ext tokenizer port + cleanup unused lexer states

we basically added a mechanism to store the token stream during parsing and exposed the entire parser stack on the tokenizer extension through an opt in flag: token_get_all($src, TOKEN_PARSE). this change allows easy future language enhancements regarding context aware parsing & scanning without further maintance on the tokenizer extension while solves known inconsistencies "parseless" tokenizer extension has when it handles `__halt_compiler()` presence.
Metabolix · Apr 30, 2015 · 1107593 · 1107593
1 parent 02a9eb4
commit 1107593
Show file tree

Hide file tree

Showing 10 changed files with 510 additions and 220 deletions.
diff --git a/Zend/zend_compile.c b/Zend/zend_compile.c
@@ -30,7 +30,6 @@
 #include "zend_interfaces.h"
 #include "zend_virtual_cwd.h"
 #include "zend_multibyte.h"
-#include "zend_language_scanner.h"
 #include "zend_inheritance.h"
 
 #define SET_NODE(target, src) do { \
@@ -568,7 +567,10 @@ static int zend_add_const_name_literal(zend_op_array *op_array, zend_string *nam
 		op.constant = zend_add_literal(CG(active_op_array), &_c); \
 	} while (0)
 
-void zend_stop_lexing(void) {
+void zend_stop_lexing(void)
+{
+	if(LANG_SCNG(on_event)) LANG_SCNG(on_event)(ON_STOP, END, 0);
+
 	LANG_SCNG(yy_cursor) = LANG_SCNG(yy_limit);
 }
 

diff --git a/Zend/zend_globals.h b/Zend/zend_globals.h
@@ -249,6 +249,12 @@ struct _zend_ini_scanner_globals {
 	int scanner_mode;
 };
 
+typedef enum {
+	ON_TOKEN,
+	ON_FEEDBACK,
+	ON_STOP
+} zend_php_scanner_event;
+
 struct _zend_php_scanner_globals {
 	zend_file_handle *yy_in;
 	zend_file_handle *yy_out;
@@ -278,6 +284,9 @@ struct _zend_php_scanner_globals {
 
 	/* initial string length after scanning to first variable */
 	int scanned_string_len;
+
+	/* hooks */
+	void (* on_event)(zend_php_scanner_event event, int token, int line);
 };
 
 #endif /* ZEND_GLOBALS_H */

diff --git a/Zend/zend_language_parser.y b/Zend/zend_language_parser.y
@@ -35,7 +35,7 @@
 #include "zend_globals.h"
 #include "zend_API.h"
 #include "zend_constants.h"
-#include "zend_language_scanner_defs.h"
+#include "zend_language_scanner.h"
 
 #define YYSIZE_T size_t
 #define yytnamerr zend_yytnamerr
@@ -49,12 +49,6 @@ static YYSIZE_T zend_yytnamerr(char*, const char*);
 #define YYFREE free
 #endif
 
-#define REWIND { \
-	zend_stack_push(&LANG_SCNG(state_stack), (void *) &LANG_SCNG(yy_state)); \
-	LANG_SCNG(yy_state) = yycST_LOOKING_FOR_SEMI_RESERVED_NAME; \
-	LANG_SCNG(yy_cursor) = (unsigned char*)LANG_SCNG(yy_text); \
-	LANG_SCNG(yy_leng)   = 0; }
-
 %}
 
 %pure_parser
@@ -290,7 +284,11 @@ semi_reserved:
 
 identifier:
 		T_STRING { $$ = $1; }
-	| 	/* if */ semi_reserved { REWIND } /* and rematch as */ T_STRING { $$ = $3; }
+	| 	semi_reserved  {
+			zval zv;
+			zend_lex_tstring(&zv);
+			$$ = zend_ast_create_zval(&zv);
+		}
 ;
 
 top_statement_list:

diff --git a/Zend/zend_language_scanner.h b/Zend/zend_language_scanner.h
@@ -50,6 +50,9 @@ typedef struct _zend_lex_state {
 	zend_encoding_filter output_filter;
 	const zend_encoding *script_encoding;
 
+	/* hooks */
+	void (* on_event)(zend_php_scanner_event event, int token, int line);
+
 	zend_ast *ast;
 	zend_arena *ast_arena;
 } zend_lex_state;
@@ -66,6 +69,7 @@ ZEND_API void zend_restore_lexical_state(zend_lex_state *lex_state);
 ZEND_API int zend_prepare_string_for_scanning(zval *str, char *filename);
 ZEND_API void zend_multibyte_yyinput_again(zend_encoding_filter old_input_filter, const zend_encoding *old_encoding);
 ZEND_API int zend_multibyte_set_filter(const zend_encoding *onetime_encoding);
+ZEND_API void zend_lex_tstring(zval *zv);
 
 END_EXTERN_C()