Merged
Conversation
rudolfochrist
requested changes
Feb 15, 2024
| @@ -0,0 +1,135 @@ | |||
| ;; copied from tarhash.lisp from quicklisp-controller: https://github.com/quicklisp/quicklisp-controller | |||
| ;; changed to use openssl to compute sha1 digest rather than depending on ironclad | |||
|
|
|||
Owner
There was a problem hiding this comment.
Can you add the license information here as a comment as well?
https://github.com/quicklisp/quicklisp-controller/blob/master/LICENSE
rudolfochrist
approved these changes
Feb 15, 2024
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
After exchange on reddit when ql-https release was posted there, I realize the sha1 provided doesn't come from git as I had thought for some reason, but is calculated by iterating over the files in the tgz sorted by filename and feeding each into the sha1 digest. This is done by quicklisp-controller using sha1 from ironclad, but as we are quicklisp client and not server side script, we shouldn't use dependencies so I took the code from quicklisp-controller and used
openssl dgst -sha1instead of using ironclad. sha1 is much more expensive to generate collisions for than md5, so this should provide a meaningful increase in security in case quicklisp server is compromised, though it is still possible to generate collisions. It'd be great if quicklisp dist provided a secure hash like sha256 but I think this is the best we can check for now