在 Python 中使用 CSV 模块删除列

Question

I realize this has been asked (many) times before, but I've been trying different solutions and none of them are working for me - clearly I'm doing something wrong, but I'm not sure what.

We're learning how to scrub data in Python, so what I'm trying to do is take a text file (that has been converted to a text file from excel) as input and output my scrubbed data. Data is a mix of text and numbers and each cell either has text or numbers but not both. I'm trying to delete certain columns, and I can't figure out how to. I would really appreciate it if I could get answers just using the csv package (or no package at all) - I know pandas is supposed to be helpful, but I'm trying to go by what we're using in class.

This is the code I currently have right now; when I run it, I just get a blank excel sheet as my output.

import csv

def airbnb_csv():

    source = '/Users/(myname)/Desktop/airbnb.txt'
    target = 'scrubbed_airbnb2.csv'

    with open(source,'r') as fp_in:
        reader = csv.reader(fp_in, delimiter=',')
        with open(target,'w') as fp_out:
            writer = csv.writer(fp_out, delimiter=',')
            for r in reader:
                writer.writerow((r[2], r[3], r[5], r[7], r[8], r[9], 
                r[10], r[11], r[13]))

I have other code that did get me a filled in excel sheet as output. This was my original code, but it got weird fast.

for row in fp_in:
     if (row[:5].isdigit()):
         v = row.split()
         v = v[:9]
         writer.writerow(v)
         if row.startswith("room_id") and not header_written:
            header_written = True
            v = row.split()

Thank you so much for any and all help/advice/hints you can give me! (no need to correct my code if it's too messy to deal with, but just wanted to add it so I didn't look like I was trying to get my homework done for me)

Answer 1

So, here's some directions.

First, not directly regarding your question, context managers can be chained, ie:

with open('input') as inp, open('output') as out:
    do_your_stuff

This will save your from some indentation pain.

More on question: there's no way to "delete" columns for most modern filesystems, so you need to read, process, write to other file (being honest, there's ways around, but that requires way more work). In your case, processing would be selecting what to write (or what to skip). The best way to do this, while keeping your code readable and maintainable is to useDictReader andDictWriter . Once you reading/writing csv rows by named entities, everything becomes easy:

fields_needed = ['price', 'rooms']
with open(source) as fp_in, open(target, 'w') as fp_out:
    reader = csv.DictReader(fp_in)
    writer = csv.DictWriter(fp_out, fieldnames=fields_needed, extrasaction='ignore')

    writer.writeheader()
    for r in reader:
        writer.writerow(r)