Skip to content
Toggle navigation
P
Projects
G
Groups
S
Snippets
Help
lichenggang
/
update_cate
This project
Loading...
Sign in
Toggle navigation
Go to a project
Project
Repository
Issues
0
Merge Requests
0
Pipelines
Wiki
Snippets
Settings
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Commit
9157c712
authored
May 07, 2019
by
lichenggang
Browse files
Options
_('Browse Files')
Download
Email Patches
Plain Diff
add eb80
parent
19bf894c
Show whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
60 additions
and
2 deletions
plat/eb80/cate_stock.py
utils/mysqlopera.py
plat/eb80/cate_stock.py
View file @
9157c712
from
utils.base
import
Module_Base
import
re
dom_pat
=
re
.
compile
(
r'\<script\>document\.write\(\"(.+)\"\)'
)
domain
=
'http:'
headers
=
{
"User-Agent"
:
"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.121 Safari/537.36"
,
}
a
=
Module_Base
()
url
=
'http://dianzi.eb80.com/gongying/'
resp
=
a
.
requests
.
get
(
url
,
headers
=
headers
)
.
text
dom
=
a
.
etree
.
HTML
(
resp
)
level1_doms1
=
dom
.
xpath
(
'//div[@class="trade"]//li[position()<25]'
)
s
=
dom_pat
.
findall
(
resp
)
level1_doms2
=
[
a
.
etree
.
HTML
(
i
)
for
i
in
s
[:
24
]]
level1_doms
=
zip
(
level1_doms1
,
level1_doms2
)
other_doms
=
dom
.
xpath
(
'//div[@class="trade"]//li[position()>=25]'
)
def
get_levels
():
levels
=
[]
for
level1_dom
in
level1_doms
:
level1
=
dict
()
level1
[
'cat_name'
]
=
level1_dom
[
0
]
.
xpath
(
'.//strong//a[1]/text()'
)[
0
]
.
strip
()
level1
[
'url'
]
=
level1_dom
[
0
]
.
xpath
(
'.//strong//a[1]/@href'
)[
0
]
level1
[
'level'
]
=
1
level1
[
'islast'
]
=
0
level1
[
'parent_id'
]
=
0
level1
[
'level2_list'
]
=
[]
# print('一级分类',level1)
for
level2_dom
in
level1_dom
[
1
]
.
xpath
(
'.//a'
):
level2
=
{}
level2
[
'cat_name'
]
=
level2_dom
.
xpath
(
'./text()'
)[
0
]
.
strip
()
level2
[
'url'
]
=
level2_dom
.
xpath
(
'./@href'
)[
0
]
level2
[
'level'
]
=
2
level2
[
'islast'
]
=
1
level1
[
'level2_list'
]
.
append
(
level2
)
levels
.
append
(
level1
)
other_level1
=
dict
()
other_level1
[
'cat_name'
]
=
'其他'
other_level1
[
'url'
]
=
''
other_level1
[
'level'
]
=
1
other_level1
[
'islast'
]
=
0
other_level1
[
'parent_id'
]
=
0
other_level1
[
'level2_list'
]
=
[]
for
level2_dom
in
other_doms
:
level2
=
{}
level2
[
'cat_name'
]
=
level2_dom
.
xpath
(
'.//strong//a[1]/text()'
)[
0
]
.
strip
()
level2
[
'url'
]
=
level2_dom
.
xpath
(
'.//strong//a[1]/@href'
)[
0
]
level2
[
'level'
]
=
2
level2
[
'islast'
]
=
1
other_level1
[
'level2_list'
]
.
append
(
level2
)
levels
.
append
(
other_level1
)
return
levels
levels
=
get_levels
()
print
(
levels
)
utils/mysqlopera.py
View file @
9157c712
...
...
@@ -5,7 +5,7 @@ import time
import
pymysql
COMPANY_PN_LIST
=
[
'hc360'
,
'huangye88'
,
'pangwo'
,
'coovee'
,
'liebiao'
,
'china'
,
'ic114'
,
'cecb2b'
,
'trustexporter'
,
'x11467'
,
'x912688'
,
'zgtghccl'
,
'herostart'
,
'testmart'
,
'cc'
,
'eastsoo'
,
'x51sole'
]
COMPANY_PN_LIST
=
[
'hc360'
,
'huangye88'
,
'pangwo'
,
'coovee'
,
'liebiao'
,
'china'
,
'ic114'
,
'cecb2b'
,
'trustexporter'
,
'x11467'
,
'x912688'
,
'zgtghccl'
,
'herostart'
,
'testmart'
,
'cc'
,
'eastsoo'
,
'x51sole'
,
'eb80'
]
def
get_env
():
...
...
@@ -18,7 +18,7 @@ def get_env():
ENV
=
get_env
()
HOST_SET
=
{
'test'
:
'192.168.1.23
2
'
,
'test'
:
'192.168.1.23
4
'
,
'test234'
:
'192.168.1.234'
,
'produce'
:
'172.18.137.37'
}
...
...
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