读取csv文件以解析日期

Question

我有一个.csv文件，其中包含以下数据：

equipement,"144444444"
Date,"Time","measure"
16/09/2016,"07:15:00","16.47777"
16/09/2016,"07:30:00","15.44454"
16/09/2016,"07:45:00","16.21114"

我在此文件上运行python代码，我的目标是将类似以下内容作为输出：

"measure","20160916071500","16.47777"
"measure","20160916073000","15.44454"
"measure","20160916074500","16.21114"

这是我的代码：

import csv
import sys
import os
import re
import fnmatch
import csv
from dateutil.parser import parse as parseDate
from datetime import datetime, time, timedelta


    file = open("myfile.csv", 'rt')
    reader = csv.reader(file)
    next(reader)
    rows = list(reader)
    firstline = rows[0]
    header = firstline[2]
    print header
    for row in rows:
        next(reader)
        print rows[0]
        if "".join(row).strip() != "":
            chaine = str(row[0]+row[1])
            #print chaine
            date = chaine[:10] + " " + chaine[11:]
            #print date
            index = parseDate(date)
            index = str(index).replace('-','')
            index = str(index).replace(':','')
            index = str(index).replace(' ','')
            data = row[2]

我的问题是，我需要执行next（reader）来跳过文件中的第一行和第二行，因为它们不包含任何日期。 但我得到这个错误：

Traceback (most recent call last): File "t.py", line 19, in <module> next(reader) StopIteration

任何想法？

Answer 1

通过执行rows = list(reader) ，您已经用尽了reader ，并将结果收集在一个名为rows的列表中。 再次执行next(reader)将提高StopIteration 。

但是，不必创建rows列表。 您可以使用for循环直接遍历reader ：

reader = csv.reader(file)
next(reader)               # skip first line
secondline = next(reader)  # capture second line
header = secondline[2]
for row in reader:         # iterate from third line to the end
    # next(reader) <-- don't do this, the for loop already does it for you
    if "".join(row).strip() != "":
        # ... your code processing row ...

Answer 2

如果您愿意，可以用熊猫解决它：

import pandas as pd

df = pd.read_csv('in.csv', skiprows=2, header=None, parse_dates=[[0,1]])
df['dt']=df["0_1"].apply(lambda x: x.strftime('%Y%m%d%H%M%S'))
df['mes'] = pd.Series(["measure"]*len(df), index=df.index)
df[['mes','dt',2]].to_csv('out.csv', quoting=True, index=None,header=None)

CSV档案：

"measure","20160916071500","16.47777"
"measure","20160916073000","15.44454"
"measure","20160916074500","16.21114"

Answer 3

您可以仅使用two for loops和一些字符串替换来获得相同的期望输出，例如以下示例（我假设您的输入名为in.csv ）：

data = list(k.strip("\n") for k in open("in.csv", 'r'))
mesure = data[1].split(",")[2]
m = list(k.replace('"', "").split(",") for k in data[2:])

final, d =[], ""
for k in m:
    for j in k[:-1]:
        if "/" in j:
            d = '"%s' % "".join(j.split("/")[::-1])
        if ":" in j:
            d += '%s"' % "".join(j.split(":"))
    final.append(",".join([mesure, d,'"%s"' % k[-1:][0]]))

for k in final:
    print(k)

输出：

"measure","20160916071500","16.47777"
"measure","20160916073000","15.44454"
"measure","20160916074500","16.21114"