Remove comma from csv file python

Try with str.replace instead:

df['size'] = df['size'].str.replace(',', '')

Optional convert to int with astype:

df['size'] = df['size'].str.replace(',', '').astype(int)

   number   name     size
0       1    Car   932123
1       2   Bike   100000
2       3  Truck  1032111

Sample Frame Used:

df = pd.DataFrame({'number': [1, 2, 3], 'name': ['Car', 'Bike', 'Truck'],
                   'size': ['9,32,123', '1,00,000', '10,32,111']})

   number   name       size
0       1    Car   9,32,123
1       2   Bike   1,00,000
2       3  Truck  10,32,111

I have a .csv that was created using input_xls.to_csv, and everything delimits correctly.  However, within some of the string records are commas (e.g. a business name like 'Widgets, Inc.').  The result is incorrect delimiting when importing the .csv into a file geodatabase using arcpy.conversion.TableToTable -  values end up shifted over into other columns.  I want to replace the commas with a space.

I cannot figure out how to use either something like

.apply(lambda x: x.replace(',', ' '))

or

 [val if val else " " for val in row] + ([" "] * (len(first_row) - len(row))) (found here: https://community.esri.com/t5/python-questions/find-and-replace-blanks-in-csv-with-python/m-p/673632) 

Any help to nudge me in the right direction would be appreciated.  Thanks.

• csv
• python

• All Posts
• Previous Topic
• Next Topic

6 Replies

What you can do is read your excel file into an array, where when appending data into the array, you call the replace command to get rid of the comma.

inRow = [str(row[0]),str(row[0]),str(row[2]),str(row[3]),str(row[4]),str(row[5]),str(row[6]),str(row[7]),str(row[8]),str(row[9]),str(row[10])]

index = 0

for i in inRow:

     if "," in i:

          i = i.replace(",","")

          inRow[index] = i

     index += 1

points.append(inRow)

And then use that array to create the csv.

        AllProjects_CSV.truncate()

        writer = csv.writer(AllProjects_CSV,delimiter=',')

        writer.writerow(fields)

        writer.writerows(points)

My example is pulling data from a SDE database but I am pretty sure you can use something like openpyxl to read through an excel sheet.

You might want to try bringing in the csv back into Excel, then  try the Excel to Table tool to see if it handles the embedded commas better when bringing back into a file geodatabase

What is your input_xls?  Is it a pandas DataFrame?

Yes, I'm using pandas .read_excel and .to_csv

I was out Friday and will come back to this later today, but plan to look further at what Matt Kramer posted above.  Thanks

Personally I cannot recreate your probleme, but you have two options.

1. change regional settings of your machine to use different seperators (semicolon for example) 

2. in the same folder as your data add file "schema.ini", where you  specify how data is seperated, additionnaly you can change the name of columns and its types. 

I never followed-up with the solution I found using csv and pandas:

    with open(cama_csv_file, 'r') as f1:

        with open(cama_csv_file_final, 'w+') as f2:

            df = pd.read_csv(f1)

            df.replace(',',' ', regex=True, inplace=True)

            df.to_csv(f2, index=False, line_terminator='\r')

    f2.close()

    f1.close()

Creating a new .csv is not an issue, because I delete all of them at the end of the process.

How do I remove commas from a CSV file?

How do I remove a comma from a dataset in Python?

How do I manage a CSV file in Python?

How do I change CSV separator in Python?