-
Notifications
You must be signed in to change notification settings - Fork 1.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Eliminate Zip-Slip vulnerability #908
Conversation
Seems like Kokoro won't run when not merging into |
5742f9f
to
5d3d2bc
Compare
Path entryPath = destination.resolve(entry.getName()); | ||
|
||
String canonicalTarget = entryPath.toFile().getCanonicalPath(); | ||
if (!canonicalTarget.startsWith(canonicalDestination + File.separator)) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could this be done with NIO? Like maybe entryPath.toAbsolutePath().startsWith(destination.toAbsolutePath())
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
A canonical path is an absolute path, but the opposite is not true. For example,
System.out.println(Paths.get("/temp1/../temp/test.txt").toAbsolutePath());
prints /temp1/../temp/test.txt
.
The code here is what https://snyk.io/research/zip-slip-vulnerability actually suggests, so I think it is safe to stick with the canonical path.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Okay, though it feels off having to use the deprecated java.io.File
API to account for a security vulnerability. Perhaps we could use Path#normalize
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I still don't think Path.normalize()
would work. A canonical path is a unique path for any given file; there can ever be only one canonical path. Computing a canonical path may involve "resolving symbolic links (on UNIX platforms), and converting drive letters to a standard case (on Microsoft Windows platforms)" according to its Javadoc. So I think it's safe to stick with the canonical path approach.
* Add unzip util method * Eliminate Zip-Slip vulnernability (#908)
On top of #906.
https://snyk.io/research/zip-slip-vulnerability