[DYOD] Data Definition Language Project (#177)

This pull request sums up all the work for project 1 of this term "Develop your Own Database" seminar. The project's high-level goal is to make the schema SQL of hyrise executable.
Specifically, this pull request - once finished - will include the extension of the parser to support CREATE INDEX and DROP INDEX statements, column and table key constraints, a subset of ALTER TABLE statements, and extended type parsing within column definitions.

Co-authored-by: Moritz Spranger <[email protected]>
Co-authored-by: Jostafarr <[email protected]>
Co-authored-by: mweisgut <[email protected]>
Co-authored-by: Marcel Weisgut <[email protected]>

Author: 4 people

src/parser/bison_parser.cpp

@@ -2,9 +2,9 @@
 #define hsql_HEADER_H 1
 #define hsql_IN_HEADER 1
 
-#line 6 "flex_lexer.h"
+#line 5 "flex_lexer.h"
 
-#line 8 "flex_lexer.h"
+#line 7 "flex_lexer.h"
 
 #define  YY_INT_ALIGNED short int
 
@@ -730,9 +730,9 @@ extern int yylex \
 #undef yyTABLES_NAME
 #endif
 
-#line 265 "flex_lexer.l"
+#line 270 "flex_lexer.l"
 
 
-#line 737 "flex_lexer.h"
+#line 736 "flex_lexer.h"
 #undef hsql_IN_HEADER
 #endif /* hsql_HEADER_H */

src/parser/bison_parser.h

@@ -56,165 +56,170 @@ static thread_local std::stringstream strbuf;
  ***************************/
 %%
 
---				BEGIN(COMMENT);
-<COMMENT>[^\n]*	/* skipping comment content until a end of line is read */;
-<COMMENT>\n 	BEGIN(INITIAL);
+--              BEGIN(COMMENT);
+<COMMENT>[^\n]* /* skipping comment content until a end of line is read */;
+<COMMENT>\n     BEGIN(INITIAL);
 
-[ \t\n]+	/* skip whitespace */;
+[ \t\n]+        /* skip whitespace */;
 
-DEALLOCATE	TOKEN(DEALLOCATE)
-PARAMETERS	TOKEN(PARAMETERS)
-INTERSECT	TOKEN(INTERSECT)
-TEMPORARY	TOKEN(TEMPORARY)
-TIMESTAMP	TOKEN(TIMESTAMP)
-DESCRIBE	TOKEN(DESCRIBE)
-DISTINCT	TOKEN(DISTINCT)
-NVARCHAR	TOKEN(NVARCHAR)
-RESTRICT	TOKEN(RESTRICT)
-TRUNCATE	TOKEN(TRUNCATE)
-ANALYZE		TOKEN(ANALYZE)
-BETWEEN		TOKEN(BETWEEN)
-CASCADE		TOKEN(CASCADE)
-COLUMNS		TOKEN(COLUMNS)
-CONTROL		TOKEN(CONTROL)
-DEFAULT		TOKEN(DEFAULT)
-EXECUTE		TOKEN(EXECUTE)
-EXPLAIN		TOKEN(EXPLAIN)
-INTEGER		TOKEN(INTEGER)
-NATURAL		TOKEN(NATURAL)
-PREPARE		TOKEN(PREPARE)
-PRIMARY		TOKEN(PRIMARY)
-SCHEMAS		TOKEN(SCHEMAS)
-SPATIAL		TOKEN(SPATIAL)
-VARCHAR		TOKEN(VARCHAR)
-VIRTUAL		TOKEN(VIRTUAL)
-BEFORE		TOKEN(BEFORE)
-COLUMN		TOKEN(COLUMN)
-CREATE		TOKEN(CREATE)
-DELETE		TOKEN(DELETE)
-DIRECT		TOKEN(DIRECT)
-DOUBLE		TOKEN(DOUBLE)
-ESCAPE		TOKEN(ESCAPE)
-EXCEPT		TOKEN(EXCEPT)
-EXISTS		TOKEN(EXISTS)
-EXTRACT 	TOKEN(EXTRACT)
-CAST		TOKEN(CAST)
-FORMAT		TOKEN(FORMAT)
-GLOBAL		TOKEN(GLOBAL)
-HAVING		TOKEN(HAVING)
-IMPORT		TOKEN(IMPORT)
-INSERT		TOKEN(INSERT)
-ISNULL		TOKEN(ISNULL)
-OFFSET		TOKEN(OFFSET)
-RENAME		TOKEN(RENAME)
-SCHEMA		TOKEN(SCHEMA)
-SELECT		TOKEN(SELECT)
-SORTED		TOKEN(SORTED)
-TABLES		TOKEN(TABLES)
-UNIQUE		TOKEN(UNIQUE)
-UNLOAD		TOKEN(UNLOAD)
-UPDATE		TOKEN(UPDATE)
-VALUES		TOKEN(VALUES)
-AFTER		TOKEN(AFTER)
-ALTER		TOKEN(ALTER)
-ARRAY		TOKEN(ARRAY)
-CROSS		TOKEN(CROSS)
-DELTA		TOKEN(DELTA)
-FLOAT		TOKEN(FLOAT)
-GROUP		TOKEN(GROUP)
-INDEX		TOKEN(INDEX)
-INNER		TOKEN(INNER)
-LIMIT		TOKEN(LIMIT)
-LOCAL		TOKEN(LOCAL)
-MERGE		TOKEN(MERGE)
-MINUS		TOKEN(MINUS)
-ORDER		TOKEN(ORDER)
-OUTER		TOKEN(OUTER)
-RIGHT		TOKEN(RIGHT)
-TABLE		TOKEN(TABLE)
-UNION		TOKEN(UNION)
-USING		TOKEN(USING)
-WHERE		TOKEN(WHERE)
-CALL		TOKEN(CALL)
-CASE		TOKEN(CASE)
-CHAR		TOKEN(CHAR)
-COPY		TOKEN(COPY)
-DATE		TOKEN(DATE)
-DATETIME	TOKEN(DATETIME)
-DESC		TOKEN(DESC)
-DROP		TOKEN(DROP)
-ELSE		TOKEN(ELSE)
-FILE		TOKEN(FILE)
-FROM		TOKEN(FROM)
-FULL		TOKEN(FULL)
-HASH		TOKEN(HASH)
-HINT		TOKEN(HINT)
-INTO		TOKEN(INTO)
-JOIN		TOKEN(JOIN)
-LEFT		TOKEN(LEFT)
-LIKE		TOKEN(LIKE)
-ILIKE		TOKEN(ILIKE)
-LOAD		TOKEN(LOAD)
-LONG		TOKEN(LONG)
-NULL		TOKEN(NULL)
-PLAN		TOKEN(PLAN)
-SHOW		TOKEN(SHOW)
-TEXT		TOKEN(TEXT)
-THEN		TOKEN(THEN)
-TIME		TOKEN(TIME)
-VIEW		TOKEN(VIEW)
-WHEN		TOKEN(WHEN)
-WITH		TOKEN(WITH)
-ADD			TOKEN(ADD)
-ALL			TOKEN(ALL)
-AND			TOKEN(AND)
-ASC			TOKEN(ASC)
-END			TOKEN(END)
-FOR			TOKEN(FOR)
-INT			TOKEN(INT)
-KEY			TOKEN(KEY)
-NOT			TOKEN(NOT)
-OFF			TOKEN(OFF)
-SET			TOKEN(SET)
-TOP			TOKEN(TOP)
-AS			TOKEN(AS)
-BY			TOKEN(BY)
-IF			TOKEN(IF)
-IN			TOKEN(IN)
-IS			TOKEN(IS)
-OF			TOKEN(OF)
-ON			TOKEN(ON)
-OR			TOKEN(OR)
-TO			TOKEN(TO)
-SECOND		TOKEN(SECOND)
-MINUTE		TOKEN(MINUTE)
-HOUR		TOKEN(HOUR)
-DAY			TOKEN(DAY)
-MONTH		TOKEN(MONTH)
-YEAR		TOKEN(YEAR)
-SECONDS		TOKEN(SECONDS)
-MINUTES		TOKEN(MINUTES)
-HOURS		TOKEN(HOURS)
-DAYS		TOKEN(DAYS)
-MONTHS		TOKEN(MONTHS)
-YEARS		TOKEN(YEARS)
-TRUE		TOKEN(TRUE)
-FALSE		TOKEN(FALSE)
+DEALLOCATE  TOKEN(DEALLOCATE)
+PARAMETERS  TOKEN(PARAMETERS)
+INTERSECT   TOKEN(INTERSECT)
+TEMPORARY   TOKEN(TEMPORARY)
+TIMESTAMP   TOKEN(TIMESTAMP)
+DESCRIBE    TOKEN(DESCRIBE)
+DISTINCT    TOKEN(DISTINCT)
+NVARCHAR    TOKEN(NVARCHAR)
+RESTRICT    TOKEN(RESTRICT)
+TRUNCATE    TOKEN(TRUNCATE)
+ANALYZE     TOKEN(ANALYZE)
+BETWEEN     TOKEN(BETWEEN)
+CASCADE     TOKEN(CASCADE)
+COLUMNS     TOKEN(COLUMNS)
+CONTROL     TOKEN(CONTROL)
+DEFAULT     TOKEN(DEFAULT)
+EXECUTE     TOKEN(EXECUTE)
+EXPLAIN     TOKEN(EXPLAIN)
+INTEGER     TOKEN(INTEGER)
+NATURAL     TOKEN(NATURAL)
+PREPARE     TOKEN(PREPARE)
+PRIMARY     TOKEN(PRIMARY)
+SCHEMAS     TOKEN(SCHEMAS)
+SPATIAL     TOKEN(SPATIAL)
+VARCHAR     TOKEN(VARCHAR)
+CHARACTER   TOKEN(CHARACTER)
+VARYING     TOKEN(VARYING)
+TIME        TOKEN(TIME)
+DECIMAL     TOKEN(DECIMAL)
+REAL        TOKEN(REAL)
+VIRTUAL     TOKEN(VIRTUAL)
+BEFORE      TOKEN(BEFORE)
+COLUMN      TOKEN(COLUMN)
+CREATE      TOKEN(CREATE)
+DELETE      TOKEN(DELETE)
+DIRECT      TOKEN(DIRECT)
+DOUBLE      TOKEN(DOUBLE)
+ESCAPE      TOKEN(ESCAPE)
+EXCEPT      TOKEN(EXCEPT)
+EXISTS      TOKEN(EXISTS)
+EXTRACT     TOKEN(EXTRACT)
+CAST        TOKEN(CAST)
+FORMAT      TOKEN(FORMAT)
+GLOBAL      TOKEN(GLOBAL)
+HAVING      TOKEN(HAVING)
+IMPORT      TOKEN(IMPORT)
+INSERT      TOKEN(INSERT)
+ISNULL      TOKEN(ISNULL)
+OFFSET      TOKEN(OFFSET)
+RENAME      TOKEN(RENAME)
+SCHEMA      TOKEN(SCHEMA)
+SELECT      TOKEN(SELECT)
+SORTED      TOKEN(SORTED)
+TABLES      TOKEN(TABLES)
+UNIQUE      TOKEN(UNIQUE)
+UNLOAD      TOKEN(UNLOAD)
+UPDATE      TOKEN(UPDATE)
+VALUES      TOKEN(VALUES)
+AFTER       TOKEN(AFTER)
+ALTER       TOKEN(ALTER)
+ARRAY       TOKEN(ARRAY)
+CROSS       TOKEN(CROSS)
+DELTA       TOKEN(DELTA)
+FLOAT       TOKEN(FLOAT)
+GROUP       TOKEN(GROUP)
+INDEX       TOKEN(INDEX)
+INNER       TOKEN(INNER)
+LIMIT       TOKEN(LIMIT)
+LOCAL       TOKEN(LOCAL)
+MERGE       TOKEN(MERGE)
+MINUS       TOKEN(MINUS)
+ORDER       TOKEN(ORDER)
+OUTER       TOKEN(OUTER)
+RIGHT       TOKEN(RIGHT)
+TABLE       TOKEN(TABLE)
+UNION       TOKEN(UNION)
+USING       TOKEN(USING)
+WHERE       TOKEN(WHERE)
+CALL        TOKEN(CALL)
+CASE        TOKEN(CASE)
+CHAR        TOKEN(CHAR)
+COPY        TOKEN(COPY)
+DATE        TOKEN(DATE)
+DATETIME    TOKEN(DATETIME)
+DESC        TOKEN(DESC)
+DROP        TOKEN(DROP)
+ELSE        TOKEN(ELSE)
+FILE        TOKEN(FILE)
+FROM        TOKEN(FROM)
+FULL        TOKEN(FULL)
+HASH        TOKEN(HASH)
+HINT        TOKEN(HINT)
+INTO        TOKEN(INTO)
+JOIN        TOKEN(JOIN)
+LEFT        TOKEN(LEFT)
+LIKE        TOKEN(LIKE)
+ILIKE       TOKEN(ILIKE)
+LOAD        TOKEN(LOAD)
+LONG        TOKEN(LONG)
+NULL        TOKEN(NULL)
+PLAN        TOKEN(PLAN)
+SHOW        TOKEN(SHOW)
+TEXT        TOKEN(TEXT)
+THEN        TOKEN(THEN)
+TIME        TOKEN(TIME)
+VIEW        TOKEN(VIEW)
+WHEN        TOKEN(WHEN)
+WITH        TOKEN(WITH)
+ADD         TOKEN(ADD)
+ALL         TOKEN(ALL)
+AND         TOKEN(AND)
+ASC         TOKEN(ASC)
+END         TOKEN(END)
+FOR         TOKEN(FOR)
+INT         TOKEN(INT)
+KEY         TOKEN(KEY)
+NOT         TOKEN(NOT)
+OFF         TOKEN(OFF)
+SET         TOKEN(SET)
+TOP         TOKEN(TOP)
+AS          TOKEN(AS)
+BY          TOKEN(BY)
+IF          TOKEN(IF)
+IN          TOKEN(IN)
+IS          TOKEN(IS)
+OF          TOKEN(OF)
+ON          TOKEN(ON)
+OR          TOKEN(OR)
+TO          TOKEN(TO)
+SECOND      TOKEN(SECOND)
+MINUTE      TOKEN(MINUTE)
+HOUR        TOKEN(HOUR)
+DAY         TOKEN(DAY)
+MONTH       TOKEN(MONTH)
+YEAR        TOKEN(YEAR)
+SECONDS     TOKEN(SECONDS)
+MINUTES     TOKEN(MINUTES)
+HOURS       TOKEN(HOURS)
+DAYS        TOKEN(DAYS)
+MONTHS      TOKEN(MONTHS)
+YEARS       TOKEN(YEARS)
+TRUE        TOKEN(TRUE)
+FALSE       TOKEN(FALSE)
 TRANSACTION TOKEN(TRANSACTION)
-BEGIN		TOKEN(BEGIN)
-ROLLBACK	TOKEN(ROLLBACK)
-COMMIT		TOKEN(COMMIT)
-INTERVAL	TOKEN(INTERVAL)
+BEGIN       TOKEN(BEGIN)
+ROLLBACK    TOKEN(ROLLBACK)
+COMMIT      TOKEN(COMMIT)
+INTERVAL    TOKEN(INTERVAL)
 
-                /* Allow =/== see https://sqlite.org/lang_expr.html#collateop */
-"=="		TOKEN(EQUALS)
-"!=" 		TOKEN(NOTEQUALS)
-"<>" 		TOKEN(NOTEQUALS)
-"<="		TOKEN(LESSEQ)
-">="		TOKEN(GREATEREQ)
-"||"		TOKEN(CONCAT)
+            /* Allow =/== see https://sqlite.org/lang_expr.html#collateop */
+"=="        TOKEN(EQUALS)
+"!="        TOKEN(NOTEQUALS)
+"<>"        TOKEN(NOTEQUALS)
+"<="        TOKEN(LESSEQ)
+">="        TOKEN(GREATEREQ)
+"||"        TOKEN(CONCAT)
 
-[-+*/(){},.;<>=^%:?[\]|]	{ return yytext[0]; }
+[-+*/(){},.;<>=^%:?[\]|]    { return yytext[0]; }
 
 [0-9]+"."[0-9]* |
 "."[0-9]*	{

src/parser/bison_parser.y

@@ -0,0 +1,41 @@
+#ifndef SQLPARSER_ALTER_STATEMENT_H
+#define SQLPARSER_ALTER_STATEMENT_H
+
+#include "SQLStatement.h"
+
+// Note: Implementations of constructors and destructors can be found in statements.cpp.
+namespace hsql {
+
+  enum ActionType {
+    DropColumn,
+  };
+
+  struct AlterAction {
+    AlterAction(ActionType type);
+    ActionType type;
+    virtual ~AlterAction();
+  };
+
+  struct DropColumnAction : AlterAction {
+    DropColumnAction(char* column_name);
+    char* columnName;
+    bool ifExists;
+
+    ~DropColumnAction() override;
+  };
+
+  // Represents SQL Alter Table statements.
+  // Example "ALTER TABLE students DROP COLUMN name;"
+  struct AlterStatement : SQLStatement {
+
+    AlterStatement(char* name, AlterAction* action);
+    ~AlterStatement() override;
+
+    char* schema;
+    bool ifTableExists;
+    char* name;
+    AlterAction* action;
+  };
+} // namespace hsql
+
+#endif