如何将 csv 文件中的数据保存到数组？

Question

I am trying to read a csv file using the module csv. Here is my code :

import csv

with open('test.csv') as csvfile:
    spamreader = csv.reader(csvfile, delimiter='|')
    daymoment = [row[0] for row in csv.reader(csvfile, delimiter='|') if row != ''][1::]
    day = [row[0] for row in spamreader]
print(daymoment)
print(day)

Here is my csv file :

|Monday|Tuesday|Wednesday|Thursday|Friday|Saturday|Sunday
Morning|2|1|1|3|5|6|7
Afternoon|3|4|4|7|8|6|5
Evening|5|23|5|6|8|7|9
Night|6|1|4|2|6|4|7

But I have the following problem : there are no values in day . I want, for example, ['Monday', '2', '3', '5', '6']

Could you help me please ?

Thank you !

Answer 1

For your CSV file, I fixed a typo on "Afternoon" and added a new column title based on your code, 'daymoment'.

Daymoment|Monday|Tuesday|Wednesday|Thursday|Friday|Saturday|Sunday
Morning|2|1|1|3|5|6|7
Afternoon|3|4|4|7|8|6|5
Evening|5|23|5|6|8|7|9
Night|6|1|4|2|6|4|7

You can read in the csv file, convert it to a pandas dataframe, and read the 'monday' column to get the data out like this:

import csv
import pandasql as ps
import pandas as pd

with open('test.csv') as csvfile:
    spamreader = csv.reader(csvfile, delimiter='|')
    mydata = pd.DataFrame([x for x in spamreader])
    mydata.columns = mydata.iloc[0]
    mydata = mydata[1:]

print(mydata['Monday'])  
1    2
2    3
3    5
4    6  

If I'm understanding correctly, you may want to select individual days from your data and grab their values in a different structure. This would require data transformation using sql (imported above), like so:

transform_sql = """
select
Daymoment,
'Monday' as Day,
Monday as Daycount
from mydata
union all
select
Daymoment,
'Tuesday' as Day,
Tuesday as Daycount
from mydata
union all
select
Daymoment,
'Wednesday' as Day,
Wednesday as Daycount
from mydata
union all
select
Daymoment,
'Thursday' as Day,
Thursday as Daycount
from mydata
union all
select
Daymoment,
'Friday' as Day,
Friday as Daycount
from mydata
union all
select
Daymoment,
'Saturday' as Day,
Saturday as Daycount
from mydata
union all
select
Daymoment,
'Sunday' as Day,
Sunday as Daycount
from mydata
"""

print(ps.sqldf(transform_sql))  
    Daymoment        Day Daycount
0     Morning     Monday        2
1   Afternoon     Monday        3
2     Evening     Monday        5
3       Night     Monday        6
4     Morning    Tuesday        1
5   Afternoon    Tuesday        4
6     Evening    Tuesday       23
7       Night    Tuesday        1
8     Morning  Wednesday        1
9   Afternoon  Wednesday        4
10    Evening  Wednesday        5
11      Night  Wednesday        4
12    Morning   Thursday        3
13  Afternoon   Thursday        7
14    Evening   Thursday        6
15      Night   Thursday        2
16    Morning     Friday        5
17  Afternoon     Friday        8
18    Evening     Friday        8
19      Night     Friday        6
20    Morning   Saturday        6
21  Afternoon   Saturday        6
22    Evening   Saturday        7
23      Night   Saturday        4
24    Morning     Sunday        7
25  Afternoon     Sunday        5
26    Evening     Sunday        9
27      Night     Sunday        7  

Now you can simply select the "day" you would like from your transformed data structure.

print(transposed_data.loc[transposed_data['Day'] == 'Monday'])  
   Daymoment     Day Daycount
0    Morning  Monday        2
1  Afternoon  Monday        3
2    Evening  Monday        5
3      Night  Monday        6

Answer 2

A csv reader reads the file line by line. When the end of file is reached, the reader is exhausted and will not return anything more. If you want to process it more than once, you have to keep the rows in a data structure, for example a list:

with io.StringIO(t) as csvfile:
    spamreader = list(csv.reader(csvfile, delimiter='|'))
    daymoment = [row[0] for row in spamreader if row[0] != ''][1::]
    day = [row[1] for row in spamreader]

print(day)
print(daymoment)

gives:

['Monday', '2', '3', '5', '6']
['Afternoon', 'Evening', 'Night']