start heredoc

c-like.l

@@ -1,79 +0,0 @@
-%{
-// Copyright (c) 2015 The golex Authors. All rights reserved.
-// Use of this source code is governed by a BSD-style
-// license that can be found in the LICENSE file.
-
-// This is an example program using golex run time library.
-package main
-
-import (
-  "bufio"
-  "go/token"
-  "io"
-  "unicode"
-
-  "github.com/cznic/golex/lex"
-)
-
-// Allocate Character classes anywhere in [0x80, 0xFF].
-const (
-	classUnicodeLeter = iota + 0x80
-	classUnicodeDigit
-	classOther
-)
-
-type lexer struct {
-	*lex.Lexer
-}
-
-func rune2Class(r rune) int {
-	if r >= 0 && r < 0x80 { // Keep ASCII as it is.
-		return int(r)
-	}
-	if unicode.IsLetter(r) {
-		return classUnicodeLeter
-	}
-	if unicode.IsDigit(r) {
-		return classUnicodeDigit
-	}
-	return classOther
-}
-
-func newLexer(src io.Reader, dst io.Writer, fName string) *lexer {
-  file := token.NewFileSet().AddFile(fName, -1, 1<<31-1)
-  lx, err := lex.New(file, bufio.NewReader(src), lex.RuneClass(rune2Class))
-  if (err != nil) { panic(err) }
-  return &lexer{lx}
-}
-
-func (l *lexer) Lex(lval *yySymType) int {
-  c := l.Enter()
-
-%}
-
-%yyc c
-%yyn c = l.Next()
-%yym l.Mark()
-
-digit         [0-9]|{unicodeDigit}
-identifier    {letter}({letter}|{digit})*
-int           {digit}+
-letter        [_a-zA-Z]|{unicodeLetter}
-unicodeDigit  \x81
-unicodeLetter \x80
-op            [-+*/]
-
-%%
-	c = l.Rule0()
-
-[ \t\r\n]+
-[/][/][^\n]+  lval.token = string(l.TokenBytes(nil)); return COMMENT
-func          lval.token = string(l.TokenBytes(nil)); return FUNC
-{identifier}  lval.token = string(l.TokenBytes(nil)); return IDENT
-{int}         lval.token = string(l.TokenBytes(nil)); return INT
-{op}          lval.token = string(l.TokenBytes(nil)); return OP
-
-%%
-	if c, ok := l.Abort(); ok { return int(c) }
-	goto yyAction
-}

php-parser.l

@@ -142,6 +142,7 @@ BNUM            0b[01]+
 EXPONENT_DNUM   (({LNUM}|{DNUM})[eE][+-]?{LNUM})
 VAR_NAME        [a-zA-Z_\x7f-\xff][a-zA-Z0-9_\x7f-\xff]*
 OPERATORS       [;:,.\[\]()|\/\^&\+-*=%!~$<>?@]
+NEW_LINE        (\r|\n|\r\n)
 
 %%
     c = l.Rule0()
@@ -256,7 +257,7 @@ OPERATORS       [;:,.\[\]()|\/\^&\+-*=%!~$<>?@]
 <PHP>\*\*                                       fmt.Println("T_POW")
 <PHP>\<\<                                       fmt.Println("T_SL")
 <PHP>\>\>                                       fmt.Println("T_SR")
-<PHP>(#|[/][/])[^\n]+                           fmt.Println("T_COMMENT"); // TODO: handle \r\n and allow ?>
+<PHP>(#|[/][/]){NEW_LINE}                       fmt.Println("T_COMMENT"); // TODO: handle \r\n and allow ?>
 <PHP>'[^']*(\\')*'                              fmt.Println("T_CONSTANT_ENCAPSED_STRING")
 <PHP>{OPERATORS}                                fmt.Printf("%s\n", l.TokenBytes(nil));
 
@@ -270,6 +271,16 @@ OPERATORS       [;:,.\[\]()|\/\^&\+-*=%!~$<>?@]
 <PROPERTY>.                                     l.ungetN(1);begin(PHP)
 
 <PHP>[\']([^\\\']*([\\][\'])*)*[\']             fmt.Printf("T_CONSTANT_ENCAPSED_STRING: %s\n", l.TokenBytes(nil));
+
+<PHP>[b]?\<\<\<[ \t]*({VAR_NAME}|([']{VAR_NAME}['])|(["]{VAR_NAME}["])){NEW_LINE}
+    tb := l.TokenBytes(nil)
+    fmt.Println(string(tb[len(tb)-2]))
+    // switch la {
+    //     case '\'' : fmt.Println("ST_NOWDOC");//begin(ST_NOWDOC)
+    //     case '"' : fmt.Println("ST_HEREDOC");//begin(ST_HEREDOC)
+    // }
+    fmt.Printf("T_START_HEREDOC: %s\n", l.TokenBytes(nil));
+
 <PHP>[b]?[\"]
     binPrefix := l.TokenBytes(nil)[0] == 'b'