From f7924ced4c8dd5319c47d865be8577bfdb988a58 Mon Sep 17 00:00:00 2001
From: Lu Wang <l423wang@uwaterloo.ca>
Date: Wed, 14 Nov 2018 18:21:45 -0500
Subject: [PATCH] Added grammar rules, building parse tree
---
Makefile | 17 +++++
SQLPGrammar.y | 202 ++++++++++++++++++++++++++++++++++++++++++++++++++
SQLPParser.c | 37 +++++++++
SQLPScanner.l | 164 ++++++++++++++++++++++++++++++++++++++++
util.c | 32 ++++++++
util.h | 38 ++++++++++
6 files changed, 490 insertions(+)
create mode 100644 Makefile
create mode 100644 SQLPGrammar.y
create mode 100644 SQLPParser.c
create mode 100644 SQLPScanner.l
create mode 100644 util.c
create mode 100644 util.h
diff --git a/Makefile b/Makefile
new file mode 100644
index 0000000..89f986e
--- /dev/null
+++ b/Makefile
@@ -0,0 +1,17 @@
+.PHONY: all clean
+
+all: SQLPParser
+
+
+SQLPParser: SQLPParser.c SQLPGrammar.y SQLPScanner.l util.o
+ bison --verbose -d SQLPGrammar.y
+ flex SQLPScanner.l
+ gcc -w SQLPParser.c util.o -o SQLPParser
+ rm -f lex.yy.c SQLPGrammar.tab.c SQLPGrammar.tab.h
+
+util.o: util.c util.h
+ gcc -c util.c
+
+clean:
+ rm -f lex.yy.c *.tab.c *.tab.h *.fasl
+
diff --git a/SQLPGrammar.y b/SQLPGrammar.y
new file mode 100644
index 0000000..4077a3c
--- /dev/null
+++ b/SQLPGrammar.y
@@ -0,0 +1,202 @@
+%{
+ #include "util.h"
+%}
+
+// %union {
+// struct node *n;
+//}
+
+%define api.value.type {struct node *}
+%token IMPLIES OR AND NOT LE GE LT GT NE HAS MAX MIN AS ASC DESC MOD ASSIGN EQ STAR COMMA DOT
+
+%token SIZE SELECTIVITY OVERLAP
+%token FREQUENCY UNIT TIME SPACE
+
+%token IDENTIFIER CONSTANT STRING_LITERAL SIZEOF
+
+%token STORE STORING DYNAMIC STATIC OF TYPE ORDERED BY
+%token INDEX LIST ARRAY BINARY TREE DISTRIBUTED POINTER
+
+%token SCHEMA CLASS ISA PROPERTIES CONSTRAINTS PROPERTY
+%token ON DETERMINED COVER QUERY GIVEN FROM SELECT WHERE ORDER
+%token PRECOMPUTED ONE EXIST FOR ALL TRANSACTION INTCLASS STRCLASS
+%token INTEGER REAL DOUBLEREAL STRING MAXLEN RANGE TO
+%token INSERT END CHANGE DELETE DECLARE RETURN UNION
+
+%start SQLPProgram
+%%
+SQLPProgram
+ : Query
+ { printf("Input Query\n");
+ struct node *n = new_node(1, SQLPProgram);
+ n->children[0] = $1;
+ printf("Printing Tree\n");
+ print_tree(n, 0);
+ }
+ ;
+
+Identifier
+ : IDENTIFIER
+ { printf("|%s| ", yytext);
+ $$ = new_node(1, Identifier);
+ $$->children[0] = $1;
+ }
+ ;
+
+Query
+ : Union_Query
+ { printf("Union Query\n");
+ $$ = new_node(1, Query);
+ $$->children[0] = $1;
+ }
+ ;
+
+Select_Query
+ : SELECT Select_List Body
+ { printf("SQLP Query\n");
+ $$ = new_node(3, Select_Query);
+ $$->children[0] = new_node(0, SELECT);
+ $$->children[1] = $2;
+ $$->children[2] = $3;
+ }
+ ;
+
+Body
+ : FROM TablePath
+ { printf("Body 1\n");
+ $$ = new_node(2, Body);
+ $$->children[0] = new_node(0, FROM);
+ $$->children[1] = $2;}
+ | FROM TablePath WHERE Bool
+ { printf("Body 2\n");
+ $$ = new_node(4, Body);
+ $$->children[0] = new_node(0, FROM);
+ $$->children[1] = $2;
+ $$->children[2] = new_node(0, WHERE);
+ $$->children[3] = $4;
+ }
+ | FROM TablePath WHERE Pred
+ { printf("Body 3\n");
+ $$ = new_node(4, Body);
+ $$->children[0] = new_node(0, FROM);
+ $$->children[1] = $2;
+ $$->children[2] = new_node(0, WHERE);
+ $$->children[3] = $4;
+ }
+ ;
+
+TablePath
+ : Identifier Identifier
+ { printf("table path1\n");
+ $$ = new_node(2, TablePath);
+ $$->children[0] = $1;
+ $$->children[1] = $2;
+ }
+ | Identifier Identifier ',' TablePath
+ { printf("table path2\n");
+ $$ = new_node(4, TablePath);
+ $$->children[0] = $1;
+ $$->children[1] = $2;
+ $$->children[2] = new_node(0, COMMA);
+ $$->children[3] = $4;
+ }
+ ;
+
+Union_Query
+ : Select_Query
+ { printf("union query 1\n");
+ $$ = new_node(1, Union_Query);
+ $$->children[0] = $1;
+ }
+ | Union_Query UNION Select_Query
+ { printf("union query 2\n");
+ $$ = new_node(3, Union_Query);
+ $$->children[0] = $1;
+ $$->children[1] = new_node(0, UNION);
+ $$->children[2] = $3;
+ }
+ ;
+
+Select_List
+ : STAR
+ { printf("star\n");
+ $$ = new_node(1, Select_List);
+ $$->children[0] = new_node(0, STAR);
+ }
+ | AttrPath
+ { printf("select list attr path\n");
+ $$ = new_node(1, Select_List);
+ $$->children[0] = $1;
+ }
+ | AttrPath ',' Select_List
+ { printf("Select list\n");
+ $$ = new_node(3, Select_List);
+ $$->children[0] = $1;
+ $$->children[1] = new_node(0, COMMA);
+ $$->children[2] = $3;
+ }
+ ;
+
+AttrPath
+ : Identifier
+ { printf("path id\n");
+ $$ = new_node(1, AttrPath);
+ $$->children[0] = $1;
+ }
+ | Identifier '.' AttrPath
+ { printf("Path Function\n");
+ $$ = new_node(3, AttrPath);
+ $$->children[0] = $1;
+ $$->children[1] = new_node(0, DOT);
+ $$->children[2] = $3;
+ }
+ ;
+Operator
+ : EQ
+ | NE
+ | LE
+ | GE
+ | LT
+ | GT
+ ;
+
+Bool
+ : AttrPath Operator AttrPath
+ { printf("AttrPath op AttrPath\n");
+ $$ = new_node(3, Bool);
+ $$->children[0] = $1;
+ $$->children[1] = $2;
+ $$->children[2] = $3;
+ }
+ | AttrPath Operator CONSTANT
+ { printf("AttrPath op Constant\n");
+ $$ = new_node(3, Bool);
+ $$->children[0] = $1;
+ $$->children[1] = $2;
+ $$->children[2] = new_node(0, CONSTANT);
+ }
+ ;
+
+Pred
+ : Bool AND Pred
+ { printf("pred and pred\n");
+ $$ = new_node(3, Pred);
+ $$->children[0] = $1;
+ $$->children[1] = new_node(0, AND);
+ $$->children[2] = $3;
+
+ }
+ | NOT Pred
+ { printf("Not Pred\n");
+ $$ = new_node(2, Pred);
+ $$->children[0] = new_node(0, NOT);
+ $$->children[1] = $2;
+
+ }
+ | EXIST '(' Query ')'
+ { printf("Exist query\n");
+ $$ = new_node(2, Pred);
+ $$->children[0] = new_node(0, EXIST);
+ $$->children[1] = $3;
+ }
+ ;
\ No newline at end of file
diff --git a/SQLPParser.c b/SQLPParser.c
new file mode 100644
index 0000000..5421d36
--- /dev/null
+++ b/SQLPParser.c
@@ -0,0 +1,37 @@
+/*
+ * Copyright (C) 1989, G. E. Weddell.
+ *
+ * This file is part of RDM.
+ *
+ * RDM is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * RDM is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with RDM. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <stdio.h>
+#include <util.h>
+#include "lex.yy.c"
+#include "SQLPGrammar.tab.c"
+
+int main()
+{
+ int Result;
+ strcpy(LineBuffer, "");
+ printf("(\n");
+ Result = yyparse();
+ if (Result)
+ printf("**error**");
+ else
+ fprintf(stderr, "checking semantics.\n");
+ printf(")\n");
+ return (Result);
+}
\ No newline at end of file
diff --git a/SQLPScanner.l b/SQLPScanner.l
new file mode 100644
index 0000000..e358f7e
--- /dev/null
+++ b/SQLPScanner.l
@@ -0,0 +1,164 @@
+%{
+/*
+ * Copyright (C) 1989, G. E. Weddell.
+ *
+ * This file is part of RDM.
+ *
+ * RDM is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * RDM is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with RDM. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "SQLPGrammar.tab.h"
+#include <stdio.h>
+#include <ctype.h>
+
+int LineNumber = 1;
+char LineBuffer[200];
+%}
+
+%p 3000
+
+D [0-9]
+L [a-zA-Z_]
+E [Ee][+-]?{D}+
+
+%%
+
+\%.*\n { LineNumber++; }
+\n { LineNumber++; strcpy(LineBuffer, ""); }
+"all" { strcat(LineBuffer, yytext); return(ALL); }
+"and" { strcat(LineBuffer, yytext); return(AND); }
+"array" { strcat(LineBuffer, yytext); return(ARRAY); }
+"as" { strcat(LineBuffer, yytext); return(AS); }
+"asc" { strcat(LineBuffer, yytext); return(ASC); }
+"binary" { strcat(LineBuffer, yytext); return(BINARY); }
+"by" { strcat(LineBuffer, yytext); return(BY); }
+"change" { strcat(LineBuffer, yytext); return(CHANGE); }
+"class" { strcat(LineBuffer, yytext); return(CLASS); }
+"constraints" { strcat(LineBuffer, yytext); return(CONSTRAINTS); }
+"cover" { strcat(LineBuffer, yytext); return(COVER); }
+"declare" { strcat(LineBuffer, yytext); return(DECLARE); }
+"delete" { strcat(LineBuffer, yytext); return(DELETE); }
+"desc" { strcat(LineBuffer, yytext); return(DESC); }
+"union" { strcat(LineBuffer, yytext); return(UNION); }
+"determined" { strcat(LineBuffer, yytext); return(DETERMINED); }
+"distributed" { strcat(LineBuffer, yytext); return(DISTRIBUTED); }
+"dynamic" { strcat(LineBuffer, yytext); return(DYNAMIC); }
+"exist" { strcat(LineBuffer, yytext); return(EXIST); }
+"for" { strcat(LineBuffer, yytext); return(FOR); }
+"frequency" { strcat(LineBuffer, yytext); return(FREQUENCY); }
+"from" { strcat(LineBuffer, yytext); return(FROM); }
+"given" { strcat(LineBuffer, yytext); return(GIVEN); }
+"has" { strcat(LineBuffer, yytext); return(HAS); }
+"implies" { strcat(LineBuffer, yytext); return(IMPLIES); }
+"index" { strcat(LineBuffer, yytext); return(INDEX); }
+"insert" { strcat(LineBuffer, yytext); return(INSERT); }
+"Integer" { strcat(LineBuffer, yytext); return(INTCLASS); }
+"isa" { strcat(LineBuffer, yytext); return(ISA); }
+"list" { strcat(LineBuffer, yytext); return(LIST); }
+"max" { strcat(LineBuffer, yytext); return(MAX); }
+"maxlen" { strcat(LineBuffer, yytext); return(MAXLEN); }
+"min" { strcat(LineBuffer, yytext); return(MIN); }
+"mod" { strcat(LineBuffer, yytext); return(MOD); }
+"not" { strcat(LineBuffer, yytext); return(NOT); }
+"of" { strcat(LineBuffer, yytext); return(OF); }
+"on" { strcat(LineBuffer, yytext); return(ON); }
+"one" { strcat(LineBuffer, yytext); return(ONE); }
+"or" { strcat(LineBuffer, yytext); return(OR); }
+"order" { strcat(LineBuffer, yytext); return(ORDER); }
+"ordered" { strcat(LineBuffer, yytext); return(ORDERED); }
+"overlap" { strcat(LineBuffer, yytext); return(OVERLAP); }
+"pointer" { strcat(LineBuffer, yytext); return(POINTER); }
+"precomputed" { strcat(LineBuffer, yytext); return(PRECOMPUTED); }
+"property" { strcat(LineBuffer, yytext); return(PROPERTY); }
+"properties" { strcat(LineBuffer, yytext); return(PROPERTIES); }
+"query" { strcat(LineBuffer, yytext); return(QUERY); }
+"range" { strcat(LineBuffer, yytext); return(RANGE); }
+"return" { strcat(LineBuffer, yytext); return(RETURN); }
+"schema" { strcat(LineBuffer, yytext); return(SCHEMA); }
+"select" { strcat(LineBuffer, yytext); return(SELECT); }
+"selectivity" { strcat(LineBuffer, yytext); return(SELECTIVITY); }
+"size" { strcat(LineBuffer, yytext); return(SIZE); }
+"space" { strcat(LineBuffer, yytext); return(SPACE); }
+"static" { strcat(LineBuffer, yytext); return(STATIC); }
+"store" { strcat(LineBuffer, yytext); return(STORE); }
+"storing" { strcat(LineBuffer, yytext); return(STORING); }
+"String" { strcat(LineBuffer, yytext); return(STRCLASS); }
+"time" { strcat(LineBuffer, yytext); return(TIME); }
+"to" { strcat(LineBuffer, yytext); return(TO); }
+"transaction" { strcat(LineBuffer, yytext); return(TRANSACTION); }
+"tree" { strcat(LineBuffer, yytext); return(TREE); }
+"type" { strcat(LineBuffer, yytext); return(TYPE); }
+"unit" { strcat(LineBuffer, yytext); return(UNIT); }
+"where" { strcat(LineBuffer, yytext); return(WHERE); }
+
+{L}({L}|{D})* { strcat(LineBuffer, yytext); return(IDENTIFIER); }
+
+{D}+ { strcat(LineBuffer, yytext); return(INTEGER); }
+{D}+{E} { strcat(LineBuffer, yytext); return(REAL); }
+{D}*"."{D}+({E})? { strcat(LineBuffer, yytext); return(REAL); }
+{D}+"."{D}*({E})? { strcat(LineBuffer, yytext); return(REAL); }
+
+\"(\\.|[^\\"])*\" { strcat(LineBuffer, yytext); return(STRING); }
+
+":=" { strcat(LineBuffer, yytext); return(ASSIGN); }
+"<=" { strcat(LineBuffer, yytext); return(LE); }
+">=" { strcat(LineBuffer, yytext); return(GE); }
+"<>" { strcat(LineBuffer, yytext); return(NE); }
+";" { strcat(LineBuffer, yytext); return(';'); }
+"{" { strcat(LineBuffer, yytext); return('{'); }
+"}" { strcat(LineBuffer, yytext); return('}'); }
+"," { strcat(LineBuffer, yytext); return(','); }
+"=" { strcat(LineBuffer, yytext); return(EQ); }
+"(" { strcat(LineBuffer, yytext); return('('); }
+")" { strcat(LineBuffer, yytext); return(')'); }
+"[" { strcat(LineBuffer, yytext); return('['); }
+"]" { strcat(LineBuffer, yytext); return(']'); }
+"." { strcat(LineBuffer, yytext); return('.'); }
+"-" { strcat(LineBuffer, yytext); return('-'); }
+"+" { strcat(LineBuffer, yytext); return('+'); }
+"*" { strcat(LineBuffer, yytext); return(STAR); }
+"/" { strcat(LineBuffer, yytext); return('/'); }
+"<" { strcat(LineBuffer, yytext); return('<'); }
+">" { strcat(LineBuffer, yytext); return('>'); }
+
+" " { strcat(LineBuffer, yytext); }
+"\r" { strcat(LineBuffer, yytext); }
+. { strcat(LineBuffer, yytext);
+ yyerror("illegal character"); }
+%%
+
+yyerror(s)
+char *s;
+{
+ int i;
+ char c;
+ fprintf( stderr, "\n%s", LineBuffer );
+ c = input();
+ while ( c != '\n' && c != 0 ) {
+ putc(c, stderr);
+ c = input();
+ }
+ putc('\n', stderr);
+ for (i=1; i <= strlen( LineBuffer ) - strlen( yytext ); i++)
+ putc(' ', stderr);
+ for (i=1; i <= strlen( yytext ); i++)
+ putc('^', stderr);
+ putc('\n', stderr);
+ fprintf ( stderr, "** error: %s found in line %d.\n\n", s, LineNumber );
+}
+
+yywrap()
+{
+ return(1);
+}
\ No newline at end of file
diff --git a/util.c b/util.c
new file mode 100644
index 0000000..e22cc07
--- /dev/null
+++ b/util.c
@@ -0,0 +1,32 @@
+#include "util.h"
+
+struct node *new_node(int count, int type) {
+ struct node *n = malloc(sizeof(struct node));
+ n->type = type;
+ n->count = count;
+ if (count > 0) n->children = malloc(sizeof(struct node*)*count);
+ return n;
+}
+
+
+const char *node_types[] = {"program", "id", "query", "select_query", "body", "table path", "union_query",
+"select list", "attrpath", "operator", "bool", "pred"};
+
+void print_tree(struct node *root, int indent) {
+ if (root == NULL) return;
+ if (root->count == 0) return;
+ for (int i = 0; i < indent; i++) {
+ printf(" ");
+ }
+ printf("type: %s\n", node_types[root->type]);
+ for (int i = 0; i < root->count; i++) {
+ print_tree(root->children[i], indent+1);
+ }
+}
+
+void destroy_tree(struct node *root) {
+ for (int i = 0; i < root->count; i++) {
+ destroy_tree(root->children[i]);
+ }
+ free(root);
+}
\ No newline at end of file
diff --git a/util.h b/util.h
new file mode 100644
index 0000000..621ccb0
--- /dev/null
+++ b/util.h
@@ -0,0 +1,38 @@
+#ifndef UTIL_H
+#define UTIL_H
+
+#include <stdio.h>
+#include <stdlib.h>
+
+struct node {
+ int type;
+ int count;
+ struct node **children;
+};
+
+
+
+enum type {
+ SQLPProgram,
+ Identifier,
+ Query,
+ Select_Query,
+ Body,
+ TablePath,
+ Union_Query,
+ Select_List,
+ AttrPath,
+ Operator,
+ Bool,
+ Pred
+};
+
+struct node *new_node(int count, int type);
+
+void print_tree(struct node *root, int indent);
+void destroy_tree(struct node *root);
+
+#endif
+
+
+
--
GitLab