import jqdatasdk

jqdatasdk.auth(tel, pwd)

# 获取行业列表
# "sw_l1": 申万一级行业
# "sw_l2": 申万二级行业
# "sw_l3": 申万三级行业
# "jq_l1": 聚宽一级行业
# "jq_l2": 聚宽二级行业
# "zjw": 证监会行业
industries_list = jqdatasdk.get_industries(name='zjw')

industries_list

# 按行业列表获取每个行业公司的列表
def get_complany_lists(industries_list):
    complany_lists = []
    for industrie in industries_list:
        complany_lists.append(jqdatasdk.get_industry_stocks(industrie))
        
    return complany_lists

# 行业内公司的员工信息
def get_employee_df_lists(complany_lists, end_date_list):
    employee_df_lists = []

    for complany_list in complany_lists:
        q=jqdatasdk.query(jqdatasdk.finance.STK_EMPLOYEE_INFO).filter(jqdatasdk.finance.STK_EMPLOYEE_INFO.code.in_(complany_list),jqdatasdk.finance.STK_EMPLOYEE_INFO.end_date.in_(end_date_list)).limit(3000)
        employee_df_lists.append(jqdatasdk.finance.run_query(q))
        
    return employee_df_lists

# 计算从2012-2017年行业人数均值
def compute_employee_info(employee_df_lists):
    dict_list = []
    for employee_df in employee_df_lists:
        employee_list = []
        employee_dict = {}
        for name, group in employee_df.groupby('end_date'):
            employee_list.append(group['employee'].sum() / len(group))
            employee_dict[name] = group['employee'].sum() / len(group)
            
        dict_list.append(employee_dict)
        
    return dict_list

industries_list = jqdatasdk.get_industries(name='zjw')
end_date_list = ['2012-12-31', '2013-12-31', '2014-12-31', '2015-12-31', '2016-12-31', '2017-12-31']
complany_lists = get_complany_lists(industries_list.index)
employee_df_lists = get_employee_df_lists(complany_lists, end_date_list)
dict_list = compute_employee_info(employee_df_lists)

import matplotlib.pyplot as plt
plt.rcParams['font.sans-serif']=['SimHei'] #用来正常显示中文标签
plt.rcParams['axes.unicode_minus']=False #用来正常显示负号

industries_dict = {}
for i in range(len(dict_list)):
    employee_dict = dict_list[i]
    industries_dict[industries_list['name'].iloc[i]] = employee_dict

def plot_industrie_info(industrie_name):
    employee_dict = industries_dict[industrie_name]
    fig = plt.figure(figsize=(16,6))
    ax1 = fig.add_subplot(1,1,1)
    ax1.plot(list(employee_dict.keys()), list(employee_dict.values()))
    ax1.set_title(industrie_name)
    plt.show()

plot_industrie_info('软件和信息技术服务业')

plot_industrie_info('互联网和相关服务')

# 人数前十行业展示
sort_dict = {}
for industrie,employee_dict in industries_dict.items():
    if employee_dict:
        sort_dict[industrie] = list(employee_dict.values())[-1]
        
sort_dict = sorted(sort_dict.items(),key = lambda x:x[1],reverse = True)

fig = plt.figure(figsize=(16,6))
ax1 = fig.add_subplot(1,1,1)
for i in range(len(sort_dict)):
    if i == 10:
        break
    
    employee_dict = industries_dict[sort_dict[i][0]]
    ax1.plot(list(employee_dict.keys()), list(employee_dict.values()), label=sort_dict[i][0])
    
plt.legend()
plt.show()

# 对行业增长情况进行排序
sort_dict = {}
for industrie,employee_dict in industries_dict.items():
    if employee_dict:
        group_percent = 0
        for value in employee_dict.values():
            if value != 0:
                group_percent = (list(employee_dict.values())[-1] - value) / value
                break
                
        sort_dict[industrie] = group_percent
        
sort_dict = sorted(sort_dict.items(),key = lambda x:x[1],reverse = True)

sort_dict[:20]

fig = plt.figure(figsize=(16,6))
ax1 = fig.add_subplot(1,1,1)
for i in range(len(sort_dict)):
    if i == 20:
        break
    
    employee_dict = industries_dict[sort_dict[i][0]]
    ax1.plot(list(employee_dict.keys()), list(employee_dict.values()), label=sort_dict[i][0])
    
plt.legend()
plt.show()