This repository has been archived by the owner on Aug 14, 2019. It is now read-only.
+ob overhaul #1110
Merged
+ob overhaul #1110
Conversation
This file contains 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
'ob', a core that implemented a permutation cipher for planet-sized atoms, contained a critical bug that allowed multiple inputs to encipher to the same output. This created collisions between some planet-sized @p values. The rewritten version corrects this bug, and also rewrites +ob to more clearly match both 1) the paper that the algorithm comes from, and 2) the parallel implementation in urbit-ob that has gone through an exhaustive test. This also adjusts two calls to 'ob' elsewhere in hoon.hoon such that they match the updated core.
This allows us to test the +ob functions exhaustively on smaller input spaces.
We can test +ob exhaustively over a smaller input space; if it behaves correctly, that gives us much greater assurance that the implementation is correct for the full planet space.
|
@Fang- nice, I'll try whipping up a pill. |
belisarius222
suggested changes
Mar 18, 2019
(The cipher's domain was documented incorrectly.)
Added a few previously-known-to-conflict cases to tests that were missing them.
|
After scratching my head for awhile as to why a pill wouldn't work, I realised I was using the one in Using the correct one persisted my changes to dojo: |
|
fwiw @jtobin I made the exact same mistake when I was learning to make modifications to hoon.hoon. It took me way too long to realize what I was doing wrong. |
I believe =/ is the superior choice and is worth standardising on.
Indent lengthy subexpressions after using =/.
joemfb
suggested changes
Mar 19, 2019
|
I'll merge this in -- don't want to sidestep @belisarius222, but I reckon the changes meet his approval, so I'll be bold. |
Closed
Merged
Closed
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
(Resolves #1105)
+ob, a core that implemented a permutation cipher for planet-sized atoms, contained a critical bug that allowed multiple inputs to encipher to the same output. This created collisions between some planet-sized@pvalues.The rewritten version corrects the bug, and also rewrites
+obto more clearly match both 1) the paper that the algorithm comes from, and 2) the parallel implementation in urbit-ob that has gone through a full-scale exhaustive test.The rewritten version is also itself easier to test; this PR adds a small-scale exhaustive test, as well as small battery of sanity checks on
+obinternals (including on@pvalues previously known to collide).NOTE: One thing I'm observing here is that dojo may be lagging behind on changes to hoon.hoon. For example, the following in
tests/sys/hoon/auras.hoonpasses:but in dojo, I still get collisions:
Should I trust the test suite here? I'd like to figure out what's going on, regardless.