Question

我正在寻找解析csv文件并聚合2列。

csv文件中的数据：

'IP Address', Severity
10.0.0.1, High
10.0.0.1, High
10.0.0.1, Low
10.0.0.1, Medium
10.0.0.2, Medium
10.0.0.2, High
10.0.0.2, Low
10.0.0.3, Medium
10.0.0.3, High
10.0.0.3, Medium

我希望获得以下行的输出：

'IP Address', Severity
10.0.0.1, High:2, Medium:1, Low:1
10.0.0.2, High:1, Medium:1, Low:1
10.0.0.3, High:1, Medium:2, Low:0

或（理想情况下）

'IP Address', High, Medium, Low
10.0.0.1, 2, 1, 1
10.0.0.2, 1, 1, 1
10.0.0.3, 1, 2, 0

我最近的来到这里： Parse CSV file and aggregate the values

我似乎无法聚合字符串（Severity）变量。

如何输出这些数据？

感谢任何帮助。

Answer 1

import csv 
from collections import defaultdict

with open('text.txt') as f, open('ofile.csv','w+') as g:
    reader,writer = csv.reader(f), csv.writer(g)
    results = defaultdict(list)
    next(reader) #skip header line
    for ip,severity in reader:
        results[ip].append(severity)
    writer.writerow(["'IP Adress'"," High"," Medium"," Low"]) #Write headers
    for ip,severities in sorted(results.iteritems()):
        writer.writerow([ip]+[severities.count(t) for t in [" High"," Medium"," Low"]])

产地：

'IP Adress', High, Medium, Low
10.0.0.1,2,1,1
10.0.0.2,1,1,1
10.0.0.3,1,2,0

Answer 2

这是我的解决方案，ag.py：

import collections
import csv
import sys

output = collections.defaultdict(collections.Counter)

with open(sys.argv[1]) as infile:
    reader = csv.reader(infile)
    reader.next() # Skip header line
    for ip,level in reader:
        level = level.strip() # Remove surrounding spaces
        output[ip][level] += 1

print "'IP Address',High,Medium,Low"
for ip, count in output.items():
    print '{0},{1[High]},{1[Medium]},{1[Low]}'.format(ip, count)

要运行解决方案，请发出以下命令：

python ag.py data.csv

讨论

output是一个字典，其键是IP，值是collections.Counter个对象。
每个计数器对象计算特定IP的“高”，“中”和“低”
我的解决方案打印到stdout，您可以修改它以打印到文件

在python中解析csv文件和聚合值

2 个答案:

讨论