如何将csv文件中的唯一个体复制到新文件中

时间:2017-12-07 02:52:27

标签: python csv arcpy

我对python很新,所以请耐心等待。我有一个csv文件,如下所示:

Animal Locations

我正在尝试遍历文件,并且对于每个唯一的个人,创建一个新的csv文件并复制行。我成功地为一只动物做了这个,但是我在创建更通用方法的语法时遇到了麻烦。这是我现在拥有的:

import arcpy
import csv
from csv import DictReader

WS = arcpy.env.workspace = raw_input("Where if your workspace")
infile = raw_input("where is your file?") 
outfile = raw_input("What is your outfile name?") 
arcpy.env.overwriteOutput = True


with open(infile, "r") as csvFile, open(outfile, "w") as out, open("outfile2.csv", "w") as out2:
    reader = csv.DictReader(csvFile)
    writer = csv.writer(out)
    writer.writerow(reader.fieldnames)
    for row in reader:
         if row["Animal"] == "1":
            values = [row[field] for field in reader.fieldnames]
            writer.writerow(values)

1 个答案:

答案 0 :(得分:1)

要将每个Animal写入自己的CSV文件,您需要为每种类型的动物打开不同的文件。这可以通过使用字典来存储每个动物的文件对象和csv编写器对象来完成。最后,这可以用于正确关闭所有文件:

import csv

output_csvs = {}    # e.g. {'1' : [file_object, csv_object]}

with open('input.csv', 'rb') as f_input:
    csv_reader = csv.reader(f_input)
    header = next(csv_reader)

    for row in csv_reader:
        animal = row[0]

        if animal in output_csvs:
            output_csvs[animal][1].writerow(row)
        else:
            f_output = open('animal_{}.csv'.format(animal), 'wb')
            csv_output = csv.writer(f_output)
            output_csvs[animal] = [f_output, csv_output]
            csv_output.writerow(header)
            csv_output.writerow(row)

for csv_file, csv_writer in output_csvs.values():
    csv_file.close()

这将为您提供一组根据动物命名的输出CSV文件,例如animal_1.csv

或者,如果数据足够小以便读入内存,则可以通过动物对其进行排序,并通过使用Python itertools.groupby()函数一次输出一个块:

from itertools import groupby
import csv

with open('input.csv', 'rb') as f_input:
    csv_reader = csv.reader(f_input)
    header = next(csv_reader)

    for animal, group in groupby(sorted(csv_reader), lambda x: x[0]):
        with open('animal_{}.csv'.format(animal), 'wb') as f_output:
            csv_output = csv.writer(f_output)
            csv_output.writerow(header)
            csv_output.writerows(group)

使用sorted()可确保将所有同类动物组合在一起。如果数据已经是这种情况,则不需要排序。

要访问这些文件,您可以使用glob.glob()

import matplotlib.pyplot as plt            
import glob

for animal_filename in glob.glob('animal_*.csv'):
    with open(animal_filename, 'rb') as f_input:
        csv_input = csv.reader(f_input)
        heading = next(csv_input)
        x, y = [], []

        for row in csv_input:
            x.append(int(row[1]))
            y.append(int(row[2]))

        fig, ax = plt.subplots()
        plt.title(animal_filename)
        ax.scatter(x, y)

plt.show()