从网上找了一圈,没有几个能用的,有代码的,也写的有问题,特此来提供一些能用的方法。
从维基百科下载bz2文件
维基百科bz2下载链接 https://dumps.wikimedia.org/zhwiki/
文件:zhwiki-20240401-pages-articles-multistream1.xml-p1p187712.bz2
将bz2文件转为txt/text格式
将以下代码复制好后,在终端运行:
# 第一个参数是你的bz2文件,第二个参数是你要转换输出的txt文件
python xml2txt.py zhwiki-20240401-pages-articles-multistream.xml.bz2 wiki_2.6GB.zh.txt
# 文件名xml2txt.py(随便起都行)
from __future__ import print_function
import logging
import os.path
import sys
from gensim.corpora import WikiCorpus
if __name__ == '__main__':
program = os.path.basename(sys.argv[0])
logger = logging.getLogger(program)
logging.basicConfig(format='%(asctime)s: %(levelname)s: %(message)s')
logging.root.setLevel(level=logging.INFO)
logger.info("running %s" % ' '.join(sys.argv))
if len(sys.argv) != 3:
print("Using: python process_wiki.py zhwiki-xxx.xml.bz2 wiki.zh.text")
sys.exit(1)
inp, outp = sys.argv[1:3]
space = b" "
i = 0
output = open(outp, 'w', encoding='utf-8')
wiki = WikiCorpus(inp, dictionary={})
for text in wiki.get_texts():
output.write(' '.join(text) + '\n') # type: ignore
i += 1
if i % 10000 == 0:
logger.info(f"Saved {i} articles")
output.close()
logger.info(f"Finished Saved {i} articles")
终端日志输出
2024-07-12 15:41:30,206: INFO: running xml2txt.py ./wiki/zhwiki-20240401-pages-articles-multistream.xml.bz2 D:/DownLoad/wiki/wiki_2.6GB.zh.txt
D:\AnaConda\Lib\site-packages\gensim\utils.py:1333: UserWarning: detected Windows; aliasing chunkize to chunkize_serial
warnings.warn("detected %s; aliasing chunkize to chunkize_serial" % entity)
Building prefix dict from the default dictionary ...
2024-07-12 15:41:35,785: DEBUG: Building prefix dict from the default dictionary ...
Loading model from cache C:\Users\26364\AppData\Local\Temp\jieba.cache
2024-07-12 15:41:35,788: DEBUG: Loading model from cache C:\Users\26364\AppData\Local\Temp\jieba.cache
Loading model cost 0.962 seconds.
2024-07-12 15:41:36,748: DEBUG: Loading model cost 0.962 seconds.
Prefix dict has been built successfully.
2024-07-12 15:41:36,748: DEBUG: Prefix dict has been built successfully.
2024-07-12 15:43:58,926: INFO: Saved 10000 articles
2024-07-12 15:45:54,696: INFO: Saved 20000 articles
2024-07-12 15:47:37,341: INFO: Saved 30000 articles
...... # 这样代表正常工作
繁体2简体:
这里我们使用 opencc:https://github.com/BYVoid/OpenCC/wiki/Download
用法:opencc -i 需要转换的文件路径 -o 转换后的文件路径 -c 配置文件路径
下载好后,配置环境变量,将opencc的exe添加到环境变量。
完成后,打开终端输入:
opencc -i wiki_2.6GB.zh.txt -o test.txt -c D:\\Opencc\\build\\share\\opencc\\t2s.json
# 注意这个-c 后面的配置文件不用自己下载,就在下载opencc的share文件夹里面
到此转换结束。