在一些 csv 文件中会有日期类型，而如果我们想按日期类型导入到数据库中，那么就需要在 insert 前以 date(datetime)的类型传入。
假设简单的 csv 文件(sample.csv)如下：
HolidayDate,Region
2019-01-01,China
2019-01-02,China
2019-04-01,China
2019-05-01,China
2019-07-07,China
2019-10-01,China
2019-01-05,US
2019-06-01,US
直接使用 cvs 库
读取到 csv 后，直接遍历完成：
import csv
from datetime import datetime
def read_csv_rows_list(file_path):
csv_row_list = []
with open(file_path, newline='') as csvfile:
csvReader = csv.reader(csvfile, delimiter=',', quotechar='|')
for row in csvReader:
csv_row_list.append(row)
return csv_row_list
csv_row_list = read_csv_rows_list('sample.csv')
csv_row_list = csv_row_list[1:]
for row in csv_row_list:
row[0] = datetime.strptime(row[0], '%Y-%m-%d')
[print(row) for row in csv_row_list]
output:
[datetime.datetime(2019, 1, 1, 0, 0), 'China']
[datetime.datetime(2019, 1, 2, 0, 0), 'China']
[datetime.datetime(2019, 4, 1, 0, 0), 'China']
[datetime.datetime(2019, 5, 1, 0, 0), 'China']
[datetime.datetime(2019, 7, 7, 0, 0), 'China']
[datetime.datetime(2019, 10, 1, 0, 0), 'China']
[datetime.datetime(2019, 1, 5, 0, 0), 'US']
[datetime.datetime(2019, 6, 1, 0, 0), 'US']
使用 pandas
如果使用 pandas 就很方便：
import pandas as pd
import numpy as np
data = pd.read_csv('sample.csv',encoding='utf-8',)
data[u'HolidayDate'] = pd.to_datetime(data[u'HolidayDate'])
# data[u'HolidayDate'].astype(str)
# data[u'HolidayDate'] = data[u'HolidayDate'].apply(lambda x :datetime.strptime(x, '%Y-%m-%d'))
data = np.array(data)#np.ndarray()
data_list = data.tolist()
[print(row) for row in data_list]
#data.to_csv('sample.output.csv',index=False, encoding='utf-8')
[Timestamp('2019-01-01 00:00:00'), 'China']
[Timestamp('2019-01-02 00:00:00'), 'China']
[Timestamp('2019-04-01 00:00:00'), 'China']
[Timestamp('2019-05-01 00:00:00'), 'China']
[Timestamp('2019-07-07 00:00:00'), 'China']
[Timestamp('2019-10-01 00:00:00'), 'China']
[Timestamp('2019-01-05 00:00:00'), 'US']
[Timestamp('2019-06-01 00:00:00'), 'US']
参考：
近期评论