如何将代码拆分为更小的函数

Question

我有一个可以运行的应用程序。但是为了更好地理解函数和 python。我正在尝试将其拆分为各种功能。

我卡在 file_IO 函数上。我确定它不起作用的原因是因为应用程序的主要部分不理解 reader 或 writer。为了更好地解释。这是应用程序的完整副本。

我也对使用 csv.DictReader 和 csv.DictWriter 感到好奇。是否对当前代码提供任何优点/缺点？

我想另一种方法是通过类，老实说我也想知道如何做到这一点。

#!/usr/bin/python

""" Description This script will take a csv file and parse it looking for specific criteria.  
A new file is then created based of the original file name containing only the desired parsed criteria.
"""

import csv
import re
import sys

searched = ['aircheck', 'linkrunner at', 'onetouch at']

def find_group(row):
    """Return the group index of a row
        0 if the row contains searched[0]
        1 if the row contains searched[1]
        etc
        -1 if not found
    """
    for col in row:
        col = col.lower()
        for j, s in enumerate(searched):
            if s in col:
                return j
        return -1



#Prompt for File Name
def file_IO():
    print "Please Enter a File Name, (Without .csv extension): ",
    base_Name = raw_input()
    print "You entered: ",base_Name

    in_Name = base_Name + ".csv"
    out_Name = base_Name + ".parsed.csv"

    print "Input File: ", in_Name
    print "OutPut Files: ", out_Name

    #Opens Input file for read and output file to write.
    in_File = open(in_Name, "rU")
    reader = csv.reader(in_File)

    out_File = open(out_Name, "wb")
    writer = csv.writer(out_File, delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL)

    return (reader, writer)

file_IO()

# Read header
header = reader.next()


stored = []
writer.writerow([header[0], header[3]])

for i, row in enumerate(reader):
    g = find_group(row)
    if g >= 0:
        stored.append((g, i, row))
stored.sort()

for g, i, row in stored:
    writer.writerow([row[0], row[3]])


# Closing Input and Output files.
in_File.close()
out_File.close()

Answer 1

如果我是你，我只会分开find_group。

import csv

def find_group(row):
    GROUPS = ['aircheck', 'linkrunner at', 'onetouch at']
    for idx, group in enumerate(GROUPS):
        if group in map(str.lower, row):
            return idx
    return -1

def get_filenames():
    # this might be the only other thing you'd want to factor
    # into a function, and frankly I don't really like getting
    # user input this way anyway....
    basename = raw_input("Enter a base filename (no extension): ")
    infilename = basename + ".csv"
    outfilename = basename + ".parsed.csv"
    return infilename, outfilename
    # notice that I don't open the files yet -- let main handle that

infilename, outfilename = get_filenames()

with open(infilename, 'rU') as inf, open(outfilename, 'wb') as outf:
    reader = csv.reader(inf)
    writer = csv.writer(outf, delimiter=',',
                        quotechar='"', quoting=csv.QUOTE_ALL)
    header = next(reader)
    writer.writerow([[header[0], header[3]])
    stored = sorted([(find_group(row),idx,row) for idx,row in
                     enumerate(reader)) if find_group(row) >= 0])
    for _, _, row in stored:
        writer.writerow([row[0], row[3]])