toadb scanner and gram paser init

Makefile

+# Makefile toadb project
+# create by senllang 2021/1/1
+# mail : study@senllang.onaliyun.com
+# Copyright (C) 2023-2023, senllang
+#
+
+
+VERSION_H = 0
+VERSION_L = 01
+
+TARGET = toadb-${VERSION_H}-${VERSION_L}
+
+##########################################
+# toplevel source code directory
+# topdir=$(shell pwd)
+topdir = .
+SRC_DIR=$(topdir)
+
+# search all directory and subdirectory
+DIRS = $(shell find $(SRC_DIR) -maxdepth 5 -type d)
+
+
+# general source file list , and objects list
+SOURCE = $(foreach dir, $(DIRS), $(wildcard $(dir)/*.c))
+OBJS = ${patsubst %.c, %.o, $(SOURCE)}
+
+# .h files take place all directories.
+INC = ${foreach dir, $(DIRS), -I$(dir)}
+
+SCANNER = scanner.c
+PARSER = grammar.c
+PARSER_H = grammar.h
+
+##########################################
+# test programme
+builddir = $(topdir)/build
+
+##########################################
+##########################################
+DEFINES += -DTEST_PRO
+LDFLAGS += -lpthread
+DEBUG_FLAGS = -g
+##########################################		
+		
+CC = gcc
+#CFLAGS += -O2
+
+CFLAGS += ${INC}
+CFLAGS += ${DEFINES}
+CFLAGS += ${LDFLAGS}
+
+CFLAGS += ${DEBUG_FLAGS}
+
+# compile start
+.PHONY: all clean 
+all: $(default) $(builddir) $(OBJS) $(TARGET) 
+
+#build dir create
+$(builddir):
+	mkdir -p $(builddir)
+
+# object compile to O_PATH
+.c.o: $(SOURCE)
+	$(CC) $(CFLAGS) -c $<
+
+$(TARGET):$(topdir)/*.o
+	$(CC) $(CFLAGS) $^ -o $@
+
+# show various value
+default:
+	echo -e $(DIRS)
+	echo -e $(SOURCE)
+	echo -e $(OBJS)
+	echo -e $(INC)
+
+parser: $(PARSER) $(SCANNER)
+$(PARSER): $(topdir)/grammar.y
+	bison --defines=${PARSER_H} -o $@ $<
+
+$(SCANNER): $(topdir)/scanner.l
+	flex -o $@ $<
+	
+clean:
+	rm -rf ${TARGET} ${OBJS} ${builddir} $(topdir)/*.o ${SCANNER} ${PARSER} ${PARSER_H}
+	

README.md

-### Cluster management template project
+# 数据库SQL解析
+分为两部分，
+词法分析器scanner
+语法分析器grammar
 
-Example [cluster management](https://docs.gitlab.com/ee/user/clusters/management_project.html) project.
+## 目前支持的SQL情况：
+支持DDL:
+create table name (column type,..);
+drop table name;
 
-This project is based on a GitLab [Project Template](https://docs.gitlab.com/ee/gitlab-basics/create-project.html).
+支持DML:
+select */ column, ... from tablename;
+insert into tablename(column,...) values(columnvalue,...);
 
-Improvements can be proposed in the [original project](https://gitlab.com/gitlab-org/project-templates/cluster-management).
+## 开发指南
+词法分析器和语法分析器使用了著名的开源工具flex和bison组合，flex能够按照正则表达式规则解析出关键字，标识符，以及其它字符串；而bison则通过它的语法推导规则，将各个部分解析成一个个表达式，每个表达式形成逆波兰式的抽象语法树。
+
+### 词法分析器编写过程的注意事项：
+词法分析器是向前看一个字符，所以对于规则在文件中的前后顺序就有要求。比如标识符字串如果放在操作符规则之后，那么在匹配时就会先匹配到操作符的分号，此时就会执行操作符的规则，而把前面的字符串会丢掉，把两者换个顺序就会优先匹配标识符，当下一个字符为分号时，先执行标识符的动作，再取分号进行分析。
+
+### 语法分析器编写过程注意事项：
+语法分析器在编写过程中遇到两个问题：
+（1）语法分析器规则里要使用当前表达式的值时，需要给表达式定义类型，默认是yylval为整型，如果是字符串时，就需通过%union定义，然后在token里绑定类型；
+（2）语法分析器第二部分规则定义部分，规则推导出的内容必须是词法分析器返回的内容，比如列名和类型中间是空格，但是空格在词法分析器中已经过滤掉了，就不能用空格来分隔了。
+
+
+
+# 数据库SQL执行
+
+
+# roadmap
+
+## 0.1版本开发
+
+功能：支持创建表，删除表，插入数据，删除数据
+数据库特性：客户端，词法/语法解析，SQL执行，数据字典，存储管理
\ No newline at end of file

grammar.y

+%{
+/*
+ *	toadb grammar 
+ * Copyright (C) 2023-2023, senllang
+*/
+
+#include <stdio.h>
+#include <stdlib.h>
+
+int grammar_main();
+%}
+
+%union {
+    char *sval;
+    char op;
+}
+
+/* tokens define */
+%token SELECT
+%token CREATE
+%token TABLE
+%token FROM
+%token DELETE
+%token DROP
+%token INSERT
+%token UPDATE
+%token <sval> IDENT
+
+
+%%
+stmt_list:  ;
+            |stmt ';'
+            | stmt_list stmt ';'
+        ;
+stmt:       select_stmt 
+                    {
+                        printf("select stmt\n");
+                    }
+            | create_stmt
+                    {
+                        printf("create stmt\n");
+                    }
+            | drop_stmt
+                    {
+                        printf("drop stmt\n");
+                    }
+
+        ;
+select_stmt:    SELECT select_opt_list
+                    {
+                        printf("select \n");
+                    }
+            |   SELECT select_opt_list FROM table_reference_list
+                    {
+                        printf("select  from \n");
+                    }
+        ;
+select_opt_list: attr_name
+            | select_opt_list ',' attr_name
+            | '*'
+        ;
+
+table_reference_list: table_reference
+                | table_reference ',' table_reference
+        ;
+
+table_reference:   IDENT
+                    {
+                        printf("list ident :%s\n", $1);
+                    }
+        ;
+
+create_stmt:        CREATE TABLE tablename '(' column_list ')'
+                        {
+                            printf("sql: create \n");
+                        }
+        ;
+drop_stmt:          DROP TABLE tablename
+                    {
+                        printf("sql drop \n");
+                    }
+        ;
+tablename:      IDENT
+                    {
+                        printf("ident :%s\n", $1);
+                    }
+        ;
+column_list:    attr_define
+                |    column_list ',' attr_define
+        ;
+attr_define:   attr_name attr_type
+                    {
+                        printf("attrdef \n");
+                    }
+        ;
+attr_name:      IDENT
+                    {
+                        printf("ident :%s\n", $1);
+                    }
+        ;
+attr_type:      IDENT
+                    {
+                        printf("ident :%s\n", $1);
+                    }
+        ;
+
+%%
+int grammar_main()
+{
+    yyparse();
+
+    return 0;
+}
+
+yyerror(char *s)
+{
+        fprintf(stderr, "error: %s\n",s);
+}

main.c

+/*
+ *	toadb main 
+ * Copyright (C) 2023-2023, senllang
+*/
+#include <stdio.h>
+#include <stdlib.h>
+#include "grammar.h"
+
+int grammar_main();
+int main(int argc, char *agrv[])
+{
+	grammar_main();
+	return 0;
+}

scanner.l

+%top{
+/*
+ *	toadb scanner 
+ * Copyright (C) 2023-2023, senllang
+*/
+}
+
+/* define list */
+%{
+#include <stdio.h>
+#include <stdlib.h>
+#include "grammar.h"
+int test = 0;
+%}
+
+%option outfile="scanner.c"
+%option noyywrap
+%option yylineno
+%option case-insensitive
+
+/* operators */
+operator    [-+&~|^/%*().!,;]
+space       [ \t\n\r\f]
+nonewline   [^\n\r]
+comment     ("--"{nonewline}*)
+whitespace		({space}+|{comment})
+identify    [a-zA-Z][a-zA-Z0-9_]*
+
+%%
+SELECT  { 
+            return SELECT; 
+        }
+FROM    {
+            return FROM; 
+        }
+CREATE  {
+            return CREATE; 
+        }
+TABLE	{
+		return TABLE;
+	}
+UPDATE  {
+            return UPDATE; 
+        }
+INSERT  {
+            return INSERT; 
+        }
+DELETE  {
+            return DELETE; 
+        }
+DROP    {
+            return DROP; 
+        }
+{identify}        {
+                    yylval.sval = strdup(yytext);
+                    return IDENT;
+                }
+{operator}       {
+                    return yytext[0];
+                }
+{whitespace}      {
+                    /* ignore */
+                }
+
+%%

test/scanner-test.l

+%top{
+/*
+ *	toadb scanner 
+ * Copyright (C) 2023-2023, senllang
+*/
+}
+
+/* define list */
+%{
+#include <stdio.h>
+#include <stdlib.h>
+int test = 0;
+%}
+
+%option outfile="scanner-test.c"
+%option noyywrap
+%option yylineno
+%option case-insensitive
+
+/* operators */
+endtag      [;]*
+seperate    [,]
+operator    [-+&~|^/%*().!]
+space       [ \t\n\r\f]
+nonewline   [^\n\r]
+comment     ("--"{nonewline}*)
+whitespace		({space}+|{comment})
+identify    [a-zA-Z][a-zA-Z0-9_]*
+
+
+%%
+select  { 
+            printf("--select \n"); 
+        }
+from    {
+            printf("--from \n"); 
+        }
+create  {
+            printf("--CREATE\n"); 
+        }
+update  {
+            printf("--update\n"); 
+        }
+insert  {
+           printf("--insert\n"); 
+        }
+delete  {
+            printf("--delete\n"); 
+        }
+drop    {
+           printf("--drop\n"); 
+        }
+{identify}        {
+                    printf("ident :%s\n", yytext); 
+                } 
+{operator}        {
+                    printf("op:%c\n",yytext[0]); 
+                }
+{endtag}          {
+                    printf("end:%c\n",yytext[0]); 
+                }
+{seperate}        {
+                    printf("se:%c\n",yytext[0]); 
+                }               
+{space}           {
+                    /* ignore */
+                }
+{whitespace}      {
+                    /* ignore */
+                }
+%%
+int main(int argc, char * argv[])
+{
+    /*            | drop_stmt
+                    {
+                        printf("drop stmt\n");
+                    }
+            | insert_stmt
+                    {
+                        printf("insert stmt\n");
+                    }
+            | delete_stmt
+                    {
+                        printf("delete stmt\n");
+                    }
+*/    
+    yylex();
+    return 0;
+}
\ No newline at end of file