Repository navigation
Conversation
There was a problem hiding this comment.
Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.
An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Follows flow-sync #101,
spec/crypto.md, Ed25519 checks: the keyAand the signature'sRmust each be a point of order exactly L. BouncyCastle'sEd25519.validatePublicKeyFullchecks exactly that, soCurves.ktnow uses it on both and drops its own small-order list and canonical check, which the method covers.isValidSigningKey, which guards keys entering the roster, uses the same check. An honest signer never produces what this refuses, so nothing changes for existing groups.verifyinCurves.ktis the only place that verifies an Ed25519 signature, so all its callers are covered: roster entries (Roster.kt), log objects (Session.verified,ObjectCheck.admits), live messages (LiveEnvelopes) and the LAN hello (LanLink, both sides). Snapshot blobs and the walk down from acutare checked by hash, not by signature, as the spec has it.Vectors:
ed25519.jsonandroster.jsoncome byte for byte from flow-sync #101 (01050b2), with their two README rows.VectorsTestnow also refusesmixedOrderKeys. The README's other rows are left as they were.Related issue
None. Follows flow-sync #101, which should land first.
Change type
Validation
All Gradle commands ran with
--offline -Dorg.gradle.jvmargs="-Xmx4096m -XX:MaxMetaspaceSize=1024m -Dfile.encoding=UTF-8 -Djava.net.preferIPv6Addresses=true".:app:testGithubDebugUnitTest --tests 'io.github.aedev.flow.sync2.*': 301 tests, 0 failed.ktlintCheck: exit 0.javaponbcprov-jdk18on-1.86.jar, the version ingradle/libs.versions.toml, showsvalidatePublicKeyFull(byte[], int). No new dependency.Before the change,
VectorsTest.ed25519_strict_verification(identity-r) andRosterTest.invalid_entries_are_rejected_for_the_named_reason(mixed-order-key) failed on the new vectors.Each of these made tests fail when tried:
Rnot checked (2 tests),Aonly partly checked (2),Anot checked at verify (1)../gradlew :app:assembleGithubDebug: not run locally, CI builds it../gradlew :app:testGithubDebugUnitTest: full suite not run locally, CI runs it.I ran any additional flavor-specific build or test tasks affected by this change.
I manually tested the affected behavior on an Android device or emulator: not done.
Test device and Android version: none.
Risk and compatibility
Cost, measured on a Mac on the JVM, 1 KiB message, 3 runs of 3,000 each. Not measured on a phone, so these say nothing about a phone's absolute numbers:
Ed25519.verifyalonevalidatePublicKeyFullonAvalidatePublicKeyFullonRCurves.verifynowSo one verify costs a little over twice what it did. An object is verified twice, once when it's kept and once when it's read. Downloading and reading a log of 2,000 objects of 866 bytes took 0.9 to 1.2 s before and 1.1 to 1.3 s after on the same Mac, in 3 runs each, which is noisy. Larger objects spend a smaller share on this. No number here is a phone's.
No nightly note: nothing an honest group sees changes.
roster.jsonand the vector README also change in the other open PRs that follow flow-sync #103 and #1831.