Question

我是Python新手，正在开设一些在线课程。我正在尝试将一些数据从段落格式转换为CSV格式（如下所示。）我能够导入包含段落格式的文本文件并将其导出为CSV但段落格式中的每一行都以单行格式输入导入电子表格。

import csv
import glob
import os

directory = raw_input("INPUT Folder:")
output = raw_input("OUTPUT Folder:")

txt_files = os.path.join(directory, '*.txt')

for txt_file in glob.glob(txt_files):
    with open(txt_file, "rb") as input_file:
        in_txt = csv.reader(input_file, delimiter='=')
        filename = os.path.splitext(os.path.basename(txt_file))[0] + '.csv'

        with open(os.path.join(output, filename), 'wb') as output_file:
            out_csv = csv.writer(output_file)
            out_csv.writerows(in_txt)

我不知道如何解析数据以将标签和空格与数值分开，并将每个段落部分组合成一行，其中包含CSV文件的引号和逗号。任何帮助将不胜感激！

段落格式：

12-03-06 15:19:36
流量：1.17365 g / m
POS：+ 9273x1Gal

12-03-06 15:19:37
流量：1.17849 g / m
POS：+ 9283x1Gal

12-03-06 15:19:38
流量：1.19849克/米
POS：+ 9293x1Gal

（重复）

所需的CSV输出（注意，我必须在+之前添加单引号以允许正确导入文本到电子表格中，否则它会以0形式输入）

“12-03-06 15:19:36”，“FLOW：”，“1.17365”，“g / m”，“POS：”，“'+”，“9273”，“x1”，“加尔“
“12-03-06 15:19:37”，“FLOW：”，“1.17849”，“g / m”，“POS：”，“'+”，“9283”，“x1”，“Gal”<登记/> “12-03-06 15:19:38”，“FLOW：”，“1.19849”，“g / m”，“POS：”，“'+”，“9293”，“x1”，“Gal”< BR />

Answer 1

我建议您使用collections.deque一次处理三行，并re.match解析您想要的项目：

# -*- coding: utf-8 -*-
from collections import deque
import csv
from functools import partial
import glob
import os
import re
import sys

if sys.hexversion < 0x3000000:
    # Python 2.x
    inp = raw_input
    open_csv_write = partial(open, mode="wb")
else:
    # Python 3.x
    inp = input
    open_csv_write = partial(open, mode="w", newline="")

POS_REG = re.compile("(POS:) ([+-])(\d+(?:\.\d+)?)(x\d+)(\w+)", re.I)

def change_ext(fn, new_ext):
    """
    Given `fn` as "path\filename.old_ext",
      return "path\filename" + new_ext
    """
    return os.path.splitext(fn)[0] + new_ext

def get_pos(line, reg=POS_REG):
    """
    Given a string like "POS: +92.73x1Gal",
      return ['POS:', '+', '92.73', 'x1', 'Gal']
    """
    match = reg.match(line)
    return list(match.groups()) if match else []

def process(inf, outcsv):
    # line queue
    q = deque(maxlen=3)
    # preload two lines
    q.append(next(inf, '').rstrip())
    q.append(next(inf, '').rstrip())
    # process rest of lines
    for line in inf:
        q.append(line.rstrip())
        if q[1].startswith('FLOW:'):
            pos = get_pos(line)
            if pos:
                row = [q[0]] + q[1].split() + pos
                outcsv.writerow(row)

def main():
    # get directories
    in_dir  = inp("Input directory: ")
    out_dir = inp("Output directory: ")

    # process file names
    in_filespec    =  os.path.join(in_dir, '*.txt')
    in_full_names  =  glob.glob(in_filespec)
    in_names       = [os.path.basename(fn)      for fn in in_full_names]
    out_names      = [change_ext(fn, ".csv")    for fn in in_names]
    out_full_names = [os.path.join(out_dir, fn) for fn in out_names]

    # operate on files
    for in_name, out_name in zip(in_full_names, out_full_names):
        with open(in_name) as inf, open_csv_write(out_name) as outf:
            outcsv = csv.writer(outf)
            process(inf, outcsv)

if __name__ == "__main__":
    main()

需要将段落格式的数据组合并转换为csv

1 个答案: