Merge pull request #37 from herminiogg/develop

Develop

build.sbt

@@ -2,7 +2,7 @@ organization := "es.weso"
 
 name := "ShExML"
 
-version := "0.1.0"
+version := "0.2.0"
 
 scalaVersion := "2.12.4"
 
@@ -28,6 +28,8 @@ libraryDependencies += "info.picocli" % "picocli" % "4.0.4"
 
 libraryDependencies += "org.slf4j" % "slf4j-nop" % "1.7.27"
 
+libraryDependencies += "com.github.tototoshi" %% "scala-csv" % "1.3.6"
+
 assemblyMergeStrategy in assembly := {
   case PathList("META-INF", xs @ _*) => MergeStrategy.discard
   case x => MergeStrategy.first

src/main/java/es/weso/antlr/ShExMLLexer.g4

@@ -4,19 +4,17 @@ lexer grammar ShExMLLexer;
 package es.weso.antlr;
 }
 
-PREFIX: 'PREFIX' -> mode(DECLARATION_CONTENT) ;
-SOURCE: 'SOURCE' -> mode(DECLARATION_CONTENT) ;
-QUERY: 'QUERY' ;
-ITERATOR: 'ITERATOR' -> mode(DECLARATION_CONTENT) ;
-FIELD: 'FIELD' -> mode(DECLARATION_CONTENT) ;
-EXPRESSION: 'EXPRESSION' ;
-UNION: 'UNION' ;
-JOIN: 'JOIN' ;
-MATCHER: 'MATCHER' ;
-AS: 'AS' ;
-MATCHING: 'MATCHING' ;
-QUERY_SET: 'QUERY_SET' ;
-EXPRESSION_SET: 'EXPRESSION_SET' ;
+PREFIX: P R E F I X -> mode(DECLARATION_CONTENT) ;
+SOURCE: S O U R C E -> mode(DECLARATION_CONTENT) ;
+QUERY: Q U E R Y ;
+ITERATOR: I T E R A T O R -> mode(DECLARATION_CONTENT) ;
+FIELD: F I E L D -> mode(DECLARATION_CONTENT) ;
+EXPRESSION: E X P R E S S I O N ;
+UNION: U N I O N ;
+JOIN: J O I N ;
+MATCHER: M A T C H E R ;
+AS: A S ;
+MATCHING: M A T C H I N G ;
 LESS_SYMBOL: '<' ;
 GREATER_SYMBOL: '>' ;
 BRACKET_LEFT: '[' ;
@@ -27,23 +25,25 @@ BRACE_LEFT: '{' ;
 BRACE_RIGHT: '}' ;
 SEMICOLON: ';' ;
 DOT: '.' ;
-AT: '@' ;
 ADD: '+' ;
 COMMA: ',' ;
 QUOTE: '"' ;
 AND: '&' | 'AND' ;
-A: 'a' ;
+A: ('a' | 'A');
 STRING_OR_VAR: LETTER (LETTER | DIGIT | '_')* ;
 URI_VAR: (LETTER | DIGIT | '_')* ':' ;
 STRINGOPERATOR: '"' (LETTER | DIGIT | ALLOWED_CHARACTERS | ' ')+ '"' ;
 XMLSCHEMADATATYPE: XMLSCHEMADATATYPEPREFIX ':' ALLOWEDTYPES ;
+SHAPELINK: '@' (URI_VAR STRING_OR_VAR | '<' STRING_OR_VAR '>') ;
+LANGTAG: '@' LETTER+ ('-' (LETTER | DIGIT)+)* ;
 COMMENT: '#' ~[\r\n]* -> skip ;
 WS: [ \t\n\r] -> skip ;
 
 mode QUERY_CONTENT ;
 URL: ('http' | 'https' | 'file') '://' (LETTER | DIGIT | ALLOWED_CHARACTERS)* ;
-JSONPATH: 'jsonpath: ' (LETTER | DIGIT | ALLOWED_CHARACTERS)+ ;
-XMLPATH: 'xpath: ' (LETTER | DIGIT | ALLOWED_CHARACTERS)+ ;
+JSONPATH: J S O N P A T H ':' ;
+XMLPATH: X P A T H ':' ;
+CSVPERROW: C S V P E R R O W ;
 QUERY_PART: (LETTER | DIGIT | ALLOWED_CHARACTERS)+ ;
 GREATER_SYMBOL_QUERY: '>' -> mode(DEFAULT_MODE) ;
 WS_QUERY: [ \t\n\r] -> skip ;
@@ -66,3 +66,28 @@ fragment ALLOWEDTYPES: ('string' | 'boolean' | 'decimal' | 'integer' | 'double'
     | 'NMTOKEN' | 'Name' | 'NCName') ;
 fragment ALLOWED_CHARACTERS: '[' | ']' | '*' | '_' | '/' | '\\' | '@' | '.' | ',' | '%' | '-' | '(' | ')'
         | '?' | '=' | '&' | '#' | '$' | ':' | '^' | '\'' ;
+fragment B:('b'|'B');
+fragment C:('c'|'C');
+fragment D:('d'|'D');
+fragment E:('e'|'E');
+fragment F:('f'|'F');
+fragment G:('g'|'G');
+fragment H:('h'|'H');
+fragment I:('i'|'I');
+fragment J:('j'|'J');
+fragment K:('k'|'K');
+fragment L:('l'|'L');
+fragment M:('m'|'M');
+fragment N:('n'|'N');
+fragment O:('o'|'O');
+fragment P:('p'|'P');
+fragment Q:('q'|'Q');
+fragment R:('r'|'R');
+fragment S:('s'|'S');
+fragment T:('t'|'T');
+fragment U:('u'|'U');
+fragment V:('v'|'V');
+fragment W:('w'|'W');
+fragment X:('x'|'X');
+fragment Y:('y'|'Y');
+fragment Z:('z'|'Z');

src/main/java/es/weso/antlr/ShExMLLexer.tokens

@@ -9,63 +9,48 @@ JOIN=8
 MATCHER=9
 AS=10
 MATCHING=11
-QUERY_SET=12
-EXPRESSION_SET=13
-LESS_SYMBOL=14
-GREATER_SYMBOL=15
-BRACKET_LEFT=16
-BRACKET_RIGHT=17
-PAREN_LEFT=18
-PAREN_RIGHT=19
-BRACE_LEFT=20
-BRACE_RIGHT=21
-SEMICOLON=22
-DOT=23
-AT=24
-ADD=25
-COMMA=26
-QUOTE=27
-AND=28
-A=29
-STRING_OR_VAR=30
-URI_VAR=31
-STRINGOPERATOR=32
-XMLSCHEMADATATYPE=33
-COMMENT=34
-WS=35
-URL=36
-JSONPATH=37
-XMLPATH=38
+LESS_SYMBOL=12
+GREATER_SYMBOL=13
+BRACKET_LEFT=14
+BRACKET_RIGHT=15
+PAREN_LEFT=16
+PAREN_RIGHT=17
+BRACE_LEFT=18
+BRACE_RIGHT=19
+SEMICOLON=20
+DOT=21
+ADD=22
+COMMA=23
+QUOTE=24
+AND=25
+A=26
+STRING_OR_VAR=27
+URI_VAR=28
+STRINGOPERATOR=29
+XMLSCHEMADATATYPE=30
+SHAPELINK=31
+LANGTAG=32
+COMMENT=33
+WS=34
+URL=35
+JSONPATH=36
+XMLPATH=37
+CSVPERROW=38
 QUERY_PART=39
 GREATER_SYMBOL_QUERY=40
 WS_QUERY=41
 LESS_SYMBOL_QUERY=42
 STRING_OR_VAR_QUERY=43
 URI_VAR_QUERY=44
 WS_DECLARATION=45
-'PREFIX'=1
-'SOURCE'=2
-'QUERY'=3
-'ITERATOR'=4
-'FIELD'=5
-'EXPRESSION'=6
-'UNION'=7
-'JOIN'=8
-'MATCHER'=9
-'AS'=10
-'MATCHING'=11
-'QUERY_SET'=12
-'EXPRESSION_SET'=13
-'['=16
-']'=17
-'('=18
-')'=19
-'{'=20
-'}'=21
-';'=22
-'.'=23
-'@'=24
-'+'=25
-','=26
-'"'=27
-'a'=29
+'['=14
+']'=15
+'('=16
+')'=17
+'{'=18
+'}'=19
+';'=20
+'.'=21
+'+'=22
+','=23
+'"'=24

src/main/java/es/weso/antlr/ShExMLParser.g4

@@ -23,15 +23,15 @@ exp: union | join | stringOperation | iteratorQuery ;
 stringOperation: iteratorQuery '+' STRINGOPERATOR '+' iteratorQuery ;
 iteratorQuery: variable '.' composedVariable ;
 composedVariable: variable | variable '.' composedVariable ;
-queryClause: JSONPATH | XMLPATH ;
+queryClause: JSONPATH QUERY_PART | XMLPATH QUERY_PART | CSVPERROW ;
 join: iteratorQuery UNION iteratorQuery JOIN iteratorQuery ;
 union: leftUnionOption UNION rightUnionOption ;
 leftUnionOption: iteratorQuery | stringOperation ;
 rightUnionOption: iteratorQuery | union | stringOperation ;
 shape: tripleElement prefixVar '[' (exp | variable) ']' '{' (predicateObject ';')* predicateObject? '}' ;
 predicateObject: predicate (objectElement | shapeLink | literalValue) ;
-objectElement: prefixVar? '[' (exp | variable) (MATCHING variable)? ']' XMLSCHEMADATATYPE? ;
-shapeLink: '@' tripleElement ;
+objectElement: prefixVar? '[' (exp | variable) (MATCHING variable)? ']' (XMLSCHEMADATATYPE | LANGTAG)? ;
+shapeLink: SHAPELINK ;
 predicate: (literalValue | A) ;
 literalValue: prefixVar variable ;
 tripleElement: predicate | LESS_SYMBOL variable GREATER_SYMBOL ;