Skip to content
Projects
Groups
Snippets
Help
Loading...
Sign in
Toggle navigation
C
crawler
Project
Project
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
Chengyang Zhong
crawler
Commits
53c7af78
Commit
53c7af78
authored
Aug 04, 2020
by
litaolemo
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
update
parent
98d8a309
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
73 additions
and
4 deletions
+73
-4
cal_ni_and_put_to_backend.py
crawler_sys/scheduler/cal_ni_and_put_to_backend.py
+73
-4
No files found.
crawler_sys/scheduler/cal_ni_and_put_to_backend.py
View file @
53c7af78
...
...
@@ -8,6 +8,9 @@ import json
import
datetime
import
time
import
sys
import
six
from
maintenance.func_send_email_with_file
import
send_file_email
from
typing
import
Dict
,
List
from
crawler.gm_upload.gm_upload
import
upload
,
upload_file
...
...
@@ -318,11 +321,15 @@ def write_data_into_mysql(res_data):
if
not
data
.
get
(
"qiniu_img_list"
):
return
None
tractate_id
=
0
if
data
[
"platform"
]
==
"weibo"
:
platform_value
=
11
elif
data
[
"platform"
]
==
"douban"
:
platform_value
=
12
try
:
sql_query
=
"""insert into api_tractate
(user_id,content,is_online,status,platform,content_level,is_excellent,create_time,last_modified,user_del,low_quality,low_quality_deal,platform_id,pgc_type,title)
values ({user_id},'{content}',{is_online},{status},{platform},{content_level},{is_excellent},'{create_time}','{last_modified}',{user_del},{low_quality},{low_quality_deal},'{platform_id}',{pgc_type},'{title}');"""
.
format
(
user_id
=
random
.
choice
(
user_id_list
),
content
=
data
[
"content"
],
is_online
=
0
,
status
=
2
,
platform
=
15
,
user_id
=
random
.
choice
(
user_id_list
),
content
=
data
[
"content"
],
is_online
=
0
,
status
=
2
,
platform
=
platform_value
,
content_level
=
data
[
"level"
],
is_excellent
=
0
,
create_time
=
now_str
,
last_modified
=
now_str
,
user_del
=
0
,
...
...
@@ -380,6 +387,68 @@ def write_data_into_mysql(res_data):
return
None
class
EnumMeta
(
type
):
"""Metaclass for Enum"""
@staticmethod
def
_find_allow_types_
(
cls
,
bases
):
all_types
=
set
(
six
.
integer_types
)
|
{
six
.
text_type
,
str
}
allow_types
=
set
()
if
Enum
is
None
:
# Enum base class
assert
cls
==
'Enum'
return
tuple
(
all_types
)
else
:
for
base
in
bases
:
if
not
issubclass
(
base
,
Enum
):
allow_types
.
add
(
base
)
if
allow_types
:
return
tuple
(
all_types
&
allow_types
)
else
:
return
tuple
(
all_types
)
class
Enum
(
six
.
with_metaclass
(
EnumMeta
,
object
)):
"""Generic enumeration.
Derive from this class to define new enumerations.
"""
def
__repr__
(
self
):
return
"<
%
s.
%
s:
%
r>"
%
(
self
.
__class__
.
__name__
,
self
.
_name_
,
self
.
_value_
)
def
__str__
(
self
):
if
self
.
_desc_
:
return
"
%
s.
%
s(
%
s)"
%
(
self
.
__class__
.
__name__
,
self
.
_name_
,
self
.
_desc_
)
else
:
return
"
%
s.
%
s"
%
(
self
.
__class__
.
__name__
,
self
.
_name_
)
def
__hash__
(
self
):
return
hash
(
self
.
_name_
)
class
TRACTATE_PLATFORM
(
Enum
):
"""
新帖子发布来源
"""
GM
=
(
"1"
,
u"更美"
)
HERA
=
(
"2"
,
u"HERA后台"
)
DOCTOR
=
(
"3"
,
u"医生端"
)
XIAOHONGSHU
=
(
"4"
,
u"小红书"
)
WEIBO
=
(
"5"
,
u"微博"
)
SOYOUNG
=
(
"6"
,
u"新氧"
)
MARK
=
(
"7"
,
u"站内打卡活动"
)
VARIETY_SHOW_YOUNG
=
(
"8"
,
"选秀节目(少年之名)打榜活动"
)
GROUP_DETAIL
=
(
"9"
,
"普通小组"
)
GROUP_TOPIC_DETAIL
=
(
"10"
,
"普通小组话题"
)
STRATEGY_WEIBO_HOTSPOT
=
(
"11"
,
"策略微博热点"
)
STRATEGY_DOUBAN_HOTSPOT
=
(
"12"
,
"策略豆瓣鹅组热点"
)
STRATEGY_TOUTIAO
=
(
"13"
,
"策略头条文章"
)
STRATEGY_ZHIHU
=
(
"14"
,
"策略知乎文章"
)
STRATEGY_XIAOHONGSHU
=
(
"15"
,
"策略小红书文章"
)
STRATEGY_SOYOUNG
=
(
"16"
,
"策略新氧文章"
)
STRATEGY_WEIBO
=
(
"17"
,
"策略微博文章"
)
def
task_main
():
# 实例化数据判断规则 注意高优先级在前 低优先级在后
push_rule_class1
=
push_rule
(
comment_count_ni
=
20
,
time_range
=
5
,
level
=
3
)
...
...
@@ -394,15 +463,15 @@ def task_main():
tractate_id
=
write_data_into_mysql
(
res_data
)
try
:
print
(
"line394"
,
tractate_id
,
res_data
)
if
res_data
[
"level"
]
>=
1
and
tractate_id
:
if
res_data
[
"level"
]
>=
3
and
tractate_id
:
title_str
=
res_data
[
"platform"
]
+
"帖子内容审核"
body_str
=
"""
问好:
有一篇新的内容需要审核,帖子号为{tractate_id}
有一篇新的
{level}星
内容需要审核,帖子号为{tractate_id}
内容如下:
{content}
"""
.
format
(
tractate_id
=
tractate_id
,
content
=
res_data
[
"content"
])
"""
.
format
(
tractate_id
=
tractate_id
,
content
=
res_data
[
"content"
]
,
level
=
res_data
[
"level"
]
)
send_file_email
(
""
,
""
,
email_group
=
[
"<hongxu@igengmei@igengmei.com>"
,
"<yangjiayue@igengmei.com>"
,
"<zhangweiwei@igengmei.com>"
,
"<liuyiting@igengmei.com>"
],
cc_group
=
[
"<duanyingrong@igengmei.com>"
,
"<litao@igengmei.com>"
],
email_msg_body_str
=
body_str
,
title_str
=
title_str
)
print
(
"send to mysql"
)
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment