Introduce SIEL expression support

Add new synapse path type called SIEL to evaluate complex expressions which are not supported in either JSONPath or XPATH. Fixes wso2/product-micro-integrator/issues/3749
wso2 · Oct 31, 2024 · 0222047 · 0222047
1 parent 695642e
commit 0222047
Show file tree

Hide file tree

Showing 34 changed files with 3,899 additions and 8 deletions.
diff --git a/modules/core/pom.xml b/modules/core/pom.xml
@@ -145,6 +145,44 @@
 					</execution>
 				</executions>
 			</plugin>
+			<plugin>
+				<groupId>org.antlr</groupId>
+				<artifactId>antlr4-maven-plugin</artifactId>
+				<executions>
+					<execution>
+						<id>antlr</id>
+						<phase>generate-sources</phase>
+						<goals>
+							<goal>antlr4</goal>
+						</goals>
+						<configuration>
+							<sourceDirectory>${project.basedir}/src/main/antlr4</sourceDirectory>
+							<arguments>
+								<argument>-visitor</argument>
+							</arguments>
+						</configuration>
+					</execution>
+				</executions>
+			</plugin>
+			<!-- Build Helper Maven Plugin -->
+			<plugin>
+				<groupId>org.codehaus.mojo</groupId>
+				<artifactId>build-helper-maven-plugin</artifactId>
+				<executions>
+					<execution>
+						<id>add-source</id>
+						<phase>generate-sources</phase>
+						<goals>
+							<goal>add-source</goal>
+						</goals>
+						<configuration>
+							<sources>
+								<source>${project.build.directory}/generated-sources/antlr4</source>
+							</sources>
+						</configuration>
+					</execution>
+				</executions>
+			</plugin>
 
 			<!-- Attach a JAR with the test classes so that we can reuse them in other 
 				modules (see http://maven.apache.org/guides/mini/guide-attached-tests.html). -->
@@ -506,5 +544,9 @@
 			<groupId>org.apache.logging.log4j</groupId>
 			<artifactId>log4j-slf4j-impl</artifactId>
 		</dependency>
+		<dependency>
+			<groupId>org.antlr</groupId>
+			<artifactId>antlr4-runtime</artifactId>
+		</dependency>
 	</dependencies>
 </project>
diff --git a/modules/core/src/main/antlr4/org/apache/synapse/util/synapse_path/ExpressionLexer.g4 b/modules/core/src/main/antlr4/org/apache/synapse/util/synapse_path/ExpressionLexer.g4
@@ -0,0 +1,114 @@
+lexer grammar ExpressionLexer;
+
+JSONPATH_FUNCTIONS: 'contains ' | 'in' | 'nin' | 'subsetof' | 'size' | 'empty' | 'empty true' | 'empty false' | '=~';
+
+// Tokens for identifiers, operators, and keywords
+VAR: 'var';
+PAYLOAD: 'payload' | '$';
+HEADERS: 'headers';
+CONFIG: 'config';
+ATTRIBUTES: 'attributes';
+AXIS2: 'axis2';
+SYNAPSE: 'synapse';
+REGISTRY: 'registry';
+SECRET: 'secret';
+BASE64ENCODE: 'base64encode';
+BASE64DECODE: 'base64decode';
+URLENCODE: 'urlEncode';
+URLDECODE: 'urlDecode';
+NOW: 'now';
+TODAY: 'today';
+FORMATDATE: 'formatDate';
+ISNUMBER: 'isNumber';
+ISSTRING: 'isString';
+ISARRAY: 'isArray';
+ISOBJECT: 'isObject';
+ROUND: 'round';
+INTEGER: 'integer';
+FLOAT: 'float';
+STRING: 'string';
+BOOLEAN: 'boolean';
+ABS: 'abs';
+FLOOR: 'floor';
+CEIL: 'ceil';
+SQRT: 'sqrt';
+LOG: 'log';
+POW: 'pow';
+LENGTH: 'length';
+TOUPPER: 'toUpper';
+TOLOWER: 'toLower';
+SUBSTRING: 'subString';
+STARTSWITH: 'startsWith';
+ENDSWITH: 'endsWith';
+CONTAINS: 'contains';
+TRIM: 'trim';
+REPLACE: 'replace';
+SPLIT: 'split';
+AND: 'and' | '&&';
+OR: 'or' | '||';
+NOT: 'not' | '!';
+
+DOUBLE_DOT : '..';
+ASTERISK : '*';
+
+// Operators
+PLUS: '+';
+MINUS: '-';
+DIV: '/';
+MODULO: '%';
+EQ: '==';
+NEQ: '!=';
+GT: '>';
+LT: '<';
+GTE: '>=';
+LTE: '<=';
+
+// Delimiters
+LPAREN: '(';
+RPAREN: ')';
+LBRACKET: '[';
+RBRACKET: ']';
+DOT: '.';
+COMMA: ',';
+COLON: ':';
+QUOTE: '"' | '\'';
+
+// Literals
+BOOLEAN_LITERAL: 'true' | 'false';
+NUMBER: '-'? [0-9]+ ('.' [0-9]+)?;
+
+
+STRING_LITERAL : ('"' (ESC | ~["\\])* '"' | '\'' (ESC | ~['\\])* '\'');
+
+
+fragment ESC
+    :   '\\' [btnfr"'\\/]  // Basic escape sequences
+    |   UNICODE_ESC
+    |   OCTAL_ESC
+    ;
+
+fragment UNICODE_ESC
+    :   '\\' 'u' HEX_DIGIT HEX_DIGIT HEX_DIGIT HEX_DIGIT
+    ;
+
+fragment OCTAL_ESC
+    :   '\\' [0-3]? [0-7] [0-7]
+    ;
+
+fragment HEX_DIGIT
+    :   [0-9a-fA-F]
+    ;
+
+NULL_LITERAL
+    : 'null'           // Define null as a recognized keyword
+    ;
+
+// Identifiers
+ID: [a-zA-Z_][a-zA-Z_0-9]*;
+GETPROPERTY: 'getProperty';
+// Special symbols for JSONPath filter expressions
+QUESTION: '?';
+AT: '@';
+
+// Whitespace
+WS: [ \t\n\r]+ -> skip;
diff --git a/modules/core/src/main/antlr4/org/apache/synapse/util/synapse_path/ExpressionParser.g4 b/modules/core/src/main/antlr4/org/apache/synapse/util/synapse_path/ExpressionParser.g4
@@ -0,0 +1,182 @@
+parser grammar ExpressionParser;
+
+options {
+    tokenVocab = ExpressionLexer;
+}
+
+expression
+    : comparisonExpression
+    | conditionalExpression
+    | EOF
+    ;
+
+conditionalExpression
+    : comparisonExpression (QUESTION expression COLON expression)?
+    ;
+
+comparisonExpression
+    : logicalExpression ( (GT | LT | GTE | LTE | EQ | NEQ) logicalExpression )*
+    | logicalExpression (EQ | NEQ) NULL_LITERAL // Allow comparison to null
+    ;
+
+logicalExpression
+    : arithmeticExpression (AND logicalExpression | OR logicalExpression)?
+    ;
+
+arithmeticExpression
+    : term ( (PLUS | MINUS) term )*
+    ;
+
+term
+    : factor ( (ASTERISK | DIV | MODULO) factor )*
+    ;
+
+factor
+    : literal
+    | functionCall
+    | variableAccess
+    | payloadAccess
+    | headerAccess
+    | configAccess
+    | attributeAccess
+    | LPAREN expression RPAREN
+    ;
+
+configAccess
+    : CONFIG DOT headerName
+    ;
+
+headerAccess
+    : HEADERS DOT headerName
+    ;
+
+headerName
+    : ID                             // Standard identifier
+    | STRING_LITERAL                 // Quoted string literal to allow spaces
+    ;
+
+attributeAccess
+    : ATTRIBUTES (DOT AXIS2 DOT propertyName       // Access syntax for attributes.axis2.property
+                | DOT SYNAPSE DOT propertyName)    // Access syntax for attributes.synapse.property
+    ;
+
+propertyName
+    : ID                             // Standard identifier
+    | STRING_LITERAL                 // Quoted string literal to allow spaces
+    ;
+
+literal
+    : arrayLiteral
+    | BOOLEAN_LITERAL
+    | NUMBER
+    | STRING_LITERAL
+    | NULL_LITERAL
+    ;
+
+variableAccess
+    : VAR ( DOT ID  // Dot notation: var.variableName
+          | (DOT)? LBRACKET STRING_LITERAL RBRACKET  // Bracket notation: var["variableName"]
+          )
+      ( (DOUBLE_DOT ASTERISK    // Handles recursive descent and wildcard, like var.variableName..*
+        | DOT ASTERISK          // Handles wildcard after dot notation, like var.variableName.*
+        | DOT ID
+        | LBRACKET arrayIndex RBRACKET   // Handles array access, like var.variableName[...]
+        | DOT LBRACKET arrayIndex RBRACKET   // Handles dot followed by bracket notation, like var.variableName.["property"]
+        )*
+      )
+    ;
+
+arrayLiteral
+    : LBRACKET (expression (COMMA expression)*)? RBRACKET // Array with zero or more literals, separated by commas
+    ;
+
+payloadAccess
+    : PAYLOAD ( (DOUBLE_DOT ASTERISK
+    | DOUBLE_DOT ID
+    | DOT ID
+    | LBRACKET arrayIndex RBRACKET
+    | DOT LBRACKET arrayIndex RBRACKET
+    | DOT ASTERISK)*
+    | DOUBLE_DOT ID (LBRACKET arrayIndex RBRACKET)? )
+    ;
+
+arrayIndex
+    : NUMBER
+    | STRING_LITERAL
+    | expression
+    | multipleArrayIndices
+    | sliceArrayIndex
+    | expression ( (PLUS | MINUS | MULT | DIV ) expression)*
+    | ASTERISK
+    | QUESTION? filterExpression
+    ;
+
+multipleArrayIndices
+    : expression (COMMA expression)+
+    ;
+
+sliceArrayIndex
+    : signedExpressions? COLON signedExpressions? (COLON signedExpressions?)?
+    ;
+
+signedExpressions
+    : MINUS? expression
+    ;
+
+filterExpression
+    : (filterComponent)+
+    ;
+
+filterComponent
+    : variableAccess
+    | payloadAccess
+    | stringOrOperator
+    | headerAccess
+    | configAccess
+    | attributeAccess
+    | functionCall
+    ;
+
+stringOrOperator
+    : QUESTION | AT | JSONPATH_FUNCTIONS| STRING_LITERAL |NUMBER | BOOLEAN_LITERAL | ID | GT | LT | GTE | LTE | EQ | NEQ
+    | PLUS | MINUS | MULT | DIV | LPAREN | RPAREN | DOT | COMMA | COLON | WS | AND | OR | NOT | ASTERISK
+    ;
+
+
+
+functionCall
+    : LENGTH LPAREN expression RPAREN
+    | TOUPPER LPAREN expression RPAREN
+    | TOLOWER LPAREN expression RPAREN
+    | SUBSTRING LPAREN expression COMMA expression (COMMA expression)? RPAREN
+    | STARTSWITH LPAREN expression COMMA expression RPAREN
+    | ENDSWITH LPAREN expression COMMA expression RPAREN
+    | CONTAINS LPAREN expression COMMA expression RPAREN
+    | TRIM LPAREN expression RPAREN
+    | REPLACE LPAREN expression COMMA expression COMMA expression RPAREN
+    | SPLIT LPAREN expression COMMA expression RPAREN
+    | ABS LPAREN expression RPAREN
+    | FLOOR LPAREN expression RPAREN
+    | CEIL LPAREN expression RPAREN
+    | SQRT LPAREN expression RPAREN
+    | LOG LPAREN expression RPAREN
+    | POW LPAREN expression COMMA expression RPAREN
+    | REGISTRY LPAREN expression RPAREN (DOT GETPROPERTY LPAREN expression RPAREN)?
+    | SECRET LPAREN expression RPAREN
+    | BASE64ENCODE LPAREN expression (COMMA expression)? RPAREN
+    | BASE64DECODE LPAREN expression RPAREN
+    | URLENCODE LPAREN expression (COMMA expression)? RPAREN
+    | URLDECODE LPAREN expression RPAREN
+    | ISNUMBER LPAREN expression RPAREN
+    | ISSTRING LPAREN expression RPAREN
+    | ISARRAY LPAREN expression RPAREN
+    | ISOBJECT LPAREN expression RPAREN
+    | NOW LPAREN RPAREN
+    | TODAY LPAREN STRING_LITERAL RPAREN
+    | FORMATDATE LPAREN expression COMMA STRING_LITERAL RPAREN
+    | ROUND LPAREN expression RPAREN
+    | INTEGER LPAREN expression RPAREN
+    | FLOAT LPAREN expression RPAREN
+    | STRING LPAREN expression RPAREN
+    | BOOLEAN LPAREN expression RPAREN
+    ;