def fetch_A_secucode_innercode_map(juyuan_conn: sql_base.Connection):
'''
给出任意一个 SecuCode, 不管是上市还是退市, 不管是变更前还是变更后, 均在这此找到唯一的InnerCode
'''
sql = """
(
select A.SecuCode, A.InnerCode, A.SecuAbbr, A.ListedDate, B.ChangeDate '退市日期'
from gildata.SecuMain A
left join
gildata.LC_ListStatus B
on A.InnerCode = B.InnerCode
and B.ChangeType = 4
where
A.SecuMarket in (83, 90)
and A.SecuCategory in (1, 41)
and A.ListedSector in (1, 2, 6, 7)
and A.ListedDate <= CURDATE()
)
UNION
(
SELECT
A.SecuCode, A.InnerCode, B.SecuAbbr, A.BeginDate '启用日期', A.StopDate '停用日期'
FROM gildata.LC_CodeChange A
JOIN gildata.SecuMain B
ON A.InnerCode = B.InnerCode
AND B.SecuMarket IN (83,90)
AND B.SecuCategory in (1, 41)
WHERE
A.CodeDefine = 1
AND A.SecuCode <> B.SecuCode
);
"""
exec_res = juyuan_conn.query(sql)
map1 = {}
for one in exec_res:
map1[one['SecuCode']] = one['InnerCode']
sql = f'''select InnerCode, SecuMarket from secumain where InnerCode in {tuple(map1.values())};'''
res = juyuan_conn.query(sql)
map2 = {}
for r in res:
map2[r['InnerCode']] = r['SecuMarket']
info = {}
for k, v in map1.items():
if map2[v] == 83:
k = "SH" + k
info[k] = v
elif map2[v] == 90:
k = "SZ" + k
info[k] = v
else:
raise
return info
class SourceAnnouncementBaseV1(object):
"""将两个爬虫表合并生成公告基础表"""
def __init__(self):
self.merge_table_name = 'announcement_base'
self.his_table = 'juchao_ant'
self.live_table = 'juchao_kuaixun'
self.batch_number = 10000
self._r_spider_conn = Connection(
host=R_SPIDER_MYSQL_HOST,
port=R_SPIDER_MYSQL_PORT,
user=R_SPIDER_MYSQL_USER,
password=R_SPIDER_MYSQL_PASSWORD,
database=R_SPIDER_MYSQL_DB,
)
self._spider_conn = Connection(
host=SPIDER_MYSQL_HOST,
port=SPIDER_MYSQL_PORT,
user=SPIDER_MYSQL_USER,
password=SPIDER_MYSQL_PASSWORD,
database=SPIDER_MYSQL_DB,
)
def daily_update(self, deadline: datetime.datetime = None):
if deadline is None:
deadline = datetime.datetime.now() - datetime.timedelta(days=1)
load_sql = '''select id, SecuCode, SecuAbbr, AntTime as PubDatetime1, \
AntTitle as Title1, AntDoc as PDFLink, CREATETIMEJZ as InsertDatetime1 from {} where \
UPDATETIMEJZ > '{}'; '''.format(self.his_table, deadline)
logger.info(load_sql)
items = []
datas = self._r_spider_conn.query(load_sql)
logger.info(len(datas))
for data in datas:
data = utils.process_secucode(data)
if data:
items.append(data)
self._spider_conn.batch_insert(items, self.merge_table_name,
['SecuCode', 'SecuAbbr', 'PubDatetime1', 'InsertDatetime1', 'Title1'])
update_sql = '''
select A.* from juchao_kuaixun A, juchao_ant B where B.UPDATETIMEJZ > '{}' and A.code = B.SecuCode \
and A.link = B.AntDoc and A.type = '公告'; '''.format(deadline)
datas = self._r_spider_conn.query(update_sql)
for data in datas:
item = {
'PubDatetime2': data.get("pub_date"),
'InsertDatetime2': data.get("CREATETIMEJZ"),
'Title2': data.get("title"),
}
self._spider_conn.table_update(self.merge_table_name, item, 'PDFLink', data.get("link"))
# res2 = juyuan_conn.query(sql2)
# map2 = {}
# for one in res2:
# map2[one['SecuCode']] = one['InnerCode']
#
#
# res3 = juyuan_conn.query(sql3)
# map3 = {}
# for one in res3:
# map3[one['SecuCode']] = one['InnerCode']
#
# delta12 = set(map1.keys()) - set(map2.keys())
# # print(delta12)
#
# delta21 = set(map2.keys() - map1.keys())
# # print(delta21)
#
# delta23 = set(map2.keys() - map3.keys())
# print(delta23)
#
# delta32 = set(map3.keys() - map2.keys())
# print(delta32)
res4 = juyuan_conn.query(sql4)
map4 = {}
for one in res4:
map4[one['SecuCode']] = one['InnerCode']
if not one['SecuCode'][0] in ['0', '3', '6']:
print(one)
class JuchaoCounter(object):
def __init__(self):
self.api = 'http://www.cninfo.com.cn/new/hisAnnouncement/query'
self.headers = {
'Accept': '*/*',
'Accept-Encoding': 'gzip, deflate',
'Accept-Language': 'zh-CN,zh;q=0.9,en;q=0.8',
'Cache-Control': 'no-cache',
'Connection': 'keep-alive',
'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
'Host': 'www.cninfo.com.cn',
'Origin': 'http://www.cninfo.com.cn',
'Pragma': 'no-cache',
'Referer': 'http://www.cninfo.com.cn/new/commonUrl/pageOfSearch?url=disclosure/list/search&lastPage=index',
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.125 Safari/537.36',
}
self._spider_conn = Connection(
host=SPIDER_MYSQL_HOST,
port=SPIDER_MYSQL_PORT,
user=SPIDER_MYSQL_USER,
password=SPIDER_MYSQL_PASSWORD,
database=SPIDER_MYSQL_DB,
)
self._r_spider_conn = Connection(
host=R_SPIDER_MYSQL_HOST,
port=R_SPIDER_MYSQL_PORT,
user=R_SPIDER_MYSQL_USER,
password=R_SPIDER_MYSQL_PASSWORD,
database=R_SPIDER_MYSQL_DB,
)
@property
def codes_map(self):
codes_map = {}
sql = '''select code, OrgId from juchao_codemap; '''
res = self._spider_conn.query(sql)
for r in res:
codes_map[r.get('OrgId')] = r.get("code")
return codes_map
def launch(self, org_id: str):
codes_map = self.codes_map
org_id_lst = sorted(list(codes_map.keys()))
position = org_id_lst.index(org_id)
print("position", position)
for org_id in org_id_lst[position:]:
code = codes_map.get(org_id)
stock_str = ','.join([code, org_id])
print(stock_str)
self.get_count(stock_str)
@retry(stop_max_attempt_number=3)
def get_count(self, stock_str: str):
time.sleep(random.randint(1, 3)/10)
post_data = {
'pageNum': 1,
'pageSize': 30,
'column': 'szse',
'tabName': 'fulltext',
'plate': '',
'stock': stock_str,
'searchkey': '',
'secid': '',
'category': '',
'trade': '',
'seDate': '',
'sortName': '',
'sortType': '',
'isHLtitle': True,
}
resp = requests.post(self.api, headers=self.headers, data=post_data, timeout=3)
if resp.status_code == 200:
text = resp.text
py_datas = json.loads(text)
total_ann = py_datas.get("totalAnnouncement")
total_rec = py_datas.get("totalRecordNum")
record = {
"SecuCode": stock_str.split(',')[0],
"TotalAnn": total_ann,
"TotalRec": total_rec,
}
# print(record)
# 记录单个 code 的结束
with open(file_name, "a") as f:
f.write(json.dumps(record)+'\n')
def check_count_bydate(self):
"""
根据时间去计算每天的个数
"""
sql = '''select SecuCode, count(*) from juchao_ant2 group by AntTime ; '''
def check_count(self):
sql = '''select SecuCode, count(*) from juchao_ant2 group by SecuCode ; '''
ret = self._r_spider_conn.query(sql)
exist_map = {}
for r in ret:
exist_map[r.get('SecuCode')] = r.get("count(*)")
# print(exist_map)
web_map = {}
with open(file_name, "r") as f:
lines = f.readlines()
for line in lines:
r = json.loads(line)
web_map[r.get("SecuCode")] = r.get("TotalAnn")
no_lst = []
big_delta_lst = []
small_delta_lst = []
for code in web_map:
_count = 0
if code in _map:
_code = _map.get(code)
_count = exist_map.get(_code)
exist_num = exist_map.get(code)
exist_num += _count
web_num = web_map.get(code)
if not exist_num:
no_lst.append(code)
elif exist_num != web_num:
delta = web_num - exist_num
if delta > 0:
big_delta_lst.append((code, delta))
# big_delta_lst.append(code)
else:
# small_delta_lst.append((code, delta))
small_delta_lst.append(code)
# print(no_lst)
# print(len(no_lst))
print(big_delta_lst)
print(len(big_delta_lst))
class UpdateInnerCode(object):
def __init__(self):
self._spider_conn = Connection(
host=SPIDER_MYSQL_HOST,
port=SPIDER_MYSQL_PORT,
user=SPIDER_MYSQL_USER,
password=SPIDER_MYSQL_PASSWORD,
database=SPIDER_MYSQL_DB,
)
self._juyuan_conn = Connection(
host=JUY_HOST,
port=JUY_PORT,
user=JUY_USER,
password=JUY_PASSWD,
database=JUY_DB,
)
self.batch_num = 10000
def max_id(self):
sql = '''select max(id) as max_id from announcement_base2; '''
return self._spider_conn.get(sql).get("max_id")
def load_innercode_map(self):
sql = '''select secucode, innercode from secumain; '''
_map = dict()
ret = self._juyuan_conn.query(sql)
for r in ret:
_map[r.get("secucode")] = r.get("innercode")
return _map
def get_old_inner_code(self, secucode: str):
sql = '''select InnerCode from LC_CodeChange where secucode = '{}';'''.format(
secucode)
r = self._juyuan_conn.get(sql)
if r:
inner_code = r.get("InnerCode")
return inner_code
def start(self, begin: int, end: int):
inner_map = self.load_innercode_map() # secucode 无前缀
# max_id = self.max_id()
# print(max_id)
base_sql = '''select id, secucode from announcement_base2 where id between {} and {} and InnerCode = 0; '''
for i in range(begin, end):
sql = base_sql.format(i * self.batch_num,
i * self.batch_num + self.batch_num)
print(sql)
datas = self._spider_conn.query(sql)
print(len(datas))
for data in datas:
inner_code = inner_map.get(data.get('secucode')[2:])
if inner_code is None:
inner_code = self.get_old_inner_code(
data.get('secucode')[2:])
if inner_code is not None:
inner_map.update(
{data.get('secucode')[2:]: inner_code})
self._spider_conn.table_update(
'announcement_base2', {'InnerCode': inner_code},
"id", data.get("id"))
else:
self._spider_conn.table_update('announcement_base2',
{'InnerCode': inner_code},
"id", data.get("id"))
class SourceAnnouncementBase(object):
"""将两个爬虫表合并生成公告基础表"""
def __init__(self):
self.merge_table_name = 'announcement_base2'
self.his_table = 'juchao_ant2'
self.live_table = 'juchao_kuaixun'
self.batch_number = 10000
self._r_spider_conn = Connection(
host=R_SPIDER_MYSQL_HOST,
port=R_SPIDER_MYSQL_PORT,
user=R_SPIDER_MYSQL_USER,
password=R_SPIDER_MYSQL_PASSWORD,
database=R_SPIDER_MYSQL_DB,
)
self._spider_conn = Connection(
host=SPIDER_MYSQL_HOST,
port=SPIDER_MYSQL_PORT,
user=SPIDER_MYSQL_USER,
password=SPIDER_MYSQL_PASSWORD,
database=SPIDER_MYSQL_DB,
)
self._juyuan_conn = Connection(
host=JUY_HOST,
port=JUY_PORT,
user=JUY_USER,
password=JUY_PASSWD,
database=JUY_DB,
)
def category_code_map(self):
return {
'category_bcgz_szsh': ('补充更正', 19),
'category_bndbg_szsh': ('半年报', 2),
'category_cqdq_szsh': ('澄清致歉', 12),
'category_dshgg_szsh': ('董事会', 18),
'category_fxts_szsh': ('风险提示', 21),
'category_gddh_szsh': ('股东大会', 15),
'category_gqbd_szsh': ('股权变动', 16),
'category_gqjl_szsh': ('股权激励', 17),
'category_gszl_szsh': ('公司治理', 24),
'category_gszq_szsh': ('公司债', 25),
'category_jj_szsh': ('解禁', 9),
'category_jshgg_szsh': ('监事会', 14),
'category_kzzq_szsh': ('可转债', 22),
'category_ndbg_szsh': ('年报', 4),
'category_pg_szsh': ('配股', 7),
'category_qtrz_szsh': ('其他融资', 23),
'category_qyfpxzcs_szsh': ('权益分派', 11),
'category_rcjy_szsh': ('日常经营', 10),
'category_sf_szsh': ('首发', 8),
'category_sjdbg_szsh': ('三季报', 3),
'category_tbclts_szsh': ('特别处理和退市', 13),
'category_tszlq_szsh': ('退市整理期', 20),
'category_yjdbg_szsh': ('一季报', 1),
'category_yjygjxz_szsh': ('业绩预告', 5),
'category_zf_szsh': ('增发', 6),
'category_zj_szsh': ('中介报告', 26),
'category_others': ('其他', 27),
}
def spider_max_id(self, table_name: str):
sql = f'''select max(id) as max_id from {table_name}; '''
max_id = self._r_spider_conn.get(sql).get("max_id")
return max_id
def check_ids(self):
_delta = []
max_id = self.spider_max_id(self.his_table)
for i in range(int(max_id / self.batch_number) + 1):
begin_id = i * self.batch_number
end_id = i * self.batch_number + self.batch_number
print()
print(begin_id, end_id)
sql1 = '''select count(id) from {} where id between {} and {}; '''.format(
self.his_table, begin_id, end_id)
sql2 = '''select count(id) from {} where id between {} and {}; '''.format(
self.merge_table_name, begin_id, end_id)
count1 = self._r_spider_conn.get(sql1).get("count(id)")
count2 = self._r_spider_conn.get(sql2).get("count(id)")
print(count1, count2)
if count1 != count2:
sl1 = '''select id from {} where id between {} and {};'''.format(
self.his_table, begin_id, end_id)
sl2 = '''select id from {} where id between {} and {};'''.format(
self.merge_table_name, begin_id, end_id)
rs1 = self._r_spider_conn.query(sl1)
ids1 = set([r.get("id") for r in rs1])
rs2 = self._r_spider_conn.query(sl2)
ids2 = set([r.get("id") for r in rs2])
print(list(ids1 - ids2))
_delta.extend(list(ids1 - ids2))
print(_delta)
return _delta
def process_diff_ids(self, diff_lst):
diff_lst = tuple(diff_lst)
category_code_map = self.category_code_map()
_sql = '''select id, SecuCode, CategoryCode, SecuAbbr, AntTime as PubDatetime1, \
AntTitle as Title1, AntDoc as PDFLink, CREATETIMEJZ as InsertDatetime1 from {} where id in {}; '''.format(
self.his_table, diff_lst)
datas = self._r_spider_conn.query(_sql)
print(len(datas))
for data in datas:
data['CategoryCode'] = category_code_map.get(
data.get("CategoryCode"))[1]
data = utils.process_secucode(data) # 暂时只要 0 3 6
if data:
print(data)
ret = self._spider_conn.table_insert(
self.merge_table_name, data, [
'SecuCode', 'SecuAbbr', 'CategoryCode', 'PubDatetime1',
'InsertDatetime1', 'Title1'
])
print(ret)
def first_load(self):
inner_map = self.load_innercode_map()
category_code_map = self.category_code_map()
load_sql = '''select id, SecuCode, CategoryCode, SecuAbbr, AntTime as PubDatetime1, \
AntTitle as Title1, AntDoc as PDFLink, CREATETIMEJZ as InsertDatetime1 from {} where \
id >= {} and id < {}; '''
max_id = self.spider_max_id(self.his_table)
for i in range(int(max_id / self.batch_number) + 1):
begin_id = i * self.batch_number
end_id = i * self.batch_number + self.batch_number
print(begin_id, end_id)
_sql = load_sql.format(self.his_table, begin_id, end_id)
print(_sql)
datas = self._r_spider_conn.query(_sql)
for data in datas:
inner_code = inner_map.get(data.get("SecuCode"))
if inner_code:
data['InnerCode'] = inner_code # 爬虫库中的 secucode 无前缀
else:
inner_code = self.get_old_inner_code(data.get("SecuCode"))
if inner_code:
data['InnerCode'] = inner_code
inner_map.update({data.get("SecuCode"): inner_code})
else:
continue
data['CategoryCode'] = category_code_map.get(
data.get("CategoryCode"))[1]
data = utils.process_secucode(data) # 暂时只要 0 3 6
if data:
self._spider_conn.table_insert(
self.merge_table_name, data, [
'SecuCode', 'SecuAbbr', 'CategoryCode',
'PubDatetime1', 'InsertDatetime1', 'Title1',
'InnerCode'
])
# 将巨潮快讯的数据补充进去
update_sql = '''select A.* from {} A, {} B \
where A.code = B.SecuCode and A.link = B.AntDoc and A.type = '公告' \
and A.id between {} and {}; '''
max_id = self.spider_max_id(self.live_table)
for j in range(int(max_id / self.batch_number) + 1):
begin_id = j * self.batch_number
end_id = j * self.batch_number + self.batch_number
print(begin_id, end_id)
_sql = update_sql.format(self.live_table, self.his_table, begin_id,
end_id)
print(_sql)
datas = self._r_spider_conn.query(_sql)
for data in datas:
item = {
'PubDatetime2': data.get("pub_date"),
'InsertDatetime2': data.get("CREATETIMEJZ"),
'Title2': data.get("title"),
}
self._spider_conn.table_update(self.merge_table_name, item,
'PDFLink', data.get("link"))
def load_innercode_map(self):
sql = '''select secucode, innercode from secumain; '''
_map = dict()
ret = self._juyuan_conn.query(sql)
for r in ret:
_map[r.get("secucode")] = r.get("innercode")
return _map
def get_old_inner_code(self, secucode: str):
sql = '''select InnerCode from LC_CodeChange where secucode = '{}';'''.format(
secucode)
r = self._juyuan_conn.get(sql)
if r:
inner_code = r.get("InnerCode")
return inner_code
class SourceAnnouncementBase(object):
"""将两个爬虫表合并生成公告基础表"""
def __init__(self):
self.merge_table_name = 'announcement_base2'
self.his_table = 'juchao_ant2'
self.live_table = 'juchao_kuaixun'
self.batch_number = 10000
self._r_spider_conn = Connection(
host=R_SPIDER_MYSQL_HOST,
port=R_SPIDER_MYSQL_PORT,
user=R_SPIDER_MYSQL_USER,
password=R_SPIDER_MYSQL_PASSWORD,
database=R_SPIDER_MYSQL_DB,
)
self._spider_conn = Connection(
host=SPIDER_MYSQL_HOST,
port=SPIDER_MYSQL_PORT,
user=SPIDER_MYSQL_USER,
password=SPIDER_MYSQL_PASSWORD,
database=SPIDER_MYSQL_DB,
)
self._juyuan_conn = Connection(
host=JUY_HOST,
port=JUY_PORT,
user=JUY_USER,
password=JUY_PASSWD,
database=JUY_DB,
)
def category_code_map(self):
return {
'category_bcgz_szsh': ('补充更正', 19),
'category_bndbg_szsh': ('半年报', 2),
'category_cqdq_szsh': ('澄清致歉', 12),
'category_dshgg_szsh': ('董事会', 18),
'category_fxts_szsh': ('风险提示', 21),
'category_gddh_szsh': ('股东大会', 15),
'category_gqbd_szsh': ('股权变动', 16),
'category_gqjl_szsh': ('股权激励', 17),
'category_gszl_szsh': ('公司治理', 24),
'category_gszq_szsh': ('公司债', 25),
'category_jj_szsh': ('解禁', 9),
'category_jshgg_szsh': ('监事会', 14),
'category_kzzq_szsh': ('可转债', 22),
'category_ndbg_szsh': ('年报', 4),
'category_pg_szsh': ('配股', 7),
'category_qtrz_szsh': ('其他融资', 23),
'category_qyfpxzcs_szsh': ('权益分派', 11),
'category_rcjy_szsh': ('日常经营', 10),
'category_sf_szsh': ('首发', 8),
'category_sjdbg_szsh': ('三季报', 3),
'category_tbclts_szsh': ('特别处理和退市', 13),
'category_tszlq_szsh': ('退市整理期', 20),
'category_yjdbg_szsh': ('一季报', 1),
'category_yjygjxz_szsh': ('业绩预告', 5),
'category_zf_szsh': ('增发', 6),
'category_zj_szsh': ('中介报告', 26),
'category_others': ('其他', 27),
}
def load_innercode_map(self):
sql = '''select secucode, innercode from secumain; '''
_map = dict()
ret = self._juyuan_conn.query(sql)
for r in ret:
_map[r.get("secucode")] = r.get("innercode")
return _map
def get_old_inner_code(self, secucode: str):
sql = '''select InnerCode from LC_CodeChange where secucode = '{}';'''.format(secucode)
r = self._juyuan_conn.get(sql)
if r:
inner_code = r.get("InnerCode")
return inner_code
def daily_update(self, deadline: datetime.datetime = None):
inner_map = self.load_innercode_map()
category_code_map = self.category_code_map()
if deadline is None:
deadline = datetime.datetime.now() - datetime.timedelta(days=1)
load_sql = '''select id, SecuCode, CategoryCode, SecuAbbr, AntTime as PubDatetime1, \
AntTitle as Title1, AntDoc as PDFLink, CREATETIMEJZ as InsertDatetime1 from {} where \
UPDATETIMEJZ > '{}'; '''.format(self.his_table, deadline)
logger.info(f"his load sql is {load_sql}")
datas = self._r_spider_conn.query(load_sql)
logger.info(f"load count is {len(datas)} from his table.")
items = []
for data in datas:
data['CategoryCode'] = category_code_map.get(data.get("CategoryCode"))[1]
inner_code = inner_map.get(data.get("SecuCode"))
if inner_code:
data['InnerCode'] = inner_code # 爬虫库中的 secucode 无前缀
else:
inner_code = self.get_old_inner_code(data.get("SecuCode"))
if inner_code:
inner_map.update({data.get("SecuCode"): inner_code})
data['InnerCode'] = inner_code
else:
continue
data = utils.process_secucode(data) # 暂时只要 0 3 6
if data:
items.append(data)
self._spider_conn.batch_insert(items, self.merge_table_name,
['SecuCode', 'SecuAbbr', 'CategoryCode', 'PubDatetime1', 'InsertDatetime1', 'Title1', 'InnerCode'])
update_sql = '''select A.* from juchao_kuaixun A, juchao_ant B where B.UPDATETIMEJZ > '{}' \
and A.code = B.SecuCode and A.link = B.AntDoc and A.type = '公告'; '''.format(deadline)
logger.info(f"live sql is {update_sql}")
datas = self._r_spider_conn.query(update_sql)
logger.info(f'load count {len(datas)} from live table.')
for data in datas:
item = {
'PubDatetime2': data.get("pub_date"),
'InsertDatetime2': data.get("CREATETIMEJZ"),
'Title2': data.get("title"),
}
self._spider_conn.table_update(self.merge_table_name, item, 'PDFLink', data.get("link"))