initial commit

This commit is contained in:
2021-06-29 19:28:14 +02:00
commit 5c7908ac4b
20 changed files with 913 additions and 0 deletions

2
.gitignore vendored Normal file
View File

@@ -0,0 +1,2 @@
build
cmake-build-debug

3
.idea/.gitignore generated vendored Normal file
View File

@@ -0,0 +1,3 @@
# Default ignored files
/shelf/
/workspace.xml

4
.idea/misc.xml generated Normal file
View File

@@ -0,0 +1,4 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="CMakeWorkspace" PROJECT_DIR="$PROJECT_DIR$" />
</project>

8
.idea/modules.xml generated Normal file
View File

@@ -0,0 +1,8 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="ProjectModuleManager">
<modules>
<module fileurl="file://$PROJECT_DIR$/.idea/msql.iml" filepath="$PROJECT_DIR$/.idea/msql.iml" />
</modules>
</component>
</project>

2
.idea/msql.iml generated Normal file
View File

@@ -0,0 +1,2 @@
<?xml version="1.0" encoding="UTF-8"?>
<module classpath="CMake" type="CPP_MODULE" version="4" />

6
.idea/vcs.xml generated Normal file
View File

@@ -0,0 +1,6 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="VcsDirectoryMappings">
<mapping directory="$PROJECT_DIR$" vcs="Git" />
</component>
</project>

30
.vscode/c_cpp_properties.json vendored Normal file
View File

@@ -0,0 +1,30 @@
{
"configurations": [
{
"name": "Mac",
"browse": {
"path": [
"${workspaceFolder}"
],
"limitSymbolsToIncludedHeaders": true
},
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/stdlib",
"/usr/local/opt/openssl/include",
"${workspaceFolder}"
],
"defines": [],
"macFrameworkPath": [
"/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.14.sdk/System/Library/Frameworks"
],
"compilerPath": "/usr/bin/clang",
"cStandard": "c11",
"cppStandard": "c++17",
"intelliSenseMode": "clang-x64",
"configurationProvider": "vector-of-bool.cmake-tools",
"compileCommands": "${workspaceFolder}/build/compile_commands.json"
}
],
"version": 4
}

20
.vscode/launch.json vendored Normal file
View File

@@ -0,0 +1,20 @@
{
// Use IntelliSense to learn about possible attributes.
// Hover to view descriptions of existing attributes.
// For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
"version": "0.2.0",
"configurations": [
{
"name": "(lldb) Launch",
"type": "cppdbg",
"request": "launch",
"program": "${workspaceFolder}/build/msql",
"args": [],
"stopAtEntry": false,
"cwd": "${fileDirname}",
"environment": [],
"externalConsole": false,
"MIMode": "lldb"
}
]
}

78
.vscode/settings.json vendored Normal file
View File

@@ -0,0 +1,78 @@
{
"files.associations": {
"regex": "cpp",
"__config": "cpp",
"__nullptr": "cpp",
"exception": "cpp",
"initializer_list": "cpp",
"new": "cpp",
"stdexcept": "cpp",
"typeinfo": "cpp",
"__functional_03": "cpp",
"__string": "cpp",
"deque": "cpp",
"functional": "cpp",
"string": "cpp",
"system_error": "cpp",
"vector": "cpp",
"__bit_reference": "cpp",
"__functional_base": "cpp",
"algorithm": "cpp",
"atomic": "cpp",
"bitset": "cpp",
"chrono": "cpp",
"iterator": "cpp",
"limits": "cpp",
"locale": "cpp",
"memory": "cpp",
"ratio": "cpp",
"tuple": "cpp",
"type_traits": "cpp",
"unordered_map": "cpp",
"__functional_base_03": "cpp",
"__hash_table": "cpp",
"__tuple": "cpp",
"utility": "cpp",
"__split_buffer": "cpp",
"ios": "cpp",
"string_view": "cpp",
"map": "cpp",
"__tree": "cpp",
"istream": "cpp",
"__locale": "cpp",
"iosfwd": "cpp",
"array": "cpp",
"hash_map": "cpp",
"list": "cpp",
"optional": "cpp",
"sstream": "cpp",
"cstddef": "cpp",
"hashtable": "cpp",
"variant": "cpp",
"__debug": "cpp",
"__errc": "cpp",
"__mutex_base": "cpp",
"__node_handle": "cpp",
"__threading_support": "cpp",
"cctype": "cpp",
"cmath": "cpp",
"cstdarg": "cpp",
"cstdint": "cpp",
"cstdio": "cpp",
"cstdlib": "cpp",
"cstring": "cpp",
"ctime": "cpp",
"cwchar": "cpp",
"cwctype": "cpp",
"fstream": "cpp",
"iomanip": "cpp",
"iostream": "cpp",
"mutex": "cpp",
"ostream": "cpp",
"streambuf": "cpp",
"thread": "cpp"
},
"C_Cpp.intelliSenseEngineFallback": "Disabled",
"cmake.configureOnOpen": true,
"C_Cpp.configurationWarnings": "Disabled"
}

26
.vscode/tasks.json vendored Normal file
View File

@@ -0,0 +1,26 @@
{
// See https://go.microsoft.com/fwlink/?LinkId=733558
// for the documentation about the tasks.json format
"version": "2.0.0",
"tasks": [
{
"label": "compile",
"type": "shell",
"command": "clang++ -o mi -g -mmacosx-version-min=10.14 -lstdc++ -std=c++1z -lssl -lcrypto -lpqxx -lpq -L/usr/local/opt/openssl/lib -I/usr/local/opt/openssl/include -L/usr/local/opt/libpqxx/lib -I/usr/local/opt/libpqxx/include -L/usr/local/opt/postgres/lib -I/usr/local/opt/postgres/include -I./ -I./stdlib *.cpp ./stdlib/*.cpp",
"group": {
"kind": "build",
"isDefault": true
}
},
{
"label": "compile O3",
"type": "shell",
"command": "clang++ -o mi -O3 -mmacosx-version-min=10.14 -lstdc++ -std=c++1z -lssl -lcrypto -lpqxx -lpq -L/usr/local/opt/openssl/lib -I/usr/local/opt/openssl/include -L/usr/local/opt/libpqxx/lib -I/usr/local/opt/libpqxx/include -L/usr/local/opt/postgres/lib -I/usr/local/opt/postgres/include -I./ -I./stdlib *.cpp ./stdlib/*.cpp",
"group": {
"kind": "build",
"isDefault": false
},
"problemMatcher": []
}
]
}

21
CMakeLists.txt Normal file
View File

@@ -0,0 +1,21 @@
cmake_minimum_required(VERSION 3.0)
set(CMAKE_CXX_STANDARD 17)
set(CMAKE_CXX_STANDARD_REQUIRED ON)
set(CMAKE_CXX_EXTENSIONS OFF)
set(CMAKE_OSX_DEPLOYMENT_TARGET "10.14")
project(msql)
set(PROJECT_NAME msql)
set(SOURCE
exception.cpp lexer.cpp parser.cpp executor.cpp main.cpp)
add_executable(${PROJECT_NAME} ${SOURCE})
target_link_libraries(${PROJECT_NAME} stdc++ m)
target_compile_options(msql PRIVATE -g)

9
exception.cpp Normal file
View File

@@ -0,0 +1,9 @@
#include "exception.h"
Exception::Exception(const std::string &msg) {
cause = msg;
}
const char* Exception::what() const noexcept { return cause.c_str(); }

15
exception.h Normal file
View File

@@ -0,0 +1,15 @@
#pragma once
#include "lexer.h"
#include <string>
class Exception : public std::exception {
private:
std::string cause;
public:
Exception(const std::string &msg);
const char* what() const noexcept;
};

29
executor.cpp Normal file
View File

@@ -0,0 +1,29 @@
#include "executor.h"
#include "exception.h"
Executor::Executor() {
// TODO init database
}
bool Executor::execute(Node& node) {
switch (node.node_type) {
case NodeType::create_table:
return execute_create_table(static_cast<CreateTableNode &>(node));
case NodeType::select_from:
return execute_select(node);
default:
// TODO error message
return false;
}
}
bool Executor::execute_create_table(CreateTableNode& node) {
return false;
}
bool Executor::execute_select(Node& node) {
return false;
}

21
executor.h Normal file
View File

@@ -0,0 +1,21 @@
#pragma once
#include "parser.h"
#include <string>
class Executor {
private:
public:
Executor();
bool execute(Node& node);
private:
bool execute_create_table(CreateTableNode& node);
bool execute_select(Node& node);
private:
};

376
lexer.cpp Normal file
View File

@@ -0,0 +1,376 @@
#include "lexer.h"
#include "exception.h"
#include <algorithm>
Token::Token(const std::string &token_str, TokenType typ) {
token_string = token_str;
type = typ;
}
void Lexer::parse(const std::string &code) {
// TODO handle empty code
tokens.clear();
// PERF something like this to prealocate ??
if (code.size() > 100) {
tokens.reserve(code.size() / 10);
}
code_str = code;
if (!code_str.empty() && code_str.back() != '\n') {
code_str.append("\n"); // TODO tempo solution to prevent possible situation when last line is a comment
}
// TODO make it constant
std::regex words_regex("[0-9]+\\.[0-9]+|[0-9][0-9_]+[0-9]|[0-9]+|[A-Za-z]+[A-Za-z0-9_#]*|[\\(\\)\\[\\]\\{\\}]|[-\\+\\*/"
",;:\?]|==|>=|<=|~=|>|<|=|;|~|\\|\\||&&|\n|\r|\r\n|'([^']|'')*'|\".*?\"|%.*?\n");
auto words_begin = std::sregex_iterator(code_str.begin(), code_str.end(), words_regex);
auto words_end = std::sregex_iterator();
for (std::sregex_iterator i = words_begin; i != words_end; ++i) {
std::smatch match = *i;
std::string match_str = match.str();
TokenType token_type = type(match_str);
if (token_type == TokenType::string_literal) {
match_str = stringLiteral(match_str);
} else {
tokens.push_back(Token{match_str, token_type});
}
}
// DEBUG IT
// debugTokens();
index = 0;
}
void Lexer::debugTokens() {
int i = 0;
for (std::vector<Token>::iterator it = tokens.begin(); it != tokens.end(); ++it) {
std::cerr << i << "\t" << it->token_string << std::endl;
i++;
}
}
Token Lexer::currentToken() { return tokens[index]; }
void Lexer::nextToken() {
if (index < tokens.size()) {
index++;
}
}
void Lexer::skipToken(TokenType type) {
if (tokenType() == type) {
nextToken();
} else {
throw Exception("ERROR unexpected token " + currentToken().token_string + ", instead of " + typeToString(type));
}
}
void Lexer::skipTokenOptional(TokenType type) {
if (tokenType() == type) {
nextToken();
}
}
TokenType Lexer::tokenType() { return index < tokens.size() ? currentToken().type : TokenType::eof; }
TokenType Lexer::nextTokenType() { return index < tokens.size() - 1 ? tokens[index + 1].type : TokenType::eof; }
TokenType Lexer::prevTokenType() { return index > 0 ? tokens[index - 1].type : TokenType::undef; }
bool Lexer::isRelationalOperator(TokenType token_type) {
return (token_type == TokenType::equal || token_type == TokenType::not_equal || token_type == TokenType::greater || token_type == TokenType::greater_equal ||
token_type == TokenType::lesser || token_type == TokenType::lesser_equal);
}
TokenType Lexer::type(const std::string &token) {
// TODO move it to class level not to reinit it again and again
std::regex int_regex("[0-9]+");
std::regex int_underscored_regex("[0-9][0-9_]+[0-9]");
std::regex double_regex("[0-9]+\\.[0-9]+");
std::regex identifier_regex("[A-Za-z]+[A-Za-z0-9_#]*");
if (token == ";")
return TokenType::semicolon;
if (token == "+")
return TokenType::plus;
if (token == "-")
return TokenType::minus;
if (token == "*")
return TokenType::multiply;
if (token == "/")
return TokenType::divide;
if (token == "(")
return TokenType::open_paren;
if (token == ")")
return TokenType::close_paren;
if (token == "=")
return TokenType::equal;
if (token == "!=")
return TokenType::not_equal;
if (token == ">")
return TokenType::greater;
if (token == ">=")
return TokenType::greater_equal;
if (token == "<")
return TokenType::lesser;
if (token == "<=")
return TokenType::lesser_equal;
if (token == "create")
return TokenType::keyword_create;
if (token == "where")
return TokenType::keyword_where;
if (token == "from")
return TokenType::keyword_from;
if (token == "table")
return TokenType::keyword_table;
if (token == "insert")
return TokenType::keyword_insert;
if (token == "into")
return TokenType::keyword_into;
if (token == "values")
return TokenType::keyword_values;
if (token == "select")
return TokenType::keyword_select;
if (token == "set")
return TokenType::keyword_set;
if (token == "copy")
return TokenType::keyword_copy;
if (token == "not")
return TokenType::keyword_not;
if (token == "null")
return TokenType::keyword_null;
if (token == "integer")
return TokenType::keyword_int;
if (token == "float")
return TokenType::keyword_float;
if (token == "varchar")
return TokenType::keyword_varchar;
if (token == "||")
return TokenType::logical_or;
if (token == "&&")
return TokenType::logical_and;
if (token == ",")
return TokenType::comma;
if (token == "\n" || token == "\r\n" || token == "\r")
return TokenType::newline;
if (token.length() > 1 && token.at(0) == '%' && (token.at(token.length() - 1) == '\n' || token.at(token.length() - 1) == '\r'))
return TokenType::comment;
// if (token.length() >= 2 && token.at(0) == '"' && token.at(token.length() - 1) == '"')
// return TokenType::string_literal;
if (token.length() >= 2 && token.at(0) == '\'' && token.at(token.length() - 1) == '\'')
return TokenType::string_literal;
if (std::regex_match(token, int_regex))
return TokenType::int_number;
if (std::regex_match(token, int_underscored_regex))
return TokenType::int_number;
if (std::regex_match(token, double_regex))
return TokenType::double_number;
if (std::regex_match(token, identifier_regex))
return TokenType::identifier;
if (index + 1 >= tokens.size())
return TokenType::eof;
return TokenType::undef;
}
std::string Lexer::stringLiteral(std::string token) {
// remove ' or " from the literal ends
bool replace = token[0]=='\'' && token[token.size()-1]=='\'';
std::string str = token.substr(1, token.size() - 2);
if (!replace) {
return str;
}
std::string out = "";
out.reserve(str.size());
for(std::string::size_type i = 0; i < str.size(); ++i) {
if (str[i] == '\'' && i < str.size() - 1) {
if (str[i+1] == '\'') {
out.append(1, '\'');
i++;
} else {
out.append(1, str[i]);
}
} else if (str[i] == '\\' && i < str.size() - 1) {
if (str[i+1] == 'n') {
out.append(1, '\n');
i++;
} else if (str[i+1] == 't') {
out.append(1, '\t');
i++;
} else {
out.append(1, str[i]);
}
} else {
out.append(1, str[i]);
}
}
return out;
}
std::string Lexer::typeToString(TokenType token_type) {
std::string txt;
switch (token_type) {
case TokenType::undef:
txt = "undef";
break;
case TokenType::identifier:
txt = "identifier";
break;
case TokenType::plus:
txt = "+";
break;
case TokenType::minus:
txt = "-";
break;
case TokenType::multiply:
txt = "*";
break;
case TokenType::divide:
txt = "/";
break;
case TokenType::equal:
txt = "==";
break;
case TokenType::not_equal:
txt = "!=";
break;
case TokenType::greater:
txt = ">";
break;
case TokenType::greater_equal:
txt = ">=";
break;
case TokenType::lesser:
txt = "<";
break;
case TokenType::lesser_equal:
txt = "<=";
break;
case TokenType::keyword_create:
txt = "create";
break;
case TokenType::keyword_where:
txt = "where";
break;
case TokenType::keyword_table:
txt = "table";
break;
case TokenType::keyword_into:
txt = "into";
break;
case TokenType::keyword_values:
txt = "values";
break;
case TokenType::keyword_select:
txt = "select";
break;
case TokenType::keyword_set:
txt = "set";
break;
case TokenType::keyword_copy:
txt = "copy";
break;
case TokenType::keyword_not:
txt = "not";
break;
case TokenType::keyword_null:
txt = "null";
break;
case TokenType::keyword_int:
txt = "integer";
break;
case TokenType::keyword_float:
txt = "float";
break;
case TokenType::keyword_varchar:
txt = "varchar";
break;
case TokenType::int_number:
txt = "int number";
break;
case TokenType::double_number:
txt = "double number";
break;
case TokenType::string_literal:
txt = "string literal";
break;
case TokenType::open_paren:
txt = "(";
break;
case TokenType::close_paren:
txt = ")";
break;
case TokenType::logical_and:
txt = "and";
break;
case TokenType::logical_or:
txt = "or";
break;
case TokenType::semicolon:
txt = ";";
break;
case TokenType::comma:
txt = ",";
break;
case TokenType::newline:
txt = "newline";
break;
case TokenType::comment:
txt = "comment";
break;
case TokenType::eof:
txt = "eof";
break;
default:
txt = "FIXME, unknown token type";
break;
}
return txt;
}

87
lexer.h Normal file
View File

@@ -0,0 +1,87 @@
#pragma once
#include <iostream>
#include <regex>
#include <stdexcept>
#include <string>
enum class TokenType {
undef,
identifier,
plus,
minus,
multiply,
divide,
equal,
not_equal,
greater,
greater_equal,
lesser,
lesser_equal,
keyword_create,
keyword_table,
keyword_where,
keyword_from,
keyword_insert,
keyword_into,
keyword_values,
keyword_select,
keyword_set,
keyword_copy,
keyword_not,
keyword_null,
keyword_int,
keyword_float,
keyword_varchar,
int_number,
double_number,
string_literal,
open_paren,
close_paren,
logical_and,
logical_or,
semicolon,
comma,
newline,
comment,
eof
};
struct Token {
std::string token_string;
TokenType type;
Token(const std::string &token_str, TokenType typ);
};
class Lexer {
private:
std::string code_str;
std::vector<Token> tokens;
int index = 0;
bool eof = false;
public:
Lexer() {};
void parse(const std::string &code);
void debugTokens();
Token currentToken();
void nextToken();
void skipToken(TokenType type);
void skipTokenOptional(TokenType type);
TokenType tokenType();
TokenType nextTokenType();
TokenType prevTokenType();
static bool isRelationalOperator(TokenType token_type);
private:
TokenType type(const std::string &token);
std::string stringLiteral(std::string token);
static std::string typeToString(TokenType token_type);
};

26
main.cpp Normal file
View File

@@ -0,0 +1,26 @@
#include "parser.h"
#include "executor.h"
// https://dev.to/joaoh82/what-would-sqlite-look-like-if-written-in-rust-part-1-2np4
// parser should get lexer as param and table executor to be able translate * or get types or so
// podporovat create as select
// drop table
int main(int argc, char *argv[]) {
Parser parser{};
Executor executor{};
std::string sql_create = "create table a (i integer not null, s varchar(64), f float)";
// std::string sql_insert = "insert into a (i, s) values(1, 'one')";
// std::string sql_inser2 = "insert into a (i, s) values(2, 'two')";
// std::string sql_inser3 = "insert into a (i, s) values(3, 'two')";
// std::string sql_update = "update a set s = 'three' where i = 3";
// std::string sql_select = "select i, s from a where i > 0";
// std::string sql_delete = "delete from a where i = 3";
auto node = parser.parse(sql_create);
executor.execute(*node.get());
return 0;
}

87
parser.cpp Normal file
View File

@@ -0,0 +1,87 @@
#include "parser.h"
#include "exception.h"
Parser::Parser() {
lexer = Lexer{};
}
std::unique_ptr<Node> Parser::parse(const std::string &code) {
lexer.parse(code);
lexer.debugTokens();
if (lexer.tokenType() == TokenType::keyword_create && lexer.nextTokenType() == TokenType::keyword_table) {
return parse_create_table();
} if (lexer.tokenType() == TokenType::keyword_select) {
return parse_select();
}
return std::make_unique<Node>(NodeType::error);
}
std::unique_ptr<Node> Parser::parse_create_table() {
std::vector<ColDefNode> cols_def {};
lexer.skipToken(TokenType::keyword_create);
lexer.skipToken(TokenType::keyword_table);
if (lexer.tokenType() != TokenType::identifier) { /* TODO handle error */ }
std::string table_name = lexer.currentToken().token_string;
lexer.nextToken();
lexer.skipToken(TokenType::open_paren);
do {
std::string column_name;
ColumnType column_type;
int column_len {1};
bool column_nullable {true};
// column name
if (lexer.tokenType() != TokenType::identifier) { /* TODO handle error */ }
column_name = lexer.currentToken().token_string;
lexer.nextToken();
// column type and optionaly len
if (lexer.tokenType() == TokenType::keyword_int) {
column_type = ColumnType::integer_type;
lexer.nextToken();
} else if (lexer.tokenType() == TokenType::keyword_float) {
column_type = ColumnType::float_type;
lexer.nextToken();
} else if (lexer.tokenType() == TokenType::keyword_varchar) {
column_type = ColumnType::varchar_type;
lexer.nextToken();
lexer.skipToken(TokenType::open_paren);
if (lexer.tokenType() == TokenType::int_number) {
column_len = std::stoi(lexer.currentToken().token_string);
lexer.nextToken();
} else { /* TODO handle error */ }
lexer.skipToken(TokenType::close_paren);
} else { /* TODO handle error */ }
if (lexer.tokenType() == TokenType::keyword_not) {
lexer.nextToken();
lexer.skipToken(TokenType::keyword_null);
column_nullable = false;
} else if (lexer.tokenType() == TokenType::keyword_null) {
lexer.nextToken();
}
cols_def.push_back(ColDefNode(column_name, column_type, column_len, column_nullable));
if (lexer.tokenType() == TokenType::comma) lexer.nextToken();
// TODO in future constraints
} while (lexer.tokenType() != TokenType::close_paren);
return std::make_unique<CreateTableNode>(table_name, cols_def);
}
std::unique_ptr<Node> Parser::parse_select() {
std::vector<Node> exec_code {};
return std::make_unique<Node>(NodeType::not_implemented_yet);
}

63
parser.h Normal file
View File

@@ -0,0 +1,63 @@
#pragma once
#include "lexer.h"
#include <stdexcept>
#include <string>
#include <vector>
enum class ColumnType {
integer_type,
float_type,
varchar_type
};
enum class NodeType {
create_table,
select_from,
column_def,
not_implemented_yet,
error
};
struct Node {
NodeType node_type;
Node(const NodeType type) : node_type(type) {}
};
struct ColDefNode : Node {
std::string name;
ColumnType type;
int length;
bool null;
ColDefNode(const std::string col_name, const ColumnType col_type, int col_len, bool nullable) :
Node(NodeType::column_def), name(col_name), type(col_type), length(col_len), null(nullable) {}
};
struct CreateTableNode : Node {
std::string table_name;
std::vector<ColDefNode> cols_defs;
CreateTableNode(const std::string name, std::vector<ColDefNode> defs) :
Node(NodeType::create_table), table_name(name), cols_defs(defs) {}
};
class Parser {
private:
public:
Parser();
std::unique_ptr<Node> parse(const std::string &code);
private:
std::unique_ptr<Node> parse_create_table();
std::unique_ptr<Node> parse_select();
private:
Lexer lexer;
};