从一组文本文件中获取随机句子

Question

我不是一个程序员，但被要求帮助我的一位教授完成一个需要基本编程知识水平的研究项目。我们从一组 22 个文本文件中提取 1000 个句子。每个文本文件包含数千个句子，全部串在一个段落中，如下所示：

By this time , the resistance was severe . Why ? Consider the houses that were coming down , to be replaced by modern flats . They are typical of the row houses for workers of the dreary industrial towns of England . These " cottages , " as they were called in Sunderland , were built for the working class by unnamed builders one hundred years ago . Many were rented , but most were by this time owned by their inhabitants . The houses sit directly on the sidewalk . There are no trees on the street . There is no front yard .

很丑。无论如何，我不确定如何从这样的文本块中随机提取句子。我非常有信心这些是我的程序必须遵循的步骤：

Select 一组 22 个文本文件中的随机文本文件。
Select 文本文件中的随机句子。
- 我认为这可以通过选择一个随机数和 "counting" 个句号 (.) 达到该数字，然后选择该句号之后的句子作为随机句子，在下一个句号处停止来完成。
将所选句子写入单独的文本文件。
再重复 999 次。

这是我到目前为止编写的代码。它不像我上面考虑的那样 "count" 句号，因为我不知道该怎么做，但我试着开始概述我知道的基本部分。我知道我的代码很丑陋，但我以前从未真正编写过任何东西。感谢您的帮助！

fileNumber = 0
sentenceNumber = 0
i = 0
fname = "x"
sentence = "x"

for i < 1000

    fileNumber = random.randint(0, 22)          #chooses a random number in order to assign one of the text files
    sentenceNumber = random.randint(0, NUMBER OF FULLSTOPS) #chooses a random number to select a random sentence from the previously selected text file

    if fileNumber == 0              #assigns the file to search based on the random number stored in fileNumber
        fname = "w_acad_1990.txt"
    else if fileNumber == 1
        fname = "w_acad_1991.txt"
    else if fileNumber == 2
        fname = "w_acad_1992.txt"
    else if fileNumber == 3
        fname = "w_acad_1993.txt"
    else if fileNumber == 4
        fname = "w_acad_1994.txt"
    else if fileNumber == 5
        fname = "w_acad_1995.txt"
    else if fileNumber == 6
        fname = "w_acad_1996.txt"
    else if fileNumber == 7
        fname = "w_acad_1997.txt"
    else if fileNumber == 8
        fname = "w_acad_1998.txt"
    else if fileNumber == 9
        fname = "w_acad_1999.txt"
    else if fileNumber == 10
        fname = "w_acad_2000.txt"
    else if fileNumber == 11
        fname = "w_acad_2001.txt"
    else if fileNumber == 12
        fname = "w_acad_2002.txt"
    else if fileNumber == 13
        fname = "w_acad_2003.txt"
    else if fileNumber == 14
        fname = "w_acad_2004.txt"
    else if fileNumber == 15
        fname = "w_acad_2005.txt"
    else if fileNumber == 16
        fname = "w_acad_2006.txt"
    else if fileNumber == 17
        fname = "w_acad_2007.txt"
    else if fileNumber == 18
        fname = "w_acad_2008.txt"
    else if fileNumber == 19
        fname = "w_acad_2009.txt"
    else if fileNumber == 20
        fname = "w_acad_2010.txt"
    else if fileNumber == 21
        fname = "w_acad_2011.txt"
    else if fileNumber == 22
        fname = "w_acad_2012.txt"
    else
        fname = "x" 

    #select a random sentence
    #write that sentence to a text file

    i++

Answer 1

import random

# Do this 1000 times:
for i in range(1000):
    fileNumber = random.randint(0,22)
    # calculate file name instead of long condition chain:
    with open("w_acad_{}.txt".format(1990+fileNumber)) as f:
        # read file and split sentences on full stop:
        sentences = f.read().split(".")
    with open("sentence_{}.txt".format(i), "w") as f:
        # write a random sentence from the file:
        f.write(random.choice(sentences) + ".")

Answer 2

我建议这样：

import re
from random import sample

sentences = []
for i in range(23):
    with open('w_acad_{}.txt'.format(i)) as f:
        sentences += re.findall(r".*?[\.\!\?]+", f.read())

selected = sample(sentences, 1000)
with open('out.txt', 'w') as f:
    f.write(''.join(selected))

首先，如果你选择句子前的文件，所有句子之间不是等概率的。最好把之前的句子都检索出来再选择。

不要打开文件 1000 次。好重啊！

不要按照其他人的建议使用 split，因为这样会丢失分隔符。在这里，我使用 re 模块中的 findall，这样我就可以使用多个定界符（不仅是 .，还有 ! 等），并且在检索句子时保留它。

最后，您可以使用 random 模块中的 sample 来 select 1000 个句子。

Answer 3

万一这对其他人有用：这是一个修改，从 txt 和 returns 中删除了一个句子，剩余的文本和删除的句子

def remove_random_sentence(txt):
    sentences = []
    sentences += re.findall(r".*?[\.\!\?]+", txt)
    if len(sentences) > 1:
        removed_sentence = sample(sentences, 1)[0]  # sample returns list, accsess first element to get string
        return txt.replace(removed_sentence, ''), removed_sentence
    else:
        return 'ONLY ONE SENTENCE', 'REMOVE'

从一组文本文件中获取随机句子

Get random sentences from a set of text files

python

random

linguistics