【getHowWordPageData.py】提取热词和相关函数定义
This commit is contained in:
+17
-1
@@ -1,4 +1,7 @@
|
||||
from utils.query import query
|
||||
import re
|
||||
import sys
|
||||
import pandas as pd
|
||||
sys.path.append('model')
|
||||
|
||||
cityList = [
|
||||
@@ -72,4 +75,17 @@ def getAllCommentsData():
|
||||
|
||||
def getAllArticleData():
|
||||
articleList = query('select * from article',[],'select')
|
||||
return articleList
|
||||
return articleList
|
||||
|
||||
def getAllHotWords():
|
||||
data = []
|
||||
df = pd.read_csv('./model/cipingTotal.csv',encoding='utf8')
|
||||
for i in df.values:
|
||||
try:
|
||||
data.append([
|
||||
re.search('[\u4e00-\u9fa5]+',str(i)).group(),
|
||||
re.search('\d+',str(i)).group()
|
||||
])
|
||||
except:
|
||||
continue
|
||||
return data
|
||||
Reference in New Issue
Block a user