.p+1] != '{' {
fnext heredoc_end;
@ -408,59 +409,59 @@ func (lex *Lexer) Lex() *Token {
*|;
backqote := |*
"{$" => {lex.ungetCnt(1); lex.setTokenPosition(token); tok = T_CURLY_OPEN; lex.call(ftargs, fentry(php)); goto _out;};
"${" => {lex.setTokenPosition(token); tok = T_DOLLAR_OPEN_CURLY_BRACES; lex.call(ftargs, fentry(string_var_name)); goto _out;};
"{$" => {lex.ungetCnt(1); lex.setTokenPosition(tkn); tok = token.T_CURLY_OPEN; lex.call(ftargs, fentry(php)); goto _out;};
"${" => {lex.setTokenPosition(tkn); tok = token.T_DOLLAR_OPEN_CURLY_BRACES; lex.call(ftargs, fentry(string_var_name)); goto _out;};
"$" varname_first => {lex.ungetCnt(2); fcall string_var;};
'`' => {lex.setTokenPosition(token); tok = TokenID(int('`')); fnext php; fbreak;};
'`' => {lex.setTokenPosition(tkn); tok = token.ID(int('`')); fnext php; fbreak;};
any_line* when is_not_backqoute_end_or_var => {
lex.setTokenPosition(token);
tok = T_ENCAPSED_AND_WHITESPACE;
lex.setTokenPosition(tkn);
tok = token.T_ENCAPSED_AND_WHITESPACE;
fbreak;
};
*|;
template_string := |*
"{$" => {lex.ungetCnt(1); lex.setTokenPosition(token); tok = T_CURLY_OPEN; lex.call(ftargs, fentry(php)); goto _out;};
"${" => {lex.setTokenPosition(token); tok = T_DOLLAR_OPEN_CURLY_BRACES; lex.call(ftargs, fentry(string_var_name)); goto _out;};
"{$" => {lex.ungetCnt(1); lex.setTokenPosition(tkn); tok = token.T_CURLY_OPEN; lex.call(ftargs, fentry(php)); goto _out;};
"${" => {lex.setTokenPosition(tkn); tok = token.T_DOLLAR_OPEN_CURLY_BRACES; lex.call(ftargs, fentry(string_var_name)); goto _out;};
"$" varname_first => {lex.ungetCnt(2); fcall string_var;};
'"' => {lex.setTokenPosition(token); tok = TokenID(int('"')); fnext php; fbreak;};
'"' => {lex.setTokenPosition(tkn); tok = token.ID(int('"')); fnext php; fbreak;};
any_line* when is_not_string_end_or_var => {
lex.setTokenPosition(token);
tok = T_ENCAPSED_AND_WHITESPACE;
lex.setTokenPosition(tkn);
tok = token.T_ENCAPSED_AND_WHITESPACE;
fbreak;
};
*|;
heredoc_end := |*
varname -- ";" => {
lex.setTokenPosition(token);
tok = T_END_HEREDOC;
lex.setTokenPosition(tkn);
tok = token.T_END_HEREDOC;
fnext php;
fbreak;
};
varname => {
lex.setTokenPosition(token);
tok = T_END_HEREDOC;
lex.setTokenPosition(tkn);
tok = token.T_END_HEREDOC;
fnext php;
fbreak;
};
*|;
string_var := |*
'$' varname => {lex.setTokenPosition(token); tok = T_VARIABLE; fbreak;};
'->' varname_first => {lex.ungetCnt(1); lex.setTokenPosition(token); tok = T_OBJECT_OPERATOR; fbreak;};
varname => {lex.setTokenPosition(token); tok = T_STRING; fbreak;};
'[' => {lex.setTokenPosition(token); tok = TokenID(int('[')); lex.call(ftargs, fentry(string_var_index)); goto _out;};
'$' varname => {lex.setTokenPosition(tkn); tok = token.T_VARIABLE; fbreak;};
'->' varname_first => {lex.ungetCnt(1); lex.setTokenPosition(tkn); tok = token.T_OBJECT_OPERATOR; fbreak;};
varname => {lex.setTokenPosition(tkn); tok = token.T_STRING; fbreak;};
'[' => {lex.setTokenPosition(tkn); tok = token.ID(int('[')); lex.call(ftargs, fentry(string_var_index)); goto _out;};
any => {lex.ungetCnt(1); fret;};
*|;
string_var_index := |*
lnum | hnum | bnum => {lex.setTokenPosition(token); tok = T_NUM_STRING; fbreak;};
'$' varname => {lex.setTokenPosition(token); tok = T_VARIABLE; fbreak;};
varname => {lex.setTokenPosition(token); tok = T_STRING; fbreak;};
whitespace_line | [\\'#] => {lex.setTokenPosition(token); tok = T_ENCAPSED_AND_WHITESPACE; lex.ret(2); goto _out;};
operators > (svi, 1) => {lex.setTokenPosition(token); tok = TokenID(int(lex.data[lex.ts])); fbreak;};
']' > (svi, 2) => {lex.setTokenPosition(token); tok = TokenID(int(']')); lex.ret(2); goto _out;};
lnum | hnum | bnum => {lex.setTokenPosition(tkn); tok = token.T_NUM_STRING; fbreak;};
'$' varname => {lex.setTokenPosition(tkn); tok = token.T_VARIABLE; fbreak;};
varname => {lex.setTokenPosition(tkn); tok = token.T_STRING; fbreak;};
whitespace_line | [\\'#] => {lex.setTokenPosition(tkn); tok = token.T_ENCAPSED_AND_WHITESPACE; lex.ret(2); goto _out;};
operators > (svi, 1) => {lex.setTokenPosition(tkn); tok = token.ID(int(lex.data[lex.ts])); fbreak;};
']' > (svi, 2) => {lex.setTokenPosition(tkn); tok = token.ID(int(']')); lex.ret(2); goto _out;};
any_line => {
c := lex.data[lex.p]
lex.error(fmt.Sprintf("WARNING: Unexpected character in input: '%c' (ASCII=%d)", c, c));
@ -468,38 +469,39 @@ func (lex *Lexer) Lex() *Token {
*|;
string_var_name := |*
varname ("[" | "}") => {lex.ungetCnt(1); lex.setTokenPosition(token); tok = T_STRING_VARNAME; fnext php; fbreak;};
varname ("[" | "}") => {lex.ungetCnt(1); lex.setTokenPosition(tkn); tok = token.T_STRING_VARNAME; fnext php; fbreak;};
any => {lex.ungetCnt(1); fnext php;};
*|;
halt_compiller_open_parenthesis := |*
whitespace_line* => {lex.addHiddenToken(token, T_WHITESPACE, lex.ts, lex.te)};
"(" => {lex.setTokenPosition(token); tok = TokenID(int('(')); fnext halt_compiller_close_parenthesis; fbreak;};
whitespace_line* => {lex.addSkippedToken(tkn, token.T_WHITESPACE, lex.ts, lex.te)};
"(" => {lex.setTokenPosition(tkn); tok = token.ID(int('(')); fnext halt_compiller_close_parenthesis; fbreak;};
any => {lex.ungetCnt(1); fnext php;};
*|;
halt_compiller_close_parenthesis := |*
whitespace_line* => {lex.addHiddenToken(token, T_WHITESPACE, lex.ts, lex.te)};
")" => {lex.setTokenPosition(token); tok = TokenID(int(')')); fnext halt_compiller_close_semicolon; fbreak;};
whitespace_line* => {lex.addSkippedToken(tkn, token.T_WHITESPACE, lex.ts, lex.te)};
")" => {lex.setTokenPosition(tkn); tok = token.ID(int(')')); fnext halt_compiller_close_semicolon; fbreak;};
any => {lex.ungetCnt(1); fnext php;};
*|;
halt_compiller_close_semicolon := |*
whitespace_line* => {lex.addHiddenToken(token, T_WHITESPACE, lex.ts, lex.te)};
";" => {lex.setTokenPosition(token); tok = TokenID(int(';')); fnext halt_compiller_end; fbreak;};
whitespace_line* => {lex.addSkippedToken(tkn, token.T_WHITESPACE, lex.ts, lex.te)};
";" => {lex.setTokenPosition(tkn); tok = token.ID(int(';')); fnext halt_compiller_end; fbreak;};
any => {lex.ungetCnt(1); fnext php;};
*|;
halt_compiller_end := |*
any_line* => { lex.addHiddenToken(token, T_HALT_COMPILER, lex.ts, lex.te); };
any_line* => { lex.addSkippedToken(tkn, token.T_HALT_COMPILER, lex.ts, lex.te); };
*|;
write exec;
}%%
token.Value = lex.data[lex.ts:lex.te]
token.ID = tok
lex.addHiddenToken(token, tok, lex.ts, lex.te);
tkn.Value = lex.data[lex.ts:lex.te]
tkn.ID = token.ID(tok)
tkn.SkippedString = lex.data[lex.sts:lex.ste]
lex.addSkippedToken(tkn, tok, lex.ts, lex.te);
return token
return tkn
}

1062
internal/scanner/scanner_test.go
View File

File diff suppressed because it is too large Load Diff

14
internal/scanner/token.go
View File

@ -1,14 +0,0 @@
package scanner
import (
"github.com/z7zmey/php-parser/pkg/position"
"github.com/z7zmey/php-parser/pkg/token"
)
// Token value returned by lexer
type Token struct {
ID TokenID
Value []byte
Tokens []token.Token
Position position.Position
}

22
internal/scanner/token_pool.go
View File

@ -1,22 +0,0 @@
package scanner
// TokenPool light version of sync.Pool for Token objects
type TokenPool struct {
pool []*Token
}
// Get returns *Token from pool or creates new object
func (tp *TokenPool) Get() *Token {
if len(tp.pool) < 1 {
return new(Token)
}
t := tp.pool[len(tp.pool)-1]
tp.pool = tp.pool[:len(tp.pool)-1]
return t
}
// Put returns *Token to pool
func (tp *TokenPool) Put(t *Token) {
tp.pool = append(tp.pool, t)
}

34
internal/scanner/token_pool_test.go
View File

@ -1,34 +0,0 @@
package scanner_test
import (
"reflect"
"testing"
"github.com/z7zmey/php-parser/internal/scanner"
)
func TestTokenPoolGetNew(t *testing.T) {
tp := new(scanner.TokenPool)
newToken := tp.Get()
if newToken == nil {
t.Errorf("*TokenPool.Get() must return new *Token object\n")
}
}
func TestTokenPoolGetFromPool(t *testing.T) {
tp := new(scanner.TokenPool)
expectedToken := &scanner.Token{
Value: []byte("test"),
}
tp.Put(expectedToken)
actualToken := tp.Get()
if !reflect.DeepEqual(expectedToken, actualToken) {
t.Errorf("*TokenPool.Put() must return *Token object from pool\n")
}
}

161
internal/scanner/tokenid_string.go
View File

@ -1,161 +0,0 @@
// Code generated by "stringer -type=TokenID -output ./tokenid_string.go"; DO NOT EDIT.
package scanner
import "strconv"
func _() {
// An "invalid array index" compiler error signifies that the constant values have changed.
// Re-run the stringer command to generate them again.
var x [1]struct{}
_ = x[T_INCLUDE-57346]
_ = x[T_INCLUDE_ONCE-57347]
_ = x[T_EXIT-57348]
_ = x[T_IF-57349]
_ = x[T_LNUMBER-57350]
_ = x[T_DNUMBER-57351]
_ = x[T_STRING-57352]
_ = x[T_STRING_VARNAME-57353]
_ = x[T_VARIABLE-57354]
_ = x[T_NUM_STRING-57355]
_ = x[T_INLINE_HTML-57356]
_ = x[T_CHARACTER-57357]
_ = x[T_BAD_CHARACTER-57358]
_ = x[T_ENCAPSED_AND_WHITESPACE-57359]
_ = x[T_CONSTANT_ENCAPSED_STRING-57360]
_ = x[T_ECHO-57361]
_ = x[T_DO-57362]
_ = x[T_WHILE-57363]
_ = x[T_ENDWHILE-57364]
_ = x[T_FOR-57365]
_ = x[T_ENDFOR-57366]
_ = x[T_FOREACH-57367]
_ = x[T_ENDFOREACH-57368]
_ = x[T_DECLARE-57369]
_ = x[T_ENDDECLARE-57370]
_ = x[T_AS-57371]
_ = x[T_SWITCH-57372]
_ = x[T_ENDSWITCH-57373]
_ = x[T_CASE-57374]
_ = x[T_DEFAULT-57375]
_ = x[T_BREAK-57376]
_ = x[T_CONTINUE-57377]
_ = x[T_GOTO-57378]
_ = x[T_FUNCTION-57379]
_ = x[T_FN-57380]
_ = x[T_CONST-57381]
_ = x[T_RETURN-57382]
_ = x[T_TRY-57383]
_ = x[T_CATCH-57384]
_ = x[T_FINALLY-57385]
_ = x[T_THROW-57386]
_ = x[T_USE-57387]
_ = x[T_INSTEADOF-57388]
_ = x[T_GLOBAL-57389]
_ = x[T_VAR-57390]
_ = x[T_UNSET-57391]
_ = x[T_ISSET-57392]
_ = x[T_EMPTY-57393]
_ = x[T_HALT_COMPILER-57394]
_ = x[T_CLASS-57395]
_ = x[T_TRAIT-57396]
_ = x[T_INTERFACE-57397]
_ = x[T_EXTENDS-57398]
_ = x[T_IMPLEMENTS-57399]
_ = x[T_OBJECT_OPERATOR-57400]
_ = x[T_DOUBLE_ARROW-57401]
_ = x[T_LIST-57402]
_ = x[T_ARRAY-57403]
_ = x[T_CALLABLE-57404]
_ = x[T_CLASS_C-57405]
_ = x[T_TRAIT_C-57406]
_ = x[T_METHOD_C-57407]
_ = x[T_FUNC_C-57408]
_ = x[T_LINE-57409]
_ = x[T_FILE-57410]
_ = x[T_COMMENT-57411]
_ = x[T_DOC_COMMENT-57412]
_ = x[T_OPEN_TAG-57413]
_ = x[T_OPEN_TAG_WITH_ECHO-57414]
_ = x[T_CLOSE_TAG-57415]
_ = x[T_WHITESPACE-57416]
_ = x[T_START_HEREDOC-57417]
_ = x[T_END_HEREDOC-57418]
_ = x[T_DOLLAR_OPEN_CURLY_BRACES-57419]
_ = x[T_CURLY_OPEN-57420]
_ = x[T_PAAMAYIM_NEKUDOTAYIM-57421]
_ = x[T_NAMESPACE-57422]
_ = x[T_NS_C-57423]
_ = x[T_DIR-57424]
_ = x[T_NS_SEPARATOR-57425]
_ = x[T_ELLIPSIS-57426]
_ = x[T_EVAL-57427]
_ = x[T_REQUIRE-57428]
_ = x[T_REQUIRE_ONCE-57429]
_ = x[T_LOGICAL_OR-57430]
_ = x[T_LOGICAL_XOR-57431]
_ = x[T_LOGICAL_AND-57432]
_ = x[T_INSTANCEOF-57433]
_ = x[T_NEW-57434]
_ = x[T_CLONE-57435]
_ = x[T_ELSEIF-57436]
_ = x[T_ELSE-57437]
_ = x[T_ENDIF-57438]
_ = x[T_PRINT-57439]
_ = x[T_YIELD-57440]
_ = x[T_STATIC-57441]
_ = x[T_ABSTRACT-57442]
_ = x[T_FINAL-57443]
_ = x[T_PRIVATE-57444]
_ = x[T_PROTECTED-57445]
_ = x[T_PUBLIC-57446]
_ = x[T_INC-57447]
_ = x[T_DEC-57448]
_ = x[T_YIELD_FROM-57449]
_ = x[T_INT_CAST-57450]
_ = x[T_DOUBLE_CAST-57451]
_ = x[T_STRING_CAST-57452]
_ = x[T_ARRAY_CAST-57453]
_ = x[T_OBJECT_CAST-57454]
_ = x[T_BOOL_CAST-57455]
_ = x[T_UNSET_CAST-57456]
_ = x[T_COALESCE-57457]
_ = x[T_SPACESHIP-57458]
_ = x[T_NOELSE-57459]
_ = x[T_PLUS_EQUAL-57460]
_ = x[T_MINUS_EQUAL-57461]
_ = x[T_MUL_EQUAL-57462]
_ = x[T_POW_EQUAL-57463]
_ = x[T_DIV_EQUAL-57464]
_ = x[T_CONCAT_EQUAL-57465]
_ = x[T_MOD_EQUAL-57466]
_ = x[T_AND_EQUAL-57467]
_ = x[T_OR_EQUAL-57468]
_ = x[T_XOR_EQUAL-57469]
_ = x[T_SL_EQUAL-57470]
_ = x[T_SR_EQUAL-57471]
_ = x[T_COALESCE_EQUAL-57472]
_ = x[T_BOOLEAN_OR-57473]
_ = x[T_BOOLEAN_AND-57474]
_ = x[T_POW-57475]
_ = x[T_SL-57476]
_ = x[T_SR-57477]
_ = x[T_IS_IDENTICAL-57478]
_ = x[T_IS_NOT_IDENTICAL-57479]
_ = x[T_IS_EQUAL-57480]
_ = x[T_IS_NOT_EQUAL-57481]
_ = x[T_IS_SMALLER_OR_EQUAL-57482]
_ = x[T_IS_GREATER_OR_EQUAL-57483]
}
const _TokenID_name = "T_INCLUDET_INCLUDE_ONCET_EXITT_IFT_LNUMBERT_DNUMBERT_STRINGT_STRING_VARNAMET_VARIABLET_NUM_STRINGT_INLINE_HTMLT_CHARACTERT_BAD_CHARACTERT_ENCAPSED_AND_WHITESPACET_CONSTANT_ENCAPSED_STRINGT_ECHOT_DOT_WHILET_ENDWHILET_FORT_ENDFORT_FOREACHT_ENDFOREACHT_DECLARET_ENDDECLARET_AST_SWITCHT_ENDSWITCHT_CASET_DEFAULTT_BREAKT_CONTINUET_GOTOT_FUNCTIONT_FNT_CONSTT_RETURNT_TRYT_CATCHT_FINALLYT_THROWT_USET_INSTEADOFT_GLOBALT_VART_UNSETT_ISSETT_EMPTYT_HALT_COMPILERT_CLASST_TRAITT_INTERFACET_EXTENDST_IMPLEMENTST_OBJECT_OPERATORT_DOUBLE_ARROWT_LISTT_ARRAYT_CALLABLET_CLASS_CT_TRAIT_CT_METHOD_CT_FUNC_CT_LINET_FILET_COMMENTT_DOC_COMMENTT_OPEN_TAGT_OPEN_TAG_WITH_ECHOT_CLOSE_TAGT_WHITESPACET_START_HEREDOCT_END_HEREDOCT_DOLLAR_OPEN_CURLY_BRACEST_CURLY_OPENT_PAAMAYIM_NEKUDOTAYIMT_NAMESPACET_NS_CT_DIRT_NS_SEPARATORT_ELLIPSIST_EVALT_REQUIRET_REQUIRE_ONCET_LOGICAL_ORT_LOGICAL_XORT_LOGICAL_ANDT_INSTANCEOFT_NEWT_CLONET_ELSEIFT_ELSET_ENDIFT_PRINTT_YIELDT_STATICT_ABSTRACTT_FINALT_PRIVATET_PROTECTEDT_PUBLICT_INCT_DECT_YIELD_FROMT_INT_CASTT_DOUBLE_CASTT_STRING_CASTT_ARRAY_CASTT_OBJECT_CASTT_BOOL_CASTT_UNSET_CASTT_COALESCET_SPACESHIPT_NOELSET_PLUS_EQUALT_MINUS_EQUALT_MUL_EQUALT_POW_EQUALT_DIV_EQUALT_CONCAT_EQUALT_MOD_EQUALT_AND_EQUALT_OR_EQUALT_XOR_EQUALT_SL_EQUALT_SR_EQUALT_COALESCE_EQUALT_BOOLEAN_ORT_BOOLEAN_ANDT_POWT_SLT_SRT_IS_IDENTICALT_IS_NOT_IDENTICALT_IS_EQUALT_IS_NOT_EQUALT_IS_SMALLER_OR_EQUALT_IS_GREATER_OR_EQUAL"
var _TokenID_index = [...]uint16{0, 9, 23, 29, 33, 42, 51, 59, 75, 85, 97, 110, 121, 136, 161, 187, 193, 197, 204, 214, 219, 227, 236, 248, 257, 269, 273, 281, 292, 298, 307, 314, 324, 330, 340, 344, 351, 359, 364, 371, 380, 387, 392, 403, 411, 416, 423, 430, 437, 452, 459, 466, 477, 486, 498, 515, 529, 535, 542, 552, 561, 570, 580, 588, 594, 600, 609, 622, 632, 652, 663, 675, 690, 703, 729, 741, 763, 774, 780, 785, 799, 809, 815, 824, 838, 850, 863, 876, 888, 893, 900, 908, 914, 921, 928, 935, 943, 953, 960, 969, 980, 988, 993, 998, 1010, 1020, 1033, 1046, 1058, 1071, 1082, 1094, 1104, 1115, 1123, 1135, 1148, 1159, 1170, 1181, 1195, 1206, 1217, 1227, 1238, 1248, 1258, 1274, 1286, 1299, 1304, 1308, 1312, 1326, 1344, 1354, 1368, 1389, 1410}
func (i TokenID) String() string {
i -= 57346
if i < 0 || i >= TokenID(len(_TokenID_index)-1) {
return "TokenID(" + strconv.FormatInt(int64(i+57346), 10) + ")"
}
return _TokenID_name[_TokenID_index[i]:_TokenID_index[i+1]]
}

4
pkg/ast/node.go
View File

@ -16,6 +16,10 @@ func (n *Node) GetNode() *Node {
return n
}
func (n *Node) GetPosition() *position.Position {
return n.Position
}
// Root node
type Root struct {
Node

2
pkg/ast/visitor/dump.go
View File

@ -121,7 +121,7 @@ func (v *Dump) printNode(n *ast.Node) {
key := token.Position(k)
v.printIndent(v.indent + 2)
v.print("token." + key.String() + ": []token.Token{\n")
v.print("token." + key.String() + ": []*token.Token{\n")
for _, tkn := range n.Tokens[key] {
v.printIndent(v.indent + 3)

4
pkg/ast/visitor/dump_test.go
View File

@ -13,7 +13,7 @@ func ExampleDump() {
stxTree := &ast.Root{
Node: ast.Node{
Tokens: token.Collection{
token.Start: []token.Token{
token.Start: []*token.Token{
{
ID: token.T_WHITESPACE,
Value: []byte(" "),
@ -44,7 +44,7 @@ func ExampleDump() {
//&ast.Root{
// Node: ast.Node{
// Tokens: token.Collection{
// token.Start: []token.Token{
// token.Start: []*token.Token{
// {
// ID: token.T_WHITESPACE,
// Value: []byte(" "),

14
pkg/ast/visitor/filter_tokens.go Normal file
View File

@ -0,0 +1,14 @@
package visitor
import (
"github.com/z7zmey/php-parser/pkg/ast"
)
type FilterTokens struct {
Null
}
func (v *FilterTokens) EnterNode(n ast.Vertex) bool {
n.GetNode().Tokens = nil
return true
}

2
pkg/parser/parser.go
View File

@ -29,7 +29,7 @@ func Parse(src []byte, ver string, cfg Config) (ast.Vertex, error) {
return nil, err
}
lexer := scanner.NewLexer(src, ver, cfg.WithTokens, cfg.ErrorHandlerFunc)
lexer := scanner.NewLexer(src, ver, cfg.ErrorHandlerFunc)
if r == -1 {
parser = php5.NewParser(lexer, cfg.ErrorHandlerFunc)

29
pkg/position/pool.go Normal file
View File

@ -0,0 +1,29 @@
package position
const DefaultBlockSize = 1024
type Pool struct {
block []Position
off int
}
func NewPool(blockSize int) *Pool {
return &Pool{
block: make([]Position, blockSize),
}
}
func (p *Pool) Get() *Position {
if len(p.block) == 0 {
return nil
}
if len(p.block) == p.off {
p.block = make([]Position, len(p.block))
p.off = 0
}
p.off++
return &p.block[p.off-1]
}

2
pkg/printer/printer_parsed_php5_test.go
View File

@ -12,7 +12,7 @@ import (
)
func parsePhp5(src string) ast.Vertex {
lexer := scanner.NewLexer([]byte(src), "5.6", true, nil)
lexer := scanner.NewLexer([]byte(src), "5.6", nil)
php5parser := php5.NewParser(lexer, nil)
php5parser.Parse()

4
pkg/printer/printer_parsed_php7_test.go
View File

@ -29,7 +29,7 @@ abstract class Bar extends Baz
// parse
lexer := scanner.NewLexer([]byte(src), "7.4", true, nil)
lexer := scanner.NewLexer([]byte(src), "7.4", nil)
php7parser := php7.NewParser(lexer, nil)
php7parser.Parse()
@ -61,7 +61,7 @@ abstract class Bar extends Baz
}
func parse(src string) ast.Vertex {
lexer := scanner.NewLexer([]byte(src), "7.4", true, nil)
lexer := scanner.NewLexer([]byte(src), "7.4", nil)
php7parser := php7.NewParser(lexer, nil)
php7parser.Parse()

4
pkg/printer/printer_test.go
View File

@ -75,7 +75,7 @@ func TestPrinterPrintFileInlineHtml(t *testing.T) {
Expr: &ast.ExprVariable{
Node: ast.Node{
Tokens: token.Collection{
token.Start: []token.Token{
token.Start: []*token.Token{
{
ID: token.ID('$'),
Value: []byte("$"),
@ -93,7 +93,7 @@ func TestPrinterPrintFileInlineHtml(t *testing.T) {
Expr: &ast.ExprVariable{
Node: ast.Node{
Tokens: token.Collection{
token.Start: []token.Token{
token.Start: []*token.Token{
{
ID: token.ID('$'),
Value: []byte("$"),

29
pkg/token/pool.go Normal file
View File

@ -0,0 +1,29 @@
package token
const DefaultBlockSize = 1024
type Pool struct {
block []Token
off int
}
func NewPool(blockSize int) *Pool {
return &Pool{
block: make([]Token, blockSize),
}
}
func (p *Pool) Get() *Token {
if len(p.block) == 0 {
return nil
}
if len(p.block) == p.off {
p.block = make([]Token, len(p.block))
p.off = 0
}
p.off++
return &p.block[p.off-1]
}

173
pkg/token/pool_bench_test.go Normal file
View File

@ -0,0 +1,173 @@
package token
import (
"testing"
)
const amount = 100000
func BenchmarkPlain(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
for i := 0; i < amount; i++ {
buf = append(buf, &Token{})
}
}
}
func BenchmarkSlice128(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 128)
for i := 0; i < amount; i++ {
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkSlice512(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 512)
for i := 0; i < amount; i++ {
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkSlice1024(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 1024)
for i := 0; i < amount; i++ {
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkSlice2048(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 2048)
for i := 0; i < amount; i++ {
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkBlockAppend128(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 128)
for i := 0; i < amount; i++ {
if len(slc) == 128 {
slc = make([]Token, 0, 128)
}
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkBlockAppend512(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 512)
for i := 0; i < amount; i++ {
if len(slc) == 512 {
slc = make([]Token, 0, 512)
}
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkBlockAppend1024(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 1024)
for i := 0; i < amount; i++ {
if len(slc) == 1024 {
slc = make([]Token, 0, 1024)
}
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkBlockAppend2048(b *testing.B) {
for n := 0; n < b.N; n++ {
buf := make([]*Token, 0, amount)
slc := make([]Token, 0, 2048)
for i := 0; i < amount; i++ {
if len(slc) == 2048 {
slc = make([]Token, 0, 2048)
}
slc = append(slc, Token{})
buf = append(buf, &slc[len(slc)-1])
}
}
}
func BenchmarkPool128(b *testing.B) {
for n := 0; n < b.N; n++ {
pool := NewPool(128)
buf := make([]*Token, 0, amount)
for i := 0; i < amount; i++ {
buf = append(buf, pool.Get())
}
}
}
func BenchmarkPool512(b *testing.B) {
for n := 0; n < b.N; n++ {
pool := NewPool(512)
buf := make([]*Token, 0, amount)
for i := 0; i < amount; i++ {
buf = append(buf, pool.Get())
}
}
}
func BenchmarkPool1024(b *testing.B) {
for n := 0; n < b.N; n++ {
pool := NewPool(1024)
buf := make([]*Token, 0, amount)
for i := 0; i < amount; i++ {
buf = append(buf, pool.Get())
}
}
}
func BenchmarkPool2048(b *testing.B) {
for n := 0; n < b.N; n++ {
pool := NewPool(2048)
buf := make([]*Token, 0, amount)
for i := 0; i < amount; i++ {
buf = append(buf, pool.Get())
}
}
}

2
pkg/token/position.go
View File

@ -62,7 +62,7 @@ const (
CloseParenthesisToken
)
type Collection map[Position][]Token
type Collection map[Position][]*Token
func (c Collection) IsEmpty() bool {
for _, v := range c {

13
pkg/token/token.go
View File

@ -1,5 +1,7 @@
package token
import "github.com/z7zmey/php-parser/pkg/position"
//go:generate stringer -type=ID -output ./token_string.go
type ID int
@ -145,6 +147,13 @@ const (
)
type Token struct {
ID ID
Value []byte
ID ID
Value []byte
Position *position.Position
SkippedTokens []*Token
Skipped []byte
}
func (t *Token) GetPosition() *position.Position {
return t.Position
}