python CSV writer - formatting

Question

Quick question on how to properly write data back into a CSV file using the python csv module. Currently i'm importing a file, pulling a column of dates and making a column of days_of_the_week using the datetime module. I want to then write out a new csv file (or overright the individual one) containing one original element and the new element.

with open('new_dates.csv') as csvfile2:
 readCSV2 = csv.reader(csvfile2, delimiter=',')
 incoming = []
 for row in readCSV2:
 readin = row[0]
 time = row[1]
 year, month, day = (int(x) for x in readin.split('-'))
 ans = datetime.date(year, month, day)
 wkday = ans.strftime("%A")
 incoming.append(wkday)
 incoming.append(time)
with open('new_dates2.csv', 'w') as out_file:
 out_file.write('\n'.join(incoming))

Input files looks like this:

2017年03月02日,09:25
2017年03月01日,06:45
2017年02月28日,23:49
2017年02月28日,19:34

When using this code I end up with an output file that looks like this:

Friday
15:23
Friday
14:41
Friday
13:54
Friday
7:13 

What I need is an output file that looks like this:

Friday,15:23
Friday,14:41
Friday,13:54
Friday,7:13 

If I change the delimiter in out_file.write to a comma I just get one element of data per column, like this:

Friday 15:23 Friday 14:41 Friday 13:54 ....

Any thoughts would be appreciated. Thanks!

• patrick: The problem might be this join operation: '\n'.join(incoming). Try replacing that \n with a space to get the desired output; for a real csv, use ","
• Justin: using a space results all data in one long row, instead of two CSV columns. And as noted the use of a comma results in each value in a different field in a single row.
• Mark Tolonen: What does the input file look like?
• Justin: print wkday print time Thursday 16:45 Wednesday 20:42 Monday 12:12 Saturday 07:08
• Mark Tolonen: Please edit the question to show the input file that creates the output. Make it reproducible.

Answer 1

Being somewhat unclear on what format you want, I've assumed you just want a single space between wkday and time. For a quick fix, instead of appending both wkday and time separately, as in your example, append them together:

...
 incoming.append('{} {}'.format(wkday,time))
...

OR, build your incoming as a list of lists:

...
 incoming.append([wkday,time])
...

and change your write to:

with open('new_dates2.csv', 'w') as out_file:
 out_file.write('\n'.join([' '.join(t) for t in incoming]))

• Justin: Using that method seems to yield the same results as replace the \n with a space. All the data goes into a single field.
• Justin: The second option makes some even stranger results. All data is in a single column and each field is one character.
• Kind Stranger: @Justin sorry, missed a chunk out - above should work now
• Justin: Traceback (most recent call last): File "./analyze.py", line 32, in <module> out_file.write('\n'.join(' '.join(incoming))) TypeError: sequence item 0: expected string, list found

Answer 2

It seems you want Friday in column 0 and the time in column 1, so you need to change your incoming to a list of lists. That means the append statement should look like this:

...
 incoming.append([wkday, time])
...

Then, it is better to use the csv.writer to write back to the file. You can write the whole incoming in one go without worrying about formatting.

with open('new_dates2.csv', 'w') as out_file:
 writer = csv.writer(out_file)
 writer.writerows(incoming)

• Justin: Results in an error: Traceback (most recent call last): File "./analyze.py", line 35, in <module> out_file.write('\n'.join(incoming)) TypeError: sequence item 0: expected string, list found
• Stuart: @Justin Check the answer as it currently stands - I edited it.

Answer 3

Basically your incoming array is a linear list. So, you should have been doing is something like following:

#your incoming array
incoming = ['Friday', '15:23', 'Friday', '14:41', 'Friday', '13:54', 'Friday', '7:13']
#actual parsing of the array for correct output
for i,j in zip(incoming[::2], incoming[1::2]):
 out_file.write(','.join((i,j)))
 out_file.write('\n')

Answer 4

You don't really need the csv module for this. I'm guessing at the input, but from the description it looks like:

2017年03月02日,09:25
2017年03月01日,06:45
2017年02月28日,23:49
2017年02月28日,19:34

This will parse it and write it in a new format:

import datetime
with open('new_dates.csv') as f1, open('new_dates2.csv','w') as f2:
 for line in f1:
 dt = datetime.datetime.strptime(line.strip(),'%Y-%m-%d,%H:%M')
 f2.write(dt.strftime('%A,%H:%M\n'))

Output file:

Thursday,09:25
Wednesday,06:45
Tuesday,23:49
Tuesday,19:34

• Stuart: This is assuming there's nothing else in the input file. OP seems to want CSV output too.
• Mark Tolonen: @Stuart The question changed since I wrote this, but still, adding a comma is a slight modification to the strftime. No assuming involved, since the OP also added the input.