List to float

[Felipe]

Hi guys,

I have a 3.1 Gb csv file which I try to clean of bad data. So, I wrote the code below:

#!/usr/bin/env python3

import csv

with open('teste.csv') as csvfile:
    readCSV = csv.reader(csvfile, delimiter=',')
    fh = open("saida.txt","a")
    
    for row in readCSV:
        if row[0] and row[1] != "":
            print(row, file = fh)
    fh.close() 

Here's the output file:

Output:
['1', '1', '1']
['1', '1', '1']
['1', '1', '1']
['1', '1', '1']
['1', '1']
['1', '1']
['1', '1', '1']

It works fine, but as I discovered later, I wrote a txt file with a "List" type. That cause the inconvenience to print " [ ] ' ' " characters on the txt file. I need a way to convert this list in my txt file to float, or even change the code to print float instead list.

I'm new on python, so I appreciate the help.

[buran]

It's not that you entered a list, the csv reader returns each row as list of str objects.


from the docs for csv.reader:

Each row read from the csv file is returned as a list of strings. No automatic data type conversion is performed.

What shall you do, depends what you want to do with the data from the row

here is how you can print or convert to a list of floats

>>> row = ['1', '2', '0']
>>> print ', '.join(row)
1, 2, 0
>>> print map(float, row)
[1.0, 2.0, 0.0]

---

Just realised that you write the rows to second file. You should use with context manager for both files:

#!/usr/bin/env python3
 
import csv
 
with open('teste.csv') as csvfile, open("saida.txt","a") as fh:
    readCSV = csv.reader(csvfile, delimiter=',')
    if row[0] and row[1]:
        fh.write('{}\n'.format(','.join(row)))

[Felipe]

It's not that you entered a list, the csv reader returns each row as list of str objects.


from the docs for csv.reader:

Each row read from the csv file is returned as a list of strings. No automatic data type conversion is performed.

What shall you do, depends what you want to do with the data from the row

here is how you can print or convert to a list of floats

>>> row = ['1', '2', '0']
>>> print ', '.join(row)
1, 2, 0
>>> print map(float, row)
[1.0, 2.0, 0.0]

---

Just realised that you write the rows to second file. You should use with contects manager for both files:

#!/usr/bin/env python3
 
import csv
 
with open('teste.csv') as csvfile, open("saida.txt","a") as fh:
    readCSV = csv.reader(csvfile, delimiter=',')
    if row[0] and row[1]:
        fh.write('{}\n'.format(','.join(row)))

Thanks for the help !!

I just added "for row in readCSV:" before "if row[0] and row[1]:", and I changed the comma in " ','.join(row))" for tabulation and worked perfect to me.

#!/usr/bin/env python3
  
import csv
  
with open('teste.csv') as csvfile, open("saida.txt","a") as fh:
    readCSV = csv.reader(csvfile, delimiter=',')
    for row in readCSV:
        if row[0] and row[1]:
            fh.write('{}\n'.format('    '.join(row)))

Now I have a file with numbers and tabulation that's the best to my use:

Output:
1    1    1
1    1    1
1    1    1
1    1    1
1    1
1    1
1    1    1

[buran]

#!/usr/bin/env python3
   
import csv
   
with open('teste.csv') as csvfile, open("saida.txt","a") as fh:
    readCSV = csv.reader(csvfile, delimiter=',')
    for row in readCSV:
        if row[0] and row[1]:
            fh.write('{}\n'.format('\t'.join(row)))

you should use \t for TAB

[Felipe]

#!/usr/bin/env python3
   
import csv
   
with open('teste.csv') as csvfile, open("saida.txt","a") as fh:
    readCSV = csv.reader(csvfile, delimiter=',')
    for row in readCSV:
        if row[0] and row[1]:
            fh.write('{}\n'.format('\t'.join(row)))

you should use \t for TAB

Ok, thanks for the tip.

[snippsat]

An alternative is that there also a write csv object.
csv.writer(csv_out, delimiter='\t')
So here you can set tab before writing.

import csv

with open('test.csv') as csvfile, open("saida.txt","w") as fh:
    read_csv = csv.reader(csvfile, delimiter=',')
    write_csv = csv.writer(fh, delimiter='\t')
    for row in read_csv:
        if row[0] and row[1]:
            write_csv.writerow(row)

[Felipe]

An alternative is that there also a write csv object.
csv.writer(csv_out, delimiter='\t')
So here you can set tab before writing.

import csv

with open('test.csv') as csvfile, open("saida.txt","w") as fh:
    read_csv = csv.reader(csvfile, delimiter=',')
    write_csv = csv.writer(fh, delimiter='\t')
    for row in read_csv:
        if row[0] and row[1]:
            write_csv.writerow(row)

Thanks !! That's a nice way too.