读取多个输入文件并为每个输入文件写入多个输出文件的脚本？

Question

我有多个如下格式的输入文件需要处理。

输入文件路径/tmp/input.

1.1.1.txt
1.1.2.txt
1.1.3.txt

但是，我想在另一个文件夹中为每个输入文件生成输出文件，假设 (/tmp/outputsmgr) 如下所示：

1.1.1_output.csv
1.1.2_output.csv
1.1.3_output.csv

问题是：

首先，我无法将输出文件写入 another/different 文件夹
其次，所有输入文件数据在处理后合并到输入文件夹中的一个文件中，如下所示，而不是每个输入文件单独的输出文件

以下所有文件都包含相同的数据 1.1.1.txt 数据应在文件 1.1.1_output.csv 中，文件 1.1.2.txt 数据应在文件 1.1.2_output.csv.[=20= 中]

1.1.1.txt_output.csv
1.1.2.txt_output.csv
1.1.3.txt_output.csv

如何修改以下代码以获得所需的结果？

import os
import csv
import re

def parseFile(fileName):
    # We are using a dictionary to store info for each file
    data = list()
#    data = dict()

    fh = open(fileName, "r")
    lines = fh.readlines()[1:]
    for line in lines:
        line = line.rstrip("\n")
        if re.search("sessmgr", line):
            splitted = line.split()
            temp = dict()
            temp["CPU"] = splitted[0]
            temp["facility"] = splitted[1]
            temp["instance"] = splitted[2]
            temp["cpu-used"] = splitted[3]
            temp["cpu-allc"] = splitted[4]
            temp["mem-used"] = splitted[5]
            temp["mem-allc"] = splitted[6]
            temp["files-used"] = splitted[7]
            temp["files-allc"] = splitted[8]
            temp["sessions-used"] = splitted[9]
            temp["sessions-allc"] = splitted[10]
#            print (splitted[2])

            data.append(temp)
#            continue;
#    print (data)
    return data

if __name__ == "__main__":
    inputsDirectory = "/tmp/input"
    outputDirectory = "/tmp/outputsmgr"
    path = os.path.abspath(inputsDirectory)
    pathout = os.path.abspath(outputDirectory)
    fileLists = ["{0}/{1}".format(path,x) for x in os.listdir(outputDirectory)]
    fileList = ["{0}/{1}".format(path,x) for x in os.listdir(inputsDirectory)]
#    print(fileList)

    csvRows = []
    for file in fileList:
        newRow = parseFile(file)
        csvRows.append(newRow)

#    print(csvRows)
    for files in fileList:
        outputFile = "output.csv"
        csvfile = open(os.path.join(files + "_" + outputFile), 'w')
        fieldnames = ["CPU",
                      "facility",
                      "instance",
                      "cpu-used",
                      "cpu-allc",
                      "mem-used",
                      "mem-allc",
                      "files-used",
                      "files-allc",
                      "sessions-used",
                      "sessions-allc"]
        writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
        headers = {}
        for n in writer.fieldnames:
            headers[n] = n
        writer.writerow(headers)
#       writer.writeheader()
        for row in csvRows:
            for obj in row:
                print (obj)
                writer.writerow(obj)

Answer 1

我认为下面的代码可以满足您的要求。它按顺序处理输入目录中的文件，并将 parseFile() 函数返回的结果写入输出目录中相应的输出文件。从每个输入文件中获取一组新的 csvRows 并将（仅）写入每个输出文件非常重要。

代码假定 outputDirectory 已经存在，但如果不是这种情况，则您需要添加代码以在处理任何文件之前创建它。 提示： 将 os.path.exists() and os.path.isdir() 与 os.makedirs() 结合使用。

import csv
import os
import re


def parseFile(filePath, fieldnames, target_re=r"sessmgr"):
    """ Yield lines of file matching target regex. """
    with open(filePath, "r") as file:
        next(file)  # Skip/ignore first line.
        for line in file:
            if re.search(target_re, line):
                yield dict(zip(fieldnames, line.split()))


if __name__ == "__main__":

    OUTPUT_FILE_SUFFIX = "output.csv"
    inputsDirectory = "/tmp/input"
    outputDirectory = "/tmp/outputsmgr"
    fieldnames = ("CPU", "facility", "instance", "cpu-used", "cpu-allc", "mem-used",
                  "mem-allc", "files-used", "files-allc", "sessions-used",
                  "sessions-allc")

    input_dir = os.path.abspath(inputsDirectory)
    output_dir = os.path.abspath(outputDirectory)

    for in_filename in os.listdir(input_dir):
        in_filepath = os.path.join(input_dir, in_filename)
        print('in_filepath:  "{}"'.format(in_filepath))

        in_rootname = os.path.splitext(in_filename)[0]
        out_filename = in_rootname + "_" + OUTPUT_FILE_SUFFIX
        out_filepath = os.path.join(output_dir, out_filename)
        print('out_filepath: "{}"'.format(out_filepath))

        with open(out_filepath, 'w') as csvfile:
            writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
            writer.writeheader()
            writer.writerows(parseFile(in_filepath, fieldnames))

读取多个输入文件并为每个输入文件写入多个输出文件的脚本？

Script to read multiple input file and writing multiple output file for each input file?

python

python-2.7

scripting

python-2.6