diff --git a/dataStructure/polynomial.py b/dataStructure/polynomial.py
index a8f0384..8224dfc 100644
--- a/dataStructure/polynomial.py
+++ b/dataStructure/polynomial.py
@@ -197,4 +197,4 @@ def menu():
 
 
 if __name__ == '__main__':
-    pass
+    menu()
diff --git a/dataStructure/unionFindSet.py b/dataStructure/unionFindSet.py
index 9a05e7f..1d893e4 100644
--- a/dataStructure/unionFindSet.py
+++ b/dataStructure/unionFindSet.py
@@ -1,6 +1,7 @@
 class unionFindSet:
     def __init__(self, S):
         self.S = {i: i for i in S}
+        self.size = {i: 1 for i in S}
 
     def find(self, x):
         if x != self.S[x]:
@@ -9,4 +10,9 @@ class unionFindSet:
 
     def union(self, a, b, key=lambda x: x):
         x, y = sorted((self.find(a), self.find(b)), key=key)
-        self.S[a] = self.S[b] = self.S[y] = x
+        self.S[y] = x
+        if x != y:
+            self.size[x] += self.size[y]
+
+    def getSize(self, x):
+        return self.size[self.find(x)]
diff --git a/dynamicProgramming/subarraySum.py b/dynamicProgramming/subarraySum.py
new file mode 100644
index 0000000..72a1a85
--- /dev/null
+++ b/dynamicProgramming/subarraySum.py
@@ -0,0 +1,28 @@
+#coding: utf-8
+''' mbinary
+#######################################################################
+# File : subarraySum.py
+# Author: mbinary
+# Mail: zhuheqin1@gmail.com
+# Blog: https://mbinary.xyz
+# Github: https://github.com/mbinary
+# Created Time: 2020-04-20  16:49
+# Description: 子数组累加和
+#######################################################################
+'''
+from typing import List
+
+
+def subarraySum(nums: List[int], k: int) -> int:
+    dic = {0: 1}
+    sm = 0
+    count = 0
+    for i in range(len(nums)):
+        sm += nums[i]
+        if((sm-k) in dic):
+            count += dic[sm-k]
+        if(sm in dic):
+            dic[sm] += 1
+        else:
+            dic[sm] = 1
+    return count
diff --git a/parser/README.md b/parser/README.md
new file mode 100644
index 0000000..8881e67
--- /dev/null
+++ b/parser/README.md
@@ -0,0 +1,603 @@
+# parser
+先进行词法分析，然后进行语法分析，然后编写递归下降程序。可以将代码形成框架，词法分析每次只需要改变正则表达式部分即可，语法分析代码只需要实现语法对应的函数.
+我这里列举了4个题目，都可以这样解答。
+
+
+* [Lisp 语法解析](#lisp-语法解析)
+    * [题目](#题目)
+    * [语法](#语法)
+    * [代码](#代码)
+* [原子的数量](#原子的数量)
+    * [题目](#题目-1)
+    * [语法](#语法-1)
+    * [代码](#代码-1)
+* [花括号展开2](#花括号展开2)
+    * [题目](#题目-2)
+    * [语法](#语法-2)
+    * [代码](#代码-2)
+* [基本计算器4](#基本计算器4)
+    * [题目](#题目-3)
+    * [语法](#语法-3)
+    * [代码](#代码-3)
+* [更多](#更多)
+
+## [Lisp 语法解析](https://leetcode-cn.com/problems/parse-lisp-expression/)
+### 题目
+
+给定一个类似 Lisp 语句的表达式 expression，求出其计算结果。
+
+表达式语法如下所示:
+
+- 表达式可以为整数，let 语法，add 语法，mult 语法，或赋值的变量。表达式的结果总是一个整数。
+- (整数可以是正整数、负整数、0)
+- let 语法表示为 (let v1 e1 v2 e2 ... vn en expr), 其中 let 语法总是以字符串 "let" 来表示，接下来会跟随一个或多个交替变量或表达式，也就是说，第一个变量 v1 被分配为表达式 e1 的值，第二个变量 v2 被分配为表达式 e2 的值，以此类推；最终 let 语法的值为 expr 表达式的值。
+- add 语法表示为 (add e1 e2)，其中 add 语法总是以字符串 "add" 来表示，该语法总是有两个表达式 e1、e2, 该语法的最终结果是 e1 表达式的值与 e2 表达式的值之和。
+- mult 语法表示为 (mult e1 e2) ，其中 mult 语法总是以字符串 "mult" 表示， 该语法总是有两个表达式 e1、e2，该语法的最终结果是 e1 表达式的值与 e2 表达式的值之积。
+- 在该题目中，变量的命名以小写字符开始，之后跟随 0 个或多个小写字符或数字。为了方便，"add"，"let"，"mult" 会被定义为 "关键字"，不会在表达式的变量命名中出现。
+- 最后，要说一下作用域的概念。计算变量名所对应的表达式时，在计算上下文中，首先检查最内层作用域（按括号计），然后按顺序依次检查外部作用域。我们将保证每一个测试的表达式都是合法的。有关作用域的更多详细信息，请参阅示例。
+
+```
+>>> (let x -2 y x y)
+-2
+>>> (mult 3 (add 2 3))
+15
+>>> (let x 2 (mult x 5))
+10
+>>> (let x 2 (mult x (let x 3 y 4 (add x y))))
+14
+>>> (let x 2 x 3 x)
+3
+>>> (let x 2 (add (let x 3 (let x 4 x)) x))
+6
+>>> (let a1 3 b2 (add a1 1) b2)
+4
+```
+### 语法
+```
+S-> '(' expr ')'
+expr -> [mult|add] item item | let {word item } item 
+item -> num | word| S 
+```
+
+### 代码
+```python [lisp-收起]
+```
+```python [lisp-展开]
+import re
+from collections import namedtuple
+left = r'(?P<LEFT>\()'
+right = r'(?P<RIGHT>\))'
+word = r'(?P<WORD>[a-z][a-z0-9]*)'
+num = r'(?P<NUM>(\-)?\d+)'
+blank = r'(?P<BLANK>\s+)'
+pt = re.compile('|'.join([left, right, word, num, blank]))
+
+token = namedtuple('token', ['type', 'value'])
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        if i.lastgroup != 'BLANK':
+            yield token(i.lastgroup, i.group(0))
+
+
+class parser(object):
+    '''grammar:
+        S-> '(' expr ')'
+        expr -> [mult|add] item item | let {word item } item 
+        item -> num | word| S 
+    '''
+
+    def config(self,s):
+        if s:
+            self.token = [i for i in genToken(s)]
+            self.lookahead = 0
+            self.vars = []
+
+    def parse(self, s):
+        self.config(s)
+        try:
+            return self.S()
+        except Exception as e:
+            return e
+
+    def match(self, curType):
+        sym = self.token[self.lookahead]
+        if sym.type == curType:
+            self.lookahead += 1
+            return sym.value
+        self.errorinfo(f'Expected {curType}, got {sym.value}')
+
+    def errorinfo(self, s, k=None):
+        if k is None:
+            k = self.lookahead
+        pre = ' '.join([t.value for t in self.token[:k]])
+        suf = ' '.join([t.value for t in self.token[k:]])
+        print(pre+' '+suf)
+        print(' '*(len(pre)+1)+'^'*len(self.token[k].value))
+        raise Exception(s)
+
+    def readVar(self, var):
+        for dic in self.vars[::-1]:
+            if var in dic:
+                return dic[var]
+        self.errorinfo(f"Undefined varible '{var}'", self.lookahead-1)
+
+    def S(self):
+        self.vars.append({})
+        self.match('LEFT')
+        ret = self.expr()
+        self.match('RIGHT')
+        self.vars.pop()
+        return ret
+
+    def expr(self):
+        op = self.match('WORD')
+        if op == 'let':
+            while self.token[self.lookahead].type != 'RIGHT':
+                if self.token[self.lookahead].type == 'WORD':
+                    var = self.match('WORD')
+                    if self.token[self.lookahead].type == 'RIGHT':
+                        return self.readVar(var)
+                    else:
+                        self.vars[-1][var] = self.item()
+                else:
+                    return self.item()
+        elif op in {'mult', 'add'}:
+            a = self.item()
+            b = self.item()
+            if op == 'mult':
+                return a*b
+            elif op == 'add':
+                return a+b
+        else:
+            self.errorinfo('Unknown keyword', self.lookahead-1)
+
+    def item(self):
+        if self.token[self.lookahead].type == 'WORD':
+            return self.readVar(self.match('WORD'))
+        elif self.token[self.lookahead].type == 'NUM':
+            return int(self.match('NUM'))
+        else:
+            return self.S()
+
+
+class Solution(object):
+    def evaluate(self, expression: str) -> int:
+        return parser().parse(expression)
+
+
+if __name__ == "__main__":
+    sol = Solution()
+
+    exprs = ['(add -1 2)',
+             '(let x -2 y x y)',
+             '(mult 3 (add 2 3))',
+             '(let x 2 (mult x 5))',
+             '(let x 2 (mult x (let x 3 y 4 (add x y))))',
+             '(let x 2 x 3 x)',
+             '(let x 2 (add (let x 3 (let x 4 x)) x))',
+             '(let a1 3 b2 (add a1 1) b2)',
+             'add 1 2)',  # wrongs
+             '(asd 1 2)',
+             '(let a 1 b)',
+             '(let a 1 b 2)'
+             ]
+    for e in exprs:
+        print('>>>', e)
+        print(sol.evaluate(e))
+```
+
+
+## [原子的数量](https://leetcode-cn.com/problems/number-of-atoms)
+### 题目
+
+给定一个化学式 formula（作为字符串），返回每种原子的数量。
+
+原子总是以一个大写字母开始，接着跟随 0 个或任意个小写字母，表示原子的名字。
+
+如果数量大于 1，原子后会跟着数字表示原子的数量。如果数量等于 1 则不会跟数字。例如，H2O 和 H2O2 是可行的，但 H1O2 这个表达是不可行的。
+
+两个化学式连在一起是新的化学式。例如 H2O2He3Mg4 也是化学式。
+
+一个括号中的化学式和数字（可选择性添加）也是化学式。例如 (H2O2) 和 (H2O2) 3 是化学式。
+
+给定一个化学式，输出所有原子的数量。格式为：第一个（按字典序）原子的名子，跟着它的数量（如果数量大于 1），然后是第二个原子的名字（按字典序），跟着它的数量（如果数量大于 1），以此类推。
+```
+>>> K4(ON(SO3)2)2
+K4N2O14S4
+>>> Mg(OH)2
+H2MgO2
+```
+
+### 语法
+```
+S-> item | S item
+item -> word | word num | '(' S ')' num 
+```
+
+### 代码
+```python [atom-收起]
+```
+```python3 [atom-展开]
+import re
+from collections import namedtuple
+left = r'(?P<LEFT>\()'
+right = r'(?P<RIGHT>\))'
+word = r'(?P<WORD>[A-Z][a-z]*)'
+num = r'(?P<NUM>\d+)'
+pt = re.compile('|'.join([left, right, word, num]))
+
+token = namedtuple('token', ['type', 'value'])
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        yield token(i.lastgroup, i.group(0))
+
+
+class parser:
+    '''grammar:
+        S-> item | S item
+        item -> word | word num | '(' S ')' num 
+    '''
+
+    def match(self, curType):
+        sym = self.token[self.lookahead]
+        if sym.type == curType:
+            self.lookahead += 1
+            return sym.value
+        raise Exception('Invalid input string')
+
+    def parse(self, s):
+        self.token = [i for i in genToken(s)]
+        self.lookahead = 0
+        return self.S()
+    def S(self):
+        dic = {}
+        while self.lookahead < len(self.token) and self.token[self.lookahead].type != 'RIGHT':
+            cur = self.item()
+            for i in cur:
+                if i in dic:
+                    dic[i] += cur[i]
+                else:
+                    dic[i] = cur[i]
+        return dic
+
+    def item(self):
+        if self.token[self.lookahead].type == 'WORD':
+            ele = self.match('WORD')
+            n = 1
+            if self.lookahead < len(self.token) and self.token[self.lookahead].type == 'NUM':
+                n = int(self.match('NUM'))
+            return {ele: n}
+        elif self.token[self.lookahead].type == 'LEFT':
+            self.match('LEFT')
+            dic = self.S()
+            self.match('RIGHT')
+            n = int(self.match("NUM"))
+            for i in dic:
+                dic[i] *= n
+            return dic
+        else:
+            print(self.token[self.lookahead])
+            raise Exception('invalid string')
+
+
+class Solution(object):
+    def countOfAtoms(self, formula):
+        """
+        :type formula: str
+        :rtype: str
+        """
+        dic = parser().parse(formula)
+        return ''.join(c+str(dic[c]) if dic[c] != 1 else c for c in sorted(dic.keys()))
+
+
+if __name__ == "__main__":
+    li = ["K4(ON(SO3)2)2","Mg(OH)2"]
+    sol = Solution()
+    for s in li:
+        print('>>>',s)
+        print(sol.countOfAtoms(s))
+```
+
+## [花括号展开2](https://leetcode-cn.com/problems/brace-expansion-ii/)
+### 题目
+如果你熟悉 Shell 编程，那么一定了解过花括号展开，它可以用来生成任意字符串。
+
+花括号展开的表达式可以看作一个由 花括号、逗号 和 小写英文字母 组成的字符串，定义下面几条语法规则：
+
+如果只给出单一的元素 x，那么表达式表示的字符串就只有 "x"。 
+例如，表达式 {a} 表示字符串 "a"。
+而表达式 {ab} 就表示字符串 "ab"。
+当两个或多个表达式并列，以逗号分隔时，我们取这些表达式中元素的并集。
+例如，表达式 {a,b,c} 表示字符串 "a","b","c"。
+而表达式 {a,b},{b,c} 也可以表示字符串 "a","b","c"。
+要是两个或多个表达式相接，中间没有隔开时，我们从这些表达式中各取一个元素依次连接形成字符串。
+例如，表达式 {a,b}{c,d} 表示字符串 "ac","ad","bc","bd"。
+表达式之间允许嵌套，单一元素与表达式的连接也是允许的。
+```
+>>> {a,b}{c{d,e}}
+['acd', 'ace', 'bcd', 'bce']
+>>> {{a,z}, a{b,c}, {ab,z}}
+['a', 'ab', 'ac', 'z']
+>>> {a,b}c{d,e}f
+['acdf', 'acef', 'bcdf', 'bcef']
+```
+### 语法
+```
+expr -> item | item ',' expr
+item -> factor | factor item
+factor -> WORD | '{' expr '}'
+```
+
+### 代码
+```python [brace-收起]
+```
+```python3 [brace-展开]
+import re
+from collections import namedtuple
+
+token = namedtuple('token', ['type', 'value'])
+
+
+left = r'(?P<LEFT>\{)'
+right = r'(?P<RIGHT>\})'
+word = r'(?P<WORD>[a-z]+)'
+comma = r'(?P<COMMA>\,)'
+blank = r'(?P<BLANK>\s)'
+pt = re.compile('|'.join([left, right, word, comma, blank]))
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        if i.lastgroup != 'BLANK':
+            yield token(i.lastgroup, i.group(0))
+
+
+class parser:
+    '''gramar
+        expr -> item | item ',' expr
+        item -> factor | factor item
+        factor -> WORD | '{' expr '}'
+    '''
+
+    def match(self, tp):
+        # print(self.p.value)
+        if tp == self.p.type:
+            val = self.p.value
+            try:
+                self.p = next(self.gen)
+            except StopIteration:
+                self.p = None
+            except Exception as e:
+                print(e)
+            return val
+        else:
+            raise Exception(f"[Error]: {tp} expected, got {self.p.type}")
+
+    def parse(self, s):
+        self.gen = genToken(s)
+        self.p = next(self.gen)
+        st = self.expr()
+        return sorted(list(st))
+
+    def expr(self):
+        ret = self.item()
+        while self.p and self.p.type == 'COMMA':
+            self.match('COMMA')
+            ret = ret.union(self.item())
+        return ret
+
+    def item(self):
+        ret = self.factor()
+        while self.p and self.p.type in ['WORD', 'LEFT']:
+            sufs = self.factor()
+            new = set()
+            for pre in ret:
+                for suf in sufs:
+                    new.add(pre+suf)
+            ret = new
+        return ret
+
+    def factor(self):
+        if self.p.type == 'LEFT':
+            self.match('LEFT')
+            ret = self.expr()
+            self.match('RIGHT')
+            return ret
+        return {self.match('WORD')}
+
+
+class Solution:
+    def braceExpansionII(self, expression):
+        return parser().parse(expression)
+
+
+if __name__ == '__main__':
+    sol = Solution()
+    li = ["{a,b}{c{d,e}}", "{{a,z}, a{b,c}, {ab,z}}", "{a,b}c{d,e}f"]
+    for i in li:
+        print('>>>', i)
+        print(sol.braceExpansionII(i))
+```
+## [基本计算器4](https://leetcode-cn.com/problems/basic-calculator-iv/)
+### 题目
+给定一个表达式 expression 如 expression = "e + 8 - a + 5" 和一个求值映射，如 {"e": 1}（给定的形式为 evalvars = ["e"] 和 evalints = [1]），返回表示简化表达式的标记列表，例如 ["-1*a","14"]
+
+表达式交替使用块和符号，每个块和符号之间有一个空格。
+块要么是括号中的表达式，要么是变量，要么是非负整数。
+块是括号中的表达式，变量或非负整数。
+变量是一个由小写字母组成的字符串（不包括数字）。请注意，变量可以是多个字母，并注意变量从不具有像 "2x" 或 "-x" 这样的前导系数或一元运算符 。
+表达式按通常顺序进行求值：先是括号，然后求乘法，再计算加法和减法。例如，expression = "1 + 2 * 3" 的答案是 ["7"]。
+
+输出格式如下：
+
+对于系数非零的每个自变量项，我们按字典排序的顺序将自变量写在一个项中。例如，我们永远不会写像 “b*a*c” 这样的项，只写 “a*b*c”。
+项的次数等于被乘的自变量的数目，并计算重复项。(例如，"a*a*b*c" 的次数为 4。)。我们先写出答案的最大次数项，用字典顺序打破关系，此时忽略词的前导系数。
+项的前导系数直接放在左边，用星号将它与变量分隔开 (如果存在的话)。前导系数 1 仍然要打印出来。
+格式良好的一个示例答案是 ["-2*a*a*a", "3*a*a*b", "3*b*b", "4*a", "5*c", "-6"] 。
+系数为 0 的项（包括常数项）不包括在内。例如，“0” 的表达式输出为 []。
+```
+>>> ((a - b) * (b - c) + (c - a)) * ((a - b) + (b - c) * (c - a)) [] []
+['-1*a*a*b*b', '2*a*a*b*c', '-1*a*a*c*c', '1*a*b*b*b', '-1*a*b*b*c', '-1*a*b*c*c', '1*a*c*c*c', '-1*b*b*b*c', '2*b*b*c*c', '-1*b*c*c*c', '2*a*a*b', '-2*a*a*c', '-2*a*b*b', '2*a*c*c', '1*b*b*b', '-1*b*b*c', '1*b*c*c', '-1*c*c*c', '-1*a*a', '1*a*b', '1*a*c', '-1*b*c']
+>>> e + 8 - a + 5 ['e'] [1]
+['-1*a', '14']
+```
+
+### 语法
+```
+expr -> expr {'+'|'-'} term | term
+term -> term '*' item | item
+item -> num | var | '(' expr ')'
+```
+
+### 代码
+
+```python [cal-收起]
+```
+```python3 [cal-展开]
+import re
+from collections import namedtuple
+left = r'(?P<LEFT>\()'
+right = r'(?P<RIGHT>\))'
+var = r'(?P<VAR>[a-z]+)'
+num = r'(?P<NUM>\d+)'
+add = r'(?P<ADD>\+)'
+sub = r'(?P<SUB>\-)'
+mul = r'(?P<MUL>\*)'
+ws = r'(?P<WS> +)'
+pt = re.compile('|'.join([left, right, var, ws, num, add, sub, mul]))
+
+token = namedtuple('token', ['type', 'value'])
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        if i.lastgroup != 'WS':
+            yield token(i.lastgroup, i.group(0))
+
+
+class parser(object):
+    '''grammar
+        expr -> expr {'+'|'-'} term | term
+        term -> term '*' item | item
+        item -> num | var | '(' expr ')'
+    '''
+
+    def __init__(self, s, vars):
+        self.token = [i for i in genToken(s)]
+        self.lookahead = 0
+        self.var = vars
+
+    def parse(self):
+        dic = self.term()
+        # terminate symbol
+        while self.lookahead < len(self.token) and not self.isType('RIGHT'):
+            assert self.isType('SUB', 'ADD')
+            sign = 1 if self.match() == '+' else -1
+            var = self.term()
+            for i in var:
+                if i in dic:
+                    dic[i] += var[i]*sign
+                else:
+                    dic[i] = var[i]*sign
+        return dic
+
+    def match(self, curType=None):
+        sym = self.token[self.lookahead]
+        # print(sym,curType)
+        if curType is None or sym.type == curType:
+            self.lookahead += 1
+            return sym.value
+        raise Exception('Invalid input string')
+
+    def isType(self, *s):
+        sym = self.token[self.lookahead]
+        return any(sym.type == i for i in s)
+
+    def term(self):
+        li = []
+        dic = self.item()
+        while self.lookahead < len(self.token) and self.isType('MUL'):
+            self.match()
+            li.append(self.item())
+        for d2 in li:
+            newDic = {}
+            for v1 in dic:
+                for v2 in d2:
+                    s = ''
+                    if v1 == '':
+                        s = v2
+                    elif v2 == '':
+                        s = v1
+                    else:
+                        s = '*'.join(sorted(v1.split('*')+v2.split('*')))
+                    if s in newDic:
+                        newDic[s] += dic[v1]*d2[v2]
+                    else:
+                        newDic[s] = dic[v1]*d2[v2]
+            dic = newDic
+        return dic
+
+    def item(self):
+        if self.isType('NUM'):
+            return {'': int(self.match())}
+        elif self.isType('VAR'):
+            name = self.match()
+            if name in self.var:
+                return {'': self.var[name]}
+            else:
+                return {name: 1}
+        elif self.isType('LEFT'):
+            self.match()
+            dic = self.parse()
+            self.match('RIGHT')
+            return dic
+        else:
+            print(self.token[self.lookahead])
+            raise Exception('invalid string')
+
+
+class Solution:
+    def basicCalculatorIV(self, expression, evalvars, evalints):
+        """
+        :type expression: str
+        :type evalvars: List[str]
+        :type evalints: List[int]
+        :rtype: List[str]
+        """
+        self.var = dict(zip(evalvars, evalints))
+        dic = parser(expression, self.var).parse()
+        n = dic.pop('') if '' in dic else 0
+        ret = []
+        li = sorted(dic, key=lambda s: (-s.count('*'), s))
+        for i in li:
+            if dic[i] != 0:
+                s = str(dic[i])
+                ret.append(s + ('*'+i) if i else s)
+        if n != 0:
+            ret.append(str(n))
+        return ret
+
+
+if __name__ == '__main__':
+    sol = Solution()
+    exprs = [
+        "((a - b) * (b - c) + (c - a)) * ((a - b) + (b - c) * (c - a))", "e + 8 - a + 5"]
+    names = [[], ["e"]]
+    vars = [[], [1]]
+    for i, j, k in zip(exprs, names, vars):
+        print('>>>', i, j, k)
+        print(sol.basicCalculatorIV(i, j, k))
+```
+
+## 更多
+- [迷你语法分析器](https://leetcode-cn.com/problems/mini-parser/)
+- [字符串解码](https://leetcode-cn.com/problems/decode-string/)
+
diff --git a/parser/brace_expansion.py b/parser/brace_expansion.py
new file mode 100644
index 0000000..1c1abd9
--- /dev/null
+++ b/parser/brace_expansion.py
@@ -0,0 +1,86 @@
+import re
+from collections import namedtuple
+
+token = namedtuple('token', ['type', 'value'])
+
+
+left = r'(?P<LEFT>\{)'
+right = r'(?P<RIGHT>\})'
+word = r'(?P<WORD>[a-z]+)'
+comma = r'(?P<COMMA>\,)'
+blank = r'(?P<BLANK>\s)'
+pt = re.compile('|'.join([left, right, word, comma, blank]))
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        if i.lastgroup != 'BLANK':
+            yield token(i.lastgroup, i.group(0))
+
+
+class parser:
+    '''gramar
+        expr -> item | item ',' expr
+        item -> factor | factor item
+        factor -> WORD | '{' expr '}'
+    '''
+
+    def match(self, tp):
+        # print(self.p.value)
+        if tp == self.p.type:
+            val = self.p.value
+            try:
+                self.p = next(self.gen)
+            except StopIteration:
+                self.p = None
+            except Exception as e:
+                print(e)
+            return val
+        else:
+            raise Exception(f"[Error]: {tp} expected, got {self.p.type}")
+
+    def parse(self, s):
+        self.gen = genToken(s)
+        self.p = next(self.gen)
+        st = self.expr()
+        return sorted(list(st))
+
+    def expr(self):
+        ret = self.item()
+        while self.p and self.p.type == 'COMMA':
+            self.match('COMMA')
+            ret = ret.union(self.item())
+        return ret
+
+    def item(self):
+        ret = self.factor()
+        while self.p and self.p.type in ['WORD', 'LEFT']:
+            sufs = self.factor()
+            new = set()
+            for pre in ret:
+                for suf in sufs:
+                    new.add(pre+suf)
+            ret = new
+        return ret
+
+    def factor(self):
+        if self.p.type == 'LEFT':
+            self.match('LEFT')
+            ret = self.expr()
+            self.match('RIGHT')
+            return ret
+        return {self.match('WORD')}
+
+
+class Solution:
+    def braceExpansionII(self, expression):
+        return parser().parse(expression)
+
+
+if __name__ == '__main__':
+    sol = Solution()
+    li = ["{a,b}{c{d,e}}", "{{a,z}, a{b,c}, {ab,z}}", "{a,b}c{d,e}f"]
+    for i in li:
+        print('>>>', i)
+        print(sol.braceExpansionII(i))
diff --git a/parser/calculator.py b/parser/calculator.py
new file mode 100644
index 0000000..c567c82
--- /dev/null
+++ b/parser/calculator.py
@@ -0,0 +1,134 @@
+import re
+from collections import namedtuple
+left = r'(?P<LEFT>\()'
+right = r'(?P<RIGHT>\))'
+var = r'(?P<VAR>[a-z]+)'
+num = r'(?P<NUM>\d+)'
+add = r'(?P<ADD>\+)'
+sub = r'(?P<SUB>\-)'
+mul = r'(?P<MUL>\*)'
+ws = r'(?P<WS> +)'
+pt = re.compile('|'.join([left, right, var, ws, num, add, sub, mul]))
+
+token = namedtuple('token', ['type', 'value'])
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        if i.lastgroup != 'WS':
+            yield token(i.lastgroup, i.group(0))
+
+
+class parser(object):
+    '''grammar
+        expr -> expr {'+'|'-'} term | term
+        term -> term '*' item | item
+        item -> num | var | '(' expr ')'
+    '''
+
+    def __init__(self, s, vars):
+        self.token = [i for i in genToken(s)]
+        self.lookahead = 0
+        self.var = vars
+
+    def parse(self):
+        dic = self.term()
+        # terminate symbol
+        while self.lookahead < len(self.token) and not self.isType('RIGHT'):
+            assert self.isType('SUB', 'ADD')
+            sign = 1 if self.match() == '+' else -1
+            var = self.term()
+            for i in var:
+                if i in dic:
+                    dic[i] += var[i]*sign
+                else:
+                    dic[i] = var[i]*sign
+        return dic
+
+    def match(self, curType=None):
+        sym = self.token[self.lookahead]
+        # print(sym,curType)
+        if curType is None or sym.type == curType:
+            self.lookahead += 1
+            return sym.value
+        raise Exception('Invalid input string')
+
+    def isType(self, *s):
+        sym = self.token[self.lookahead]
+        return any(sym.type == i for i in s)
+
+    def term(self):
+        li = []
+        dic = self.item()
+        while self.lookahead < len(self.token) and self.isType('MUL'):
+            self.match()
+            li.append(self.item())
+        for d2 in li:
+            newDic = {}
+            for v1 in dic:
+                for v2 in d2:
+                    s = ''
+                    if v1 == '':
+                        s = v2
+                    elif v2 == '':
+                        s = v1
+                    else:
+                        s = '*'.join(sorted(v1.split('*')+v2.split('*')))
+                    if s in newDic:
+                        newDic[s] += dic[v1]*d2[v2]
+                    else:
+                        newDic[s] = dic[v1]*d2[v2]
+            dic = newDic
+        return dic
+
+    def item(self):
+        if self.isType('NUM'):
+            return {'': int(self.match())}
+        elif self.isType('VAR'):
+            name = self.match()
+            if name in self.var:
+                return {'': self.var[name]}
+            else:
+                return {name: 1}
+        elif self.isType('LEFT'):
+            self.match()
+            dic = self.parse()
+            self.match('RIGHT')
+            return dic
+        else:
+            print(self.token[self.lookahead])
+            raise Exception('invalid string')
+
+
+class Solution:
+    def basicCalculatorIV(self, expression, evalvars, evalints):
+        """
+        :type expression: str
+        :type evalvars: List[str]
+        :type evalints: List[int]
+        :rtype: List[str]
+        """
+        self.var = dict(zip(evalvars, evalints))
+        dic = parser(expression, self.var).parse()
+        n = dic.pop('') if '' in dic else 0
+        ret = []
+        li = sorted(dic, key=lambda s: (-s.count('*'), s))
+        for i in li:
+            if dic[i] != 0:
+                s = str(dic[i])
+                ret.append(s + ('*'+i) if i else s)
+        if n != 0:
+            ret.append(str(n))
+        return ret
+
+
+if __name__ == '__main__':
+    sol = Solution()
+    exprs = [
+        "((a - b) * (b - c) + (c - a)) * ((a - b) + (b - c) * (c - a))", "e + 8 - a + 5"]
+    names = [[], ["e"]]
+    vars = [[], [1]]
+    for i, j, k in zip(exprs, names, vars):
+        print('>>>', i, j, k)
+        print(sol.basicCalculatorIV(i, j, k))
diff --git a/parser/lisp_expression.py b/parser/lisp_expression.py
new file mode 100644
index 0000000..6a83c76
--- /dev/null
+++ b/parser/lisp_expression.py
@@ -0,0 +1,124 @@
+import re
+from collections import namedtuple
+left = r'(?P<LEFT>\()'
+right = r'(?P<RIGHT>\))'
+word = r'(?P<WORD>[a-z][a-z0-9]*)'
+num = r'(?P<NUM>(\-)?\d+)'
+blank = r'(?P<BLANK>\s+)'
+pt = re.compile('|'.join([left, right, word, num, blank]))
+
+token = namedtuple('token', ['type', 'value'])
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        if i.lastgroup != 'BLANK':
+            yield token(i.lastgroup, i.group(0))
+
+
+class parser(object):
+    '''grammar:
+        S-> '(' expr ')'
+        expr -> [mult|add] item item | let {word item } item 
+        item -> num | word| S 
+    '''
+
+    def config(self,s):
+        if s:
+            self.token = [i for i in genToken(s)]
+            self.lookahead = 0
+            self.vars = []
+
+    def parse(self, s):
+        self.config(s)
+        try:
+            return self.S()
+        except Exception as e:
+            return e
+
+    def match(self, curType):
+        sym = self.token[self.lookahead]
+        if sym.type == curType:
+            self.lookahead += 1
+            return sym.value
+        self.errorinfo(f'Expected {curType}, got {sym.value}')
+
+    def errorinfo(self, s, k=None):
+        if k is None:
+            k = self.lookahead
+        pre = ' '.join([t.value for t in self.token[:k]])
+        suf = ' '.join([t.value for t in self.token[k:]])
+        print(pre+' '+suf)
+        print(' '*(len(pre)+1)+'^'*len(self.token[k].value))
+        raise Exception(s)
+
+    def readVar(self, var):
+        for dic in self.vars[::-1]:
+            if var in dic:
+                return dic[var]
+        self.errorinfo(f"Undefined varible '{var}'", self.lookahead-1)
+
+    def S(self):
+        self.vars.append({})
+        self.match('LEFT')
+        ret = self.expr()
+        self.match('RIGHT')
+        self.vars.pop()
+        return ret
+
+    def expr(self):
+        op = self.match('WORD')
+        if op == 'let':
+            while self.token[self.lookahead].type != 'RIGHT':
+                if self.token[self.lookahead].type == 'WORD':
+                    var = self.match('WORD')
+                    if self.token[self.lookahead].type == 'RIGHT':
+                        return self.readVar(var)
+                    else:
+                        self.vars[-1][var] = self.item()
+                else:
+                    return self.item()
+        elif op in {'mult', 'add'}:
+            a = self.item()
+            b = self.item()
+            if op == 'mult':
+                return a*b
+            elif op == 'add':
+                return a+b
+        else:
+            self.errorinfo('Unknown keyword', self.lookahead-1)
+
+    def item(self):
+        if self.token[self.lookahead].type == 'WORD':
+            return self.readVar(self.match('WORD'))
+        elif self.token[self.lookahead].type == 'NUM':
+            return int(self.match('NUM'))
+        else:
+            return self.S()
+
+
+class Solution(object):
+    def evaluate(self, expression: str) -> int:
+        return parser().parse(expression)
+
+
+if __name__ == "__main__":
+    sol = Solution()
+
+    exprs = ['(add -1 2)',
+             '(let x -2 y x y)',
+             '(mult 3 (add 2 3))',
+             '(let x 2 (mult x 5))',
+             '(let x 2 (mult x (let x 3 y 4 (add x y))))',
+             '(let x 2 x 3 x)',
+             '(let x 2 (add (let x 3 (let x 4 x)) x))',
+             '(let a1 3 b2 (add a1 1) b2)',
+             'add 1 2)',  # wrongs
+             '(asd 1 2)',
+             '(let a 1 b)',
+             '(let a 1 b 2)'
+             ]
+    for e in exprs:
+        print('>>>', e)
+        print(sol.evaluate(e))
diff --git a/parser/number_of_atoms.py b/parser/number_of_atoms.py
new file mode 100644
index 0000000..db7f571
--- /dev/null
+++ b/parser/number_of_atoms.py
@@ -0,0 +1,81 @@
+import re
+from collections import namedtuple
+left = r'(?P<LEFT>\()'
+right = r'(?P<RIGHT>\))'
+word = r'(?P<WORD>[A-Z][a-z]*)'
+num = r'(?P<NUM>\d+)'
+pt = re.compile('|'.join([left, right, word, num]))
+
+token = namedtuple('token', ['type', 'value'])
+
+
+def genToken(s):
+    scanner = pt.scanner(s)
+    for i in iter(scanner.match, None):
+        yield token(i.lastgroup, i.group(0))
+
+
+class parser:
+    '''grammar:
+        S-> item | S item
+        item -> word | word num | '(' S ')' num 
+    '''
+
+    def match(self, curType):
+        sym = self.token[self.lookahead]
+        if sym.type == curType:
+            self.lookahead += 1
+            return sym.value
+        raise Exception('Invalid input string')
+
+    def parse(self, s):
+        self.token = [i for i in genToken(s)]
+        self.lookahead = 0
+        return self.S()
+    def S(self):
+        dic = {}
+        while self.lookahead < len(self.token) and self.token[self.lookahead].type != 'RIGHT':
+            cur = self.item()
+            for i in cur:
+                if i in dic:
+                    dic[i] += cur[i]
+                else:
+                    dic[i] = cur[i]
+        return dic
+
+    def item(self):
+        if self.token[self.lookahead].type == 'WORD':
+            ele = self.match('WORD')
+            n = 1
+            if self.lookahead < len(self.token) and self.token[self.lookahead].type == 'NUM':
+                n = int(self.match('NUM'))
+            return {ele: n}
+        elif self.token[self.lookahead].type == 'LEFT':
+            self.match('LEFT')
+            dic = self.S()
+            self.match('RIGHT')
+            n = int(self.match("NUM"))
+            for i in dic:
+                dic[i] *= n
+            return dic
+        else:
+            print(self.token[self.lookahead])
+            raise Exception('invalid string')
+
+
+class Solution(object):
+    def countOfAtoms(self, formula):
+        """
+        :type formula: str
+        :rtype: str
+        """
+        dic = parser().parse(formula)
+        return ''.join(c+str(dic[c]) if dic[c] != 1 else c for c in sorted(dic.keys()))
+
+
+if __name__ == "__main__":
+    li = ["K4(ON(SO3)2)2","Mg(OH)2"]
+    sol = Solution()
+    for s in li:
+        print('>>>',s)
+        print(sol.countOfAtoms(s))