正则表达式捕获不包含目录链接的所有文件路径
我正在尝试编写一个正则表达式来匹配具有以下特征的文件路径：
- 不包含
/./或/../ - 必须嵌套至少一个子目录深
/tmp/media内的内容 - 必须以
.log结尾
\/tmp\/media\/(?!.*\.?\.\/)+(?:.*\.log)
这是我想要的结果（如果字符串不匹配，我引用不满足的要求）：
/tmp/media/log.log //NO-MATCH (2)
/tmp/media/test/log.log //MATCH
/tmp/media/../log.log //NO-MATCH (1)
/tmp/media/./log.log //NO-MATCH (1)
/tmp/media/test/../log.log //NO-MATCH (1)
/tmp/media/../test/log.log //NO-MATCH (1)
/tmp/media/.t/log.log //MATCH
/tmp/media/.../log.log //MATCH
/tmp/log.log //NO-MATCH (2)
/tmp/media/test/log.notlog //NO-MATCH (3)
/tmp/media/test/./log.log //NO-MATCH (1)
我已阅读 这个问题，并成功实现了一些答案，但 /tmp/media/log.log 是当我不希望它匹配时仍然匹配。我怀疑这是因为 /media/ 和 log.log 之间的空字符串以某种方式满足 (?!.*\.?\.\/).
谁能解释我做错了什么并帮助我纠正我的表达？
I am trying to write a regular expression to match file paths with the following characteristics:
- do not contain
/./or/../ - must be nested at least one subdirectory deep within
/tmp/media - must end in
.log
Here is what I have so far:
\/tmp\/media\/(?!.*\.?\.\/)+(?:.*\.log)
Here are my desired results (if the string doesn't match, I cite the requirement which is not satisfied):
/tmp/media/log.log //NO-MATCH (2)
/tmp/media/test/log.log //MATCH
/tmp/media/../log.log //NO-MATCH (1)
/tmp/media/./log.log //NO-MATCH (1)
/tmp/media/test/../log.log //NO-MATCH (1)
/tmp/media/../test/log.log //NO-MATCH (1)
/tmp/media/.t/log.log //MATCH
/tmp/media/.../log.log //MATCH
/tmp/log.log //NO-MATCH (2)
/tmp/media/test/log.notlog //NO-MATCH (3)
/tmp/media/test/./log.log //NO-MATCH (1)
I have read this question, and successfully implemented some of the answer, but /tmp/media/log.log is still matching when I do not want it to. I suspect it is because the empty string between /media/ and log.log somehow satisfies (?!.*\.?\.\/).
Can anyone explain what I'm doing wrong and help me correct my expression?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（3）
您可以尝试以下操作：
rubular 链接
You can try this:
rubular link
有必要成为正则表达式吗？正如您提供的链接中所建议的，您可以使用 readlink。另外，在对其他约束进行正则表达式匹配之前检查
/./和/../会更容易 - 会让你的代码更干净。Is it necessary to be a regexp? As suggested in the link you provided, you can use readlink. Also, it would be easier to just check for
/./and/../before doing the regex matching for other constraints - would make your code cleaner.您应该首先真正规范您的路径。这将避免尝试阻止正则表达式本身中的
.和..目录。完成此操作后，正则表达式将变为：在不知道您正在使用哪种语言进行编程的情况下，我无法建议路径规范化机制。
You should really canonize your paths first. This will avoid attempting to block
.and..directories in the regex itself. Once you have done that, the regex becomes:Without knowing which language you are programming in, I cannot suggest a path-canonization mechanism.