CSV 计算脚本未保存正确数量的列答案

【问题标题】：CSV calulation script not saving the correct amount of columnsCSV 计算脚本未保存正确数量的列
【发布时间】：2013-11-21 11:36:48
【问题描述】：

@Stackoverflow

我有一个脚本，它根据匹配的 3 个条件将三列添加到我的 csv 末尾。

我了解问题所在，代码是从 rdr 重新读取行，而 wrt 是修改后的行，在初始循环之后我尝试遍历 wrt，但代码无法做到这一点。

我的代码：

import csv
import datetime
import copy
from collections import defaultdict

with open(r"FullMergedData.csv") as i, open(r"FullMergedDataWSPSR.csv", "wb") as o:
    rdr = csv.reader(i, delimiter ="|")
    wrt = csv.writer(o, delimiter ="|")

    data, currdate = defaultdict(lambda:[0, 0, 0, 0]), None
    calclist = [6, 7, 19, 23, 25, 26, 35, 62, 64]

    for calc in calclist:
        for counter2, line in enumerate(rdr):
            if counter2 != 0:
                date, name = datetime.datetime.strptime(line[50], '%d/%m/%Y'), line[calc]

                if date != currdate or not currdate:
                    for v in data.itervalues():
                        v[:2] = v[2:]
                    currdate = date

                top,bottom = data[name][0:2]
                try:
                    quotient = float(top)/bottom
                except ZeroDivisionError:
                    quotient = 0
                wrt.writerow(line + data[name][:2]+ [quotient])

                data[name][3] += 1
                if line[38] == "1": data[name][2] += 1

如果有人可以帮助使此代码正常工作，我将不胜感激。非常感谢 AEA

产生回溯的代码

import csv
import datetime
import copy
from collections import defaultdict

with open(r"FullMergedData.csv") as i, open(r"FullMergedDataWSPSR.csv", "wb") as o:
    rdr = csv.reader(i, delimiter ="|")
    wrt = csv.writer(o, delimiter ="|")

    data, currdate = defaultdict(lambda:[0, 0, 0, 0]), None
    calclist = [6, 7, 19, 23, 25, 26, 35, 62, 64]

    for counter, calc in enumerate(calclist):
        if counter == 0:
            for counter2, line in enumerate(rdr):
                if counter2 != 0:
                    date, name = datetime.datetime.strptime(line[50], '%d/%m/%Y'), line[calc]

                    if date != currdate or not currdate:
                        for v in data.itervalues():
                            v[:2] = v[2:]
                        currdate = date

                    top,bottom = data[name][0:2]
                    try:
                        quotient = float(top)/bottom
                    except ZeroDivisionError:
                        quotient = 0
                    wrt.writerow(line + data[name][:2]+ [quotient])

                    data[name][3] += 1
                    if line[38] == "1": data[name][2] += 1 

        else:
            for line in wrt:
                    date, name = datetime.datetime.strptime(line[50], '%d/%m/%Y'), line[calc]

                    if date != currdate or not currdate:
                        for v in data.itervalues():
                            v[:2] = v[2:]
                        currdate = date

                    top,bottom = data[name][0:2]
                    try:
                        quotient = float(top)/bottom
                    except ZeroDivisionError:
                        quotient = 0
                    wrt.writerow(line + data[name][:2]+ [quotient])

                    data[name][3] += 1
                    if line[38] == "1": data[name][2] += 1

追溯

Traceback (most recent call last):
  File "C:\mergecalculator.py", line 35, in <module>
    for line in wrt:
TypeError: '_csv.writer' object is not iterable

【问题讨论】：

您看到了什么错误？可以发回溯吗？
@SunPowered 添加了我尝试过的代码 + 回溯
你为什么要遍历一个 writer 对象？你想用for line in wrt:做什么？
您不能遍历 csv 写入器。对于类文件对象，迭代器逐行读取文件。你确定你不想遍历rdr，即for line in rdr:
@SunPowered 我没有尝试就发布了第一段代码的原因是因为我明白这是错误的。所需的输出应该包括 27 个额外的列，而不是只产生 3 个。因为它没有查看修改后的数据。

标签： python loops python-2.7 csv

【解决方案1】：

我不是 100% 确定这是你想要的，但试试这个逻辑。您想处理 calclist 中的每个项目以生成额外的 3 列。在处理完 calclist 的所有元素后，将生成的额外列添加到该行并写入 csv 文件。

...
for counter2, line in enumerate(rdr):
    extra_cols = []
    for calc in calclist:
        # process data to get extra 3 columns
        extra_cols.extend(data[name][:2]+ [quotient])
    wrt.writerow(line + extra_cols)

【讨论】：