如何在Python中获取由逗号分隔的数字组？

Question

我有以下文本:

Cluster 7: {4, 15, 21, 28, 33, 35, 43, 47, 53, 57, 59, 66,
       69, 70, 74, 86, 87, 88, 90, 114, 136, 148, 201,
       202, 212, 220, 227, 250, 252, 253, 259, 262, 267,
       270, 282, 296, 318, 319, 323, 326, 341}
Cluster 8: {9, 10, 11, 20, 39, 55, 79, 101, 108, 143, 149,
       221, 279, 284, 285, 286, 287, 327, 333, 334, 335,
       336}
Cluster 9: {3, 64, 83, 93, 150, 153, 264, 269, 320, 321, 322}
Cluster 10: {94, 123, 147}

我希望你能提取每组中的数字并进行聚类分析，相关内容与IT技术有关。我已经尝试了使用正则表达式但效果不佳。

我已尝试过：

regex="(Cluster \d+): \{((\d+)[,\}][\n ]+)+|(?:(\d+),[\n ])"

但是这些组不匹配。

我想要的输出结果是：

["Cluster 7", '4', '15', '21', '28', '33', '35', '43', '47', '53', '57', '59', '66', '69', '70', '74', '86', '87', '88', '90', '114', '136', '148', '201', '202', '212', '220', '227', '250', '252', '253', '259', '262', '267', '270', '282', '296', '318', '319', '323', '326', '341', "Cluster 8", '9', '10', '11', '20', '39', '55', '79', '101', '108', '143', '149', '221', '279', '284', '285', '286', '287', '327', '333', '334', '335', '336', "Cluster 9", '3', '64', '83', '93', '150', '153', '264', '269', '320', '321', '322', "Cluster 10", "94", "123", "147"]

也许这不是最好的方法来做这件事。

谢谢。

- Rafael Rios

Answer 1

我不建议使用正则表达式来处理这个问题。你的文本是符合 yaml 规范的，可以直接使用 oyaml 这样的有序 yaml 加载器进行加载。

import oyaml as yaml   # pip install oyaml
data = yaml.load(text)

要将那个字典解包成所需的“扁平”结构，只需使用列表推导式：

[x for (k, v) in data.items() for x in (k, *v)]

_{注: 我是 oyaml 的作者。}

- wim

Answer 2

你可以创建一个更通用的正则表达式：

import re
s = '\nCluster 7: {4, 15, 21, 28, 33, 35, 43, 47, 53, 57, 59, 66,\n       69, 70, 74, 86, 87, 88, 90, 114, 136, 148, 201,\n       202, 212, 220, 227, 250, 252, 253, 259, 262, 267,\n       270, 282, 296, 318, 319, 323, 326, 341}\nCluster 8: {9, 10, 11, 20, 39, 55, 79, 101, 108, 143, 149,\n       221, 279, 284, 285, 286, 287, 327, 333, 334, 335,\n       336}\nCluster 9: {3, 64, 83, 93, 150, 153, 264, 269, 320, 321, 322}\nCluster 10: {94, 123, 147}\n'
data = re.findall('Cluster \d+|\d+', s)

输出：

['Cluster 7', '4', '15', '21', '28', '33', '35', '43', '47', '53', '57', '59', '66', '69', '70', '74', '86', '87', '88', '90', '114', '136', '148', '201', '202', '212', '220', '227', '250', '252', '253', '259', '262', '267', '270', '282', '296', '318', '319', '323', '326', '341', 'Cluster 8', '9', '10', '11', '20', '39', '55', '79', '101', '108', '143', '149', '221', '279', '284', '285', '286', '287', '327', '333', '334', '335', '336', 'Cluster 9', '3', '64', '83', '93', '150', '153', '264', '269', '320', '321', '322', 'Cluster 10', '94', '123', '147']

- Ajax1234

Answer 3

在此处查看使用正则表达式的示例

\w+(?: +\w+)?

• \w+ 匹配一个或多个单词字符
• (?: +\w+)? 可选地匹配以下内容
  • + 匹配一个或多个空格
  • \w+ 匹配一个或多个单词字符

在此查看使用的代码

import re

s = "Cluster 7: {4, 15, 21, 28, 33, 35, 43, 47, 53, 57, 59, 66,\n       69, 70, 74, 86, 87, 88, 90, 114, 136, 148, 201,\n       202, 212, 220, 227, 250, 252, 253, 259, 262, 267,\n       270, 282, 296, 318, 319, 323, 326, 341}\nCluster 8: {9, 10, 11, 20, 39, 55, 79, 101, 108, 143, 149,\n       221, 279, 284, 285, 286, 287, 327, 333, 334, 335,\n       336}\nCluster 9: {3, 64, \n3, 93, 150, 153, 264, 269, 320, 321, 322}\nCluster 10: {94, 123, 147}"
print(re.findall(r"\w+(?: +\w+)?", s))

结果：

['Cluster 7', '4', '15', '21', '28', '33', '35', '43', '47', '53', '57', '59', '66', '69', '70', '74', '86', '87', '88', '90', '114', '136', '148', '201', '202', '212', '220', '227', '250', '252', '253', '259', '262', '267', '270', '282', '296', '318', '319', '323', '326', '341', 'Cluster 8', '9', '10', '11', '20', '39', '55', '79', '101', '108', '143', '149', '221', '279', '284', '285', '286', '287', '327', '333', '334', '335', '336', 'Cluster 9', '3', '64', '83', '93', '150', '153', '264', '269', '320', '321', '322', 'Cluster 10', '94', '123', '147']

- ctwheels