Finalize separation of parsers

ortus-boxlang · Mar 27, 2024 · 0cd0123 · 0cd0123
1 parent 8037850
commit 0cd0123
Show file tree

Hide file tree

Showing 78 changed files with 5,384 additions and 744 deletions.
diff --git a/modules/test/bifs/Hello.cfc → modules/test/bifs/Hello.bx b/modules/test/bifs/Hello.cfc → modules/test/bifs/Hello.bx
@@ -19,7 +19,7 @@
  */
 @BoxBIF 'moduleHelloWorld'
 @BoxMember { "string" : { name : "foo" }, "array" : {} }
-component{
+class{
 
 	/**
 	 * The execution of this BIF with amazing BoxLang arguments

diff --git a/modules/test/bifs/ModuleNow.cfc → modules/test/bifs/ModuleNow.bx b/modules/test/bifs/ModuleNow.cfc → modules/test/bifs/ModuleNow.bx
@@ -2,7 +2,7 @@
  * No aliases
  */
 @BoxMember "string"
-component{
+class{
 
 	/**
 	 * The execution of this BIF with amazing BoxLang arguments

diff --git a/src/main/antlr/BaseScriptGrammar.g4 b/src/main/antlr/BaseScriptGrammar.g4
@@ -9,30 +9,18 @@ options {
  	public ortus.boxlang.runtime.services.ComponentService componentService = ortus.boxlang.runtime.BoxRuntime.getInstance().getComponentService();
  }
 
-// This is the top level rule, which allow imports always, followed by a component, or an interface, or just a bunch of statements.
-script:
-	importStatement* (
-		boxClass
-		| interface
-		| functionOrStatement*
-	)
-	| EOF;
-
 // marks the end of simple statements (no body)
 eos: SEMICOLON;
 
-// TODO: This belongs only in the BL grammar. import java:foo.bar.Baz as myAlias;
-importStatement:
-	IMPORT (prefix = identifier COLON)? fqn (DOT STAR)? (
-		AS alias = identifier
-	)? eos?;
+// This is the top level rule, which allow a component, or an interface, or just a bunch of statements.
+script: ( boxClass | interface | functionOrStatement*) | EOF;
 
-// include "myFile.cfm";
+// include "myFile.bxm";
 include: INCLUDE expression;
 
 // class {}
 boxClass:
-	javadoc? (preannotation)* ABSTRACT? COMPONENT postannotation* LBRACE property*
+	javadoc? (preannotation)* ABSTRACT? CLASS_NAME postannotation* LBRACE property*
 		functionOrStatement* RBRACE;
 
 interface:
@@ -60,7 +48,8 @@ functionParam: (REQUIRED)? (type)? identifier (
 		EQUALSIGN expression
 	)? postannotation*;
 
-// @MyAnnotation "value" true
+// @MyAnnotation "value". This is BL specific, so it's disabled in the CF grammar, but defined here
+// in the base grammar for better rule reuse.
 preannotation: AT fqn (literalExpression)*;
 
 // foo=bar baz="bum"
@@ -84,7 +73,7 @@ type:
 	NUMERIC
 	| STRING
 	| BOOLEAN
-	| COMPONENT
+	| CLASS_NAME
 	| INTERFACE
 	| ARRAY
 	| STRUCT
@@ -164,29 +153,13 @@ component:
 	// http url="google.com" {}
 	(componentName componentAttributes statementBlock)
 	// http url="google.com";
-	| (componentName componentAttributes eos)
-	// cfhttp( url="google.com" ){}   -- Only needed for CF parser
-	| (
-		prefixedIdentifier LPAREN delimitedComponentAttributes? RPAREN statementBlock
-	)
-	// cfhttp( url="google.com" )   -- Only needed for CF parser
-	| (
-		prefixedIdentifier LPAREN delimitedComponentAttributes? RPAREN
-	);
-
-// cfSomething
-prefixedIdentifier: PREFIXEDIDENTIFIER;
+	| (componentName componentAttributes eos);
 
 // foo="bar" baz="bum" qux
 componentAttributes: (componentAttribute)*;
 
 componentAttribute: identifier (EQUALSIGN expression)?;
 
-// foo="bar", baz="bum"
-delimitedComponentAttributes: (componentAttribute) (
-		COMMA componentAttribute
-	)*;
-
 /*
  ++foo
  foo++
@@ -351,8 +324,7 @@ reservedKeyword:
 	| CASE
 	| CASTAS
 	| CATCH
-	| CLASS
-	| COMPONENT
+	| CLASS_NAME
 	| CONTAIN
 	| CONTAINS
 	| CONTINUE
@@ -418,9 +390,7 @@ reservedKeyword:
 	| LE
 	| NEQ
 	| NOT
-	| OR
-	| PREFIX
-	| PREFIXEDIDENTIFIER;
+	| OR;
 
 // ANY NEW LEXER RULES IN DEFAULT MODE FOR WORDS NEED ADDED HERE
 
@@ -430,7 +400,7 @@ scope: REQUEST | VARIABLES | SERVER;
 
 /*
  ```
- <cfset components="here">
+ <bx:set components="here">
  ```
  */
 componentIsland:
@@ -501,8 +471,8 @@ structMember:
 	identifier (COLON | EQUALSIGN) expression
 	| stringLiteral (COLON | EQUALSIGN) expression;
 
-// +foo -bar
-unary: (MINUS | PLUS | BITWISE_COMPLEMENT) expression;
+// +foo -bar b~baz
+unary: (MINUS | PLUS | bitwiseCompliment) expression;
 
 // new java:String( param1 )
 new:
@@ -536,11 +506,12 @@ notTernaryExpression:
 	| notTernaryExpression (STAR | SLASH | PERCENT | BACKSLASH) notTernaryExpression
 	| notTernaryExpression (PLUS | MINUS | MOD) notTernaryExpression
 	| notTernaryExpression (
-		BITWISE_SIGNED_LEFT_SHIFT
-		| BITWISE_SIGNED_RIGHT_SHIFT
-		| BITWISE_UNSIGNED_RIGHT_SHIFT
+		bitwiseSignedLeftShift
+		| bitwiseSignedRightShift
+		| bitwiseUnsignedRightShift
 	) notTernaryExpression
-	| notTernaryExpression (XOR | INSTANCEOF) notTernaryExpression
+	| notTernaryExpression XOR notTernaryExpression
+	| notTernaryExpression instanceOf notTernaryExpression
 	| notTernaryExpression (AMPERSAND notTernaryExpression)+
 	| notTernaryExpression (
 		eq
@@ -559,18 +530,44 @@ notTernaryExpression:
 		| NOT CONTAINS
 		| TEQ
 	) notTernaryExpression // Comparision
-	| notTernaryExpression BITWISE_AND notTernaryExpression // Bitwise AND operator
-	| notTernaryExpression BITWISE_XOR notTernaryExpression // Bitwise XOR operator
-	| notTernaryExpression BITWISE_OR notTernaryExpression // Bitwise OR operator
+	| notTernaryExpression bitwiseAnd notTernaryExpression // Bitwise AND operator
+	| notTernaryExpression bitwiseXOR notTernaryExpression // Bitwise XOR operator
+	| notTernaryExpression bitwiseOr notTernaryExpression // Bitwise OR operator
 	| notTernaryExpression ELVIS notTernaryExpression // Elvis operator
 	| notTernaryExpression IS notTernaryExpression // IS operator
-	| notTernaryExpression CASTAS notTernaryExpression // CastAs operator
-	| notTernaryExpression INSTANCEOF notTernaryExpression // InstanceOf operator
+	| notTernaryExpression castAs notTernaryExpression
 	| notTernaryExpression DOES NOT CONTAIN notTernaryExpression
 	| notOrBang notTernaryExpression
 	| notTernaryExpression (and | or) notTernaryExpression;
 // Logical
 
+// foo b<< bar
+bitwiseSignedLeftShift: BITWISE_SIGNED_LEFT_SHIFT;
+
+// foo b>> bar
+bitwiseSignedRightShift: BITWISE_SIGNED_RIGHT_SHIFT;
+
+// foo b>>> bar
+bitwiseUnsignedRightShift: BITWISE_UNSIGNED_RIGHT_SHIFT;
+
+// foo b& bar
+bitwiseAnd: BITWISE_AND;
+
+// foo b^ bar
+bitwiseXOR: BITWISE_XOR;
+
+// foo |b bar
+bitwiseOr: BITWISE_OR;
+
+// b~baz
+bitwiseCompliment: BITWISE_COMPLEMENT;
+
+// foo castAs bar
+castAs: CASTAS;
+
+// foo instanceOf bar
+instanceOf: INSTANCEOF;
+
 and: AND | AMPAMP;
 
 eq: EQ | EQUAL | EQEQ;

diff --git a/src/main/antlr/BaseScriptLexer.g4 b/src/main/antlr/BaseScriptLexer.g4
@@ -18,8 +18,6 @@ BREAK: 'BREAK';
 CASE: 'CASE';
 CASTAS: 'CASTAS';
 CATCH: 'CATCH';
-CLASS: 'CLASS';
-COMPONENT: 'COMPONENT';
 CONTAIN: 'CONTAIN';
 CONTAINS: 'CONTAINS';
 CONTINUE: 'CONTINUE';
@@ -78,6 +76,9 @@ WHEN: 'WHEN';
 WHILE: 'WHILE';
 XOR: 'XOR';
 
+// ALWAYS OVERRIDEN IN CF OR BOX LEXERS
+CLASS_NAME: 'sdf';
+
 AND: 'AND';
 AMPAMP: '&&';
 
@@ -144,7 +145,6 @@ MODEQUAL: '%=';
 PLUS: '+';
 PLUSPLUS: '++';
 TEQ: '===';
-PREFIX: 'CF';
 
 // BITWISE OPERATORS
 BITWISE_OR: 'b|';
@@ -185,7 +185,6 @@ FLOAT_LITERAL_DECIMAL_ONLY_E_NOTATION: DOT DIGIT+ E_NOTATION;
 FLOAT_LITERAL_DECIMAL_ONLY: DOT DIGIT+;
 
 INTEGER_LITERAL: DIGIT+;
-PREFIXEDIDENTIFIER: PREFIX IDENTIFIER;
 IDENTIFIER: [a-z_$]+ ( [_]+ | [a-z]+ | DIGIT)*;
 
 COMPONENT_ISLAND_START: '```' -> pushMode(componentIsland);

diff --git a/src/main/antlr/BaseTemplateLexer.g4 b/src/main/antlr/BaseTemplateLexer.g4
@@ -33,10 +33,6 @@ COMMENT: '<!---' .*? '--->' -> channel(HIDDEN);
 
 WS: (' ' | '\t' | '\r'? '\n')+;
 
-SCRIPT_OPEN: '<cfscript' .*? '>' -> pushMode(XFSCRIPT);
-
-OUTPUT_START:
-	'<cfoutput' -> pushMode(POSSIBLE_COMPONENT), pushMode(COMPONENT_MODE), pushMode(OUTPUT_MODE);
 COMPONENT_OPEN: '<' -> pushMode(POSSIBLE_COMPONENT);
 
 HASHHASH: '##' -> type(CONTENT_TEXT);
@@ -49,9 +45,8 @@ CONTENT_TEXT: ~[<#]+;
 // *********************************************************************************************************************
 mode POSSIBLE_COMPONENT;
 
-PREFIX: 'cf' -> pushMode(COMPONENT_MODE);
-SLASH_PREFIX: '/cf' -> pushMode(END_COMPONENT);
-ANY: . -> type(CONTENT_TEXT), popMode;
+// This mode is overridden in the CF and Box lexers
+IGNORED: [.];
 
 // *********************************************************************************************************************
 mode COMPONENT_MODE;
@@ -108,6 +103,12 @@ fragment COMPONENT_NameChar:
 
 fragment COMPONENT_NameStartChar: [a-z_];
 
+// *********************************************************************************************************************
+mode XFSCRIPT;
+
+// This mode is overridden in the CF and Box lexers
+IGNORED2: [.];
+
 // *********************************************************************************************************************
 mode OUTPUT_MODE;
 
@@ -155,15 +156,6 @@ COMPONENT_NAME2:
 COMPONENT_CLOSE2:
 	'>' -> popMode, popMode, type(COMPONENT_CLOSE);
 
-// *********************************************************************************************************************
-mode XFSCRIPT;
-
-fragment COMPONENT_WHITESPACE2: [ \t\r\n]*;
-SCRIPT_END_BODY:
-	'</' COMPONENT_WHITESPACE2 'cfscript' COMPONENT_WHITESPACE2 '>' -> popMode;
-
-SCRIPT_BODY: .+?;
-
 // *********************************************************************************************************************
 mode ATTVALUE;
 

diff --git a/src/main/antlr/BoxScriptGrammar.g4 b/src/main/antlr/BoxScriptGrammar.g4
@@ -4,4 +4,19 @@ options {
 	tokenVocab = BoxScriptLexer;
 }
 
-import BaseScriptGrammar;
+import BaseScriptGrammar;
+
+// This is the top level rule, which allow imports always, followed by a component, or an interface, or just a bunch of statements.
+script:
+	importStatement* (
+		boxClass
+		| interface
+		| functionOrStatement*
+	)
+	| EOF;
+
+// TODO: This belongs only in the BL grammar. import java:foo.bar.Baz as myAlias;
+importStatement:
+	IMPORT (prefix = identifier COLON)? fqn (DOT STAR)? (
+		AS alias = identifier
+	)? eos?;
diff --git a/src/main/antlr/BoxScriptLexer.g4 b/src/main/antlr/BoxScriptLexer.g4
@@ -4,4 +4,6 @@ options {
 	caseInsensitive = true;
 }
 
-import BaseScriptLexer;
+import BaseScriptLexer;
+
+CLASS_NAME: 'CLASS';
diff --git a/src/main/antlr/BoxTemplateLexer.g4 b/src/main/antlr/BoxTemplateLexer.g4
@@ -4,4 +4,25 @@ options {
 	caseInsensitive = true;
 }
 
-import BaseTemplateLexer;
+import BaseTemplateLexer;
+
+SCRIPT_OPEN: '<bx:script' .*? '>' -> pushMode(XFSCRIPT);
+
+OUTPUT_START:
+	'<bx:output' -> pushMode(POSSIBLE_COMPONENT), pushMode(COMPONENT_MODE), pushMode(OUTPUT_MODE);
+
+// *********************************************************************************************************************
+mode XFSCRIPT;
+
+fragment COMPONENT_WHITESPACE2: [ \t\r\n]*;
+SCRIPT_END_BODY:
+	'</' COMPONENT_WHITESPACE2 'bx:script' COMPONENT_WHITESPACE2 '>' -> popMode;
+
+SCRIPT_BODY: .+?;
+
+// *********************************************************************************************************************
+mode POSSIBLE_COMPONENT;
+
+PREFIX: 'bx:' -> pushMode(COMPONENT_MODE);
+SLASH_PREFIX: '/bx:' -> pushMode(END_COMPONENT);
+ANY: . -> type(CONTENT_TEXT), popMode;