feat: complete Query statements of FlinkSQL (#93)

* feat: add inlineDataValueClasue rule

* test: update tests of select statements

* feat: support flinksql window TVF grammar

* test: flink sql windown TVF statement test

* feat: support grouping sets grammar

* test: window TVF Aggregation and Group Window Aggregation tests

* test: supplemental selectAggregation with test cases

* test: add Having statement test case

* feat: support flinkSql over aggregation grammar

* test: add over aggregation grammar test cases

* test: flink sql join statement test cases

* test: flink sql set Operations grammar test cases

* test: flink sql limit clause test case

* feat: remove allPlusUid and replace with uid

* feat: support flink sql pattern recognition grammar

* test: flink sql pattern recognition tests

* feat: add flink sql with clause rule

* test: flink sql with clasue select tests

* feat: rebuild flink sql parser

Author: HaydenOrz

src/grammar/flinksql/FlinkSqlLexer.g4

@@ -273,6 +273,24 @@ ENFORCED:                     'ENFORCED';
 METADATA:                     'METADATA';
 VIRTUAL:                      'VIRTUAL';
 ZONE:                         'ZONE';
+TUMBLE:                       'TUMBLE';
+HOP:                          'HOP';
+CUMULATE:                     'CUMULATE';
+DESCRIPTOR:                   'DESCRIPTOR';
+TIMECOL:                      'TIMECOL';
+SIZE:                         'SIZE';
+OFFSET:                       'OFFSET';
+STEP:                         'STEP';
+SLIDE:                        'SLIDE';
+SESSION:                      'SESSION';
+MATCH_RECOGNIZE:              'MATCH_RECOGNIZE';
+MEASURES:                     'MEASURES';
+PATTERN:                      'PATTERN';
+ONE:                          'ONE';
+PER:                          'PER';
+KW_SKIP:                      'SKIP';
+PAST:                         'PAST';
+DEFINE:                       'DEFINE';
 
 // DATA TYPE Keywords
 
@@ -330,6 +348,8 @@ LS_BRACKET:                          '[';
 RS_BRACKET:                          ']';
 LR_BRACKET:                          '(';
 RR_BRACKET:                          ')';
+LB_BRACKET:                          '{';
+RB_BRACKET:                          '}';
 COMMA:                               ',';
 SEMICOLON:                           ';';
 AT_SIGN:                             '@';
@@ -345,8 +365,8 @@ PENCENT_SIGN:                        '%';
 DOUBLE_VERTICAL_SIGN:                '||';
 DOUBLE_HYPNEN_SIGN:                  '--';
 SLASH_SIGN:                          '/';
+QUESTION_MARK_SIGN:                  '?'; 
 DOT_ID:                              '.' ID_LITERAL_FRAG;
-PLUS_DOT_ID:                          (':' | '.') PLUS_ID_LITERAL;
 STRING_LITERAL:                      DQUOTA_STRING | SQUOTA_STRING | BQUOTA_STRING;
 DIG_LITERAL:                         DEC_DIGIT+;
 REAL_LITERAL:                        (DEC_DIGIT+)? '.' DEC_DIGIT+
@@ -355,14 +375,13 @@ REAL_LITERAL:                        (DEC_DIGIT+)? '.' DEC_DIGIT+
                                      | DEC_DIGIT+ EXPONENT_NUM_PART;
 BIT_STRING:                          BIT_STRING_L;
 ID_LITERAL:                          ID_LITERAL_FRAG;
-PLUS_ID_LITERAL:                     PLUS_ID_LITERAL_FRAG;
 FILE_PATH:                           FILE_PATH_STRING;
+DOUBLE_ARROW:                        '=>';                
 
 fragment FILE_PATH_STRING:           ([/\\] (~([/\\ ]))*)+;
 fragment JAR_FILE_PARTTARN:          '`' ( '\\'. | '``' | ~('`'|'\\'))* '`';
 fragment EXPONENT_NUM_PART:          'E' [-+]? DEC_DIGIT+;
 fragment ID_LITERAL_FRAG:            [A-Z_0-9a-z]*?[A-Z_a-z]+?[A-Z_0-9a-z]*;
-fragment PLUS_ID_LITERAL_FRAG:       [A-Z_0-9a-z*@#^$%&{}]*?[A-Z_a-z*@#^$%&{}]+?[A-Z_0-9a-z*@#^$%&{}]*;
 fragment DEC_DIGIT:                  [0-9];
 fragment DEC_LETTER:                 [A-Za-z];
 fragment DQUOTA_STRING:              '"' ( '\\'. | '""' | ~('"'| '\\') )* '"';

src/grammar/flinksql/FlinkSqlParser.g4

@@ -134,7 +134,7 @@ physicalColumnDefinition
     ;
 
 columnName
-    : plusUid | expression
+    : uid | expression
     ;
 
 columnNameList
@@ -380,7 +380,7 @@ insertMulStatement
 
 queryStatement
     : valuesCaluse
-    | WITH withItem (COMMA withItem)* queryStatement
+    | withClause queryStatement
     | '(' queryStatement ')'
     | left=queryStatement operator=(INTERSECT | UNION | EXCEPT) ALL? right=queryStatement orderByCaluse? limitClause?
     | selectClause orderByCaluse? limitClause?
@@ -391,6 +391,10 @@ valuesCaluse
     : VALUES expression (COMMA expression )*
     ;
 
+withClause
+    : WITH withItem (COMMA withItem)*
+    ;
+
 withItem
     : withItemName (LR_BRACKET columnName (COMMA columnName)* RR_BRACKET)? AS LR_BRACKET queryStatement RR_BRACKET
     ;
@@ -401,14 +405,21 @@ withItemName
 
 selectStatement
     : selectClause fromClause whereClause? groupByClause? havingClause? windowClause?
+    | selectClause fromClause matchRecognizeClause
     ;
 
 selectClause
     : SELECT setQuantifier? (ASTERISK_SIGN | projectItemDefinition (COMMA projectItemDefinition)*)
     ;
 
 projectItemDefinition
-    : expression (AS? expression)?
+    : overWindowItem
+    | expression (AS? expression)?
+    ;
+
+overWindowItem
+    : primaryExpression OVER windowSpec AS strictIdentifier
+    | primaryExpression OVER errorCapturingIdentifier AS strictIdentifier
     ;
 
 fromClause
@@ -419,6 +430,8 @@ tableExpression
     : tableReference (COMMA tableReference)*
     | tableExpression NATURAL? (LEFT | RIGHT | FULL | INNER)? OUTER? JOIN tableExpression joinCondition?
     | tableExpression CROSS JOIN tableExpression
+    | inlineDataValueClause
+    | windoTVFClause
     ;
 
 tableReference
@@ -444,6 +457,46 @@ dateTimeExpression
     : expression
     ;
 
+inlineDataValueClause
+    : LR_BRACKET valuesDefinition RR_BRACKET tableAlias
+    ;
+
+windoTVFClause
+    : TABLE LR_BRACKET windowTVFExression RR_BRACKET
+    ;
+
+windowTVFExression
+    : windoTVFName LR_BRACKET windowTVFParam (COMMA windowTVFParam)* RR_BRACKET
+    ;
+
+windoTVFName
+    : TUMBLE
+    | HOP
+    | CUMULATE
+;
+
+windowTVFParam
+    : TABLE timeAttrColumn
+    | columnDescriptor
+    | timeIntervalExpression
+    | DATA DOUBLE_ARROW TABLE timeAttrColumn
+    | TIMECOL DOUBLE_ARROW columnDescriptor
+    | timeIntervalParamName DOUBLE_ARROW timeIntervalExpression
+    ;
+
+timeIntervalParamName
+    : DATA
+    | TIMECOL
+    | SIZE
+    | OFFSET
+    | STEP
+    | SLIDE
+    ;
+
+columnDescriptor
+    : DESCRIPTOR LR_BRACKET uid RR_BRACKET
+    ;
+
 joinCondition
     : ON booleanExpression
     | USING LR_BRACKET uid (COMMA uid)* RR_BRACKET
@@ -459,27 +512,38 @@ groupByClause
 
 groupItemDefinition
     : expression
+    | groupWindowFunction
     | LR_BRACKET RR_BRACKET
     | LR_BRACKET expression (COMMA expression)* RR_BRACKET
-    | CUBE LR_BRACKET expression (COMMA expression)* RR_BRACKET
-    | ROLLUP LR_BRACKET expression (COMMA expression)* RR_BRACKET
-    | GROUPING SETS LR_BRACKET groupItemDefinition (COMMA groupItemDefinition)* RR_BRACKET
+    | groupingSetsNotaionName LR_BRACKET expression (COMMA expression)* RR_BRACKET
+    | groupingSets LR_BRACKET groupItemDefinition (COMMA groupItemDefinition)* RR_BRACKET
     ;
 
-havingClause
-    : HAVING booleanExpression
+groupingSets
+    : GROUPING SETS
     ;
 
-orderByCaluse
-    : ORDER BY orderItemDefition (COMMA orderItemDefition)*
+groupingSetsNotaionName
+    : CUBE
+    | ROLLUP
     ;
 
-orderItemDefition
-    : expression (ASC | DESC)?
+groupWindowFunction
+    : groupWindowFunctionName LR_BRACKET timeAttrColumn COMMA timeIntervalExpression RR_BRACKET
     ;
 
-limitClause
-    : LIMIT (ALL | limit=expression)
+groupWindowFunctionName
+    : TUMBLE
+    | HOP
+    | SESSION
+    ;
+
+timeAttrColumn
+    : uid
+    ;
+
+havingClause
+    : HAVING booleanExpression
     ;
 
 windowClause
@@ -492,26 +556,99 @@ namedWindow
 
 windowSpec
     : name=errorCapturingIdentifier?
-    '('
-        (ORDER BY sortItem (',' sortItem)*)?
-        (PARTITION BY expression (',' expression)*)?
+    LR_BRACKET
+        partitionByClause?
+        orderByCaluse?
         windowFrame?
-    ')'
+    RR_BRACKET
     ;
 
-sortItem
+matchRecognizeClause
+    : MATCH_RECOGNIZE 
+    LR_BRACKET 
+        partitionByClause?
+        orderByCaluse?
+        measuresClause?
+        outputMode?
+        afterMatchStrategy?
+        patternDefination?
+        patternVariablesDefination
+    RR_BRACKET ( AS? strictIdentifier )?
+    ;
+
+orderByCaluse
+    : ORDER BY orderItemDefition (COMMA orderItemDefition)*
+    ;
+
+orderItemDefition
     : expression ordering=(ASC | DESC)? (NULLS nullOrder=(LAST | FIRST))?
     ;
 
+limitClause
+    : LIMIT (ALL | limit=expression)
+    ;
+
+partitionByClause
+    : PARTITION BY expression (COMMA expression)*
+    ;
+
+quantifiers
+    : (ASTERISK_SIGN)
+    | (ADD_SIGN)
+    | (QUESTION_MARK_SIGN)
+    | (LB_BRACKET DIG_LITERAL COMMA DIG_LITERAL RB_BRACKET)
+    | (LB_BRACKET DIG_LITERAL COMMA  RB_BRACKET)
+    | (LB_BRACKET COMMA DIG_LITERAL RB_BRACKET)
+    ;
+
+measuresClause
+    : MEASURES projectItemDefinition (COMMA projectItemDefinition)*
+    ;
+
+patternDefination
+    : PATTERN 
+    LR_BRACKET
+        patternVariable+
+    RR_BRACKET 
+    withinClause?
+    ;
+
+patternVariable
+    : unquotedIdentifier quantifiers?
+    ;
+
+outputMode
+    : ALL ROWS PER MATCH
+    | ONE ROW PER MATCH
+    ;
+
+afterMatchStrategy
+    : AFTER MATCH KW_SKIP PAST LAST ROW 
+    | AFTER MATCH KW_SKIP TO NEXT ROW
+    | AFTER MATCH KW_SKIP TO LAST unquotedIdentifier
+    | AFTER MATCH KW_SKIP TO FIRST unquotedIdentifier
+    ;
+
+patternVariablesDefination
+    : DEFINE projectItemDefinition (COMMA projectItemDefinition)*
+    ;
+
 windowFrame
-    : RANGE frameBound
-    | ROWS frameBound
+    : RANGE BETWEEN timeIntervalExpression frameBound
+    | ROWS BETWEEN DIG_LITERAL frameBound
     ;
 
 frameBound
-    : expression PRECEDING
+    : PRECEDING AND CURRENT ROW
+    ;
+
+withinClause
+    : WITHIN timeIntervalExpression
     ;
 
+timeIntervalExpression
+    : INTERVAL STRING_LITERAL ID_LITERAL
+    ;
 
 // expression
 
@@ -686,10 +823,6 @@ uid
     : ID_LITERAL DOT_ID*?
     ;
 
-plusUid  // 匹配 xxx.$xx xx:xxxx 等字符
-    : (ID_LITERAL | PLUS_ID_LITERAL) (DOT_ID | PLUS_DOT_ID)*?
-    ;
-
 withOption
     : WITH tablePropertyList
     ;