chore: simplify ADR-028 and address.Module #14017

robert-zaremba · 2022-11-24T23:40:06Z

Description

Author Checklist

All items are required. Please add a note to the item if the item is not applicable and
please add links to any relevant follow up issues.

I have...

Reviewers Checklist

All items are required. Please add a note if the item is not applicable and please add
your handle next to the items reviewed if you only reviewed selected items.

I have...

confirmed the correct type prefix in the PR title
confirmed ! in the type prefix if API or client breaking change
confirmed all author checklist items have been addressed
reviewed state machine logic
reviewed API design and naming
reviewed documentation is accurate
reviewed tests and test coverage
manually tested (if applicable)

robert-zaremba · 2022-11-24T23:44:04Z

types/address/hash.go

-func Module(moduleName string, key []byte) []byte {
+// is constructed from a module name and a sequence of derivation keys (at least one
+// derivation key must be provided).
+func Module(moduleName string, derivationKey []byte, derivationKeys ...[]byte) []byte {


I was thinking to "merge" derivationKey into derivationKeys, but then we would need to panic if someone will provide an empty list.

I think as is is great, requires at least 1 derivationKey implicitly which is communicated to the caller.

I don't see any reason to require at least one derivation key for modules. This is inconsistent with what we've discussed for ADR 033. There is always the root module address with no derivation keys.

Suggested change

func Module(moduleName string, derivationKey []byte, derivationKeys ...[]byte) []byte {

func Module(moduleName string, derivationKeys ...[]byte) []byte {

wouldn't this conflict with the existing authtypes.NewModuleAddress?

No. That uses 20 byte addresses

My preference is to keep backwards compatibility of existing root module accounts.

func Module(moduleName string, derivationKeys ...[]byte) []byte { if len(derivationKeys) == 0 { return authtypes.NewModuleAddress() } else { /* do like current code with successive derivations */ } }

so it's:

20 byte address for root module accounts

32 byte addresses for sub/derived module accounts

So we need to make decision.

Up again, I'm going to update this PR by implementing #14017 (comment) (which is what @AmauryM seconded: #14017 (comment))

julienrbrt

Lgtm!

docs/architecture/adr-028-public-key-addresses.md

Co-authored-by: Julien Robert <julien@rbrt.fr>

robert-zaremba · 2022-11-25T11:02:46Z

Adding 0.47 backport.

types/address/hash.go

odeke-em

Thank you for this change @robert-zaremba, just some suggestions, please take a look. Thank you Julien for the co-review.

types/address/hash.go

odeke-em · 2022-11-26T23:43:08Z

types/address/hash.go

-func Module(moduleName string, key []byte) []byte {
+// is constructed from a module name and a sequence of derivation keys (at least one
+// derivation key must be provided).
+func Module(moduleName string, derivationKey []byte, derivationKeys ...[]byte) []byte {


I think as is is great, requires at least 1 derivationKey implicitly which is communicated to the caller.

x/auth/types/credentials.go

types/address/hash_test.go

aaronc · 2022-12-05T16:44:08Z

types/address/hash.go

-func Module(moduleName string, key []byte) []byte {
+// is constructed from a module name and a sequence of derivation keys (at least one
+// derivation key must be provided).
+func Module(moduleName string, derivationKey []byte, derivationKeys ...[]byte) []byte {


I don't see any reason to require at least one derivation key for modules. This is inconsistent with what we've discussed for ADR 033. There is always the root module address with no derivation keys.

Suggested change

func Module(moduleName string, derivationKey []byte, derivationKeys ...[]byte) []byte {

func Module(moduleName string, derivationKeys ...[]byte) []byte {

aaronc · 2022-12-05T16:45:07Z

@AmauryM would be great to get your review of this too.

docs/architecture/adr-028-public-key-addresses.md

x/auth/types/credentials.go

Co-authored-by: Aleksandr Bezobchuk <alexanderbez@users.noreply.github.com>

robert-zaremba · 2023-01-04T18:57:27Z

I'm fine with address.Module reverting to the pre-ADR 028 20-byte addresses (compatible with existing module accounts) when there are no derivation keys.

Personally, I would prefer to have only the "full" ADR-28 addresses for root module accounts. But simplification and compatibility is more important here.

aaronc · 2023-01-04T19:01:27Z

I'm fine with address.Module reverting to the pre-ADR 028 20-byte addresses (compatible with existing module accounts) when there are no derivation keys.

Personally, I would prefer to have only the "full" ADR-28 addresses for root module accounts. But simplification and compatibility is more important here.

For security purposes? All root accounts should be initialized on startup

robert-zaremba · 2023-01-04T19:05:37Z

For security purposes?

Mostly for having one general solution (with clear domains - what ADR-028 is mostly about).

I think security is OK here, because we have that module name hash. The concern is if someone is doing something weird and creating a module name which is a hash and could collide with an existing public key hash. Or generates multiple module accounts rather than deriving.

x/auth/types/credentials.go

sonarqubecloud · 2023-01-09T22:34:39Z

[Cosmos SDK] Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities
0 Security Hotspots
1 Code Smell

No Coverage information
No Duplication information

aaronc

Happy to give my pre-approval on this now but I think we should sort out what to do for the root module hash. Maybe on the next community call as you mentioned @robert-zaremba

Co-authored-by: Aaron Craelius <aaron@regen.network>

julienrbrt · 2023-01-12T16:58:17Z

I believe it closes #13782 too right?

julienrbrt · 2023-01-12T17:41:36Z

Can we add a changelog? Then, given the discussion from the community call, we can add automerge.

adu-crypto

It seems a little late to see this pr since it would be merged soon.
After reading through the pr, I think this is mainly talking about the derivation of module type accounts. But I got some questions about the change and would appreciate it if anyone could explain to me:

Could this possibly bring any consensus-breaking change?
what are the main benefits of this new pattern of module account derivation change

adu-crypto · 2023-01-13T02:59:32Z

docs/architecture/adr-028-public-key-addresses.md


 ```go
-func Module(moduleName string, key []byte) []byte{


so currently we are not using this function to generate root module account, instead we are using authtypes.NewModuleAddress(modulenName), right?

Exactly, now authtypes.NewModuleAddress(modulenName) falls back to address.Module now, but, the behavior stays the same as authtypes.NewModuleAddress(modulenName): https://github.com/cosmos/cosmos-sdk/pull/14017/files#diff-6cfded83a08be06034f746e83ffa6bbd6c35430cd8ac05015e6bd942c7da1030R73-R76

docs/architecture/adr-028-public-key-addresses.md

Co-authored-by: Julien Robert <julien@rbrt.fr> Co-authored-by: Aleksandr Bezobchuk <alexanderbez@users.noreply.github.com> Co-authored-by: Amaury <1293565+amaurym@users.noreply.github.com> Co-authored-by: Aaron Craelius <aaron@regen.network> Co-authored-by: Marko <marbar3778@yahoo.com> (cherry picked from commit fd3bac0) # Conflicts: # CHANGELOG.md

Co-authored-by: Robert Zaremba <robert@zaremba.ch> Co-authored-by: Julien Robert <julien@rbrt.fr>

chore: simplify ADR-028 and address.Module

fb63a19

robert-zaremba added C:Types T: ADR An issue or PR relating to an architectural decision record labels Nov 24, 2022

robert-zaremba requested a review from a team as a code owner November 24, 2022 23:40

github-actions bot added C:x/auth C:x/group Type: ADR labels Nov 24, 2022

formatting

a76c8e9

robert-zaremba commented Nov 24, 2022

View reviewed changes

julienrbrt approved these changes Nov 25, 2022

View reviewed changes

docs/architecture/adr-028-public-key-addresses.md Outdated Show resolved Hide resolved

docs/architecture/adr-028-public-key-addresses.md Outdated Show resolved Hide resolved

docs/architecture/adr-028-public-key-addresses.md Outdated Show resolved Hide resolved

julienrbrt assigned julienrbrt, amaury1093 and aaronc Nov 25, 2022

robert-zaremba and others added 2 commits November 25, 2022 12:00

Apply suggestions from code review

8758fb7

Co-authored-by: Julien Robert <julien@rbrt.fr>

update example 2

a4feb8c

robert-zaremba added the backport/v0.47.x PR scheduled for inclusion in the v0.47's next stable release label Nov 25, 2022

Merge branch 'main' into robert/update-adr28

a69b7ab

julienrbrt reviewed Nov 25, 2022

View reviewed changes

types/address/hash.go Outdated Show resolved Hide resolved

odeke-em reviewed Nov 26, 2022

View reviewed changes

fedekunze reviewed Nov 27, 2022

View reviewed changes

types/address/hash_test.go Outdated Show resolved Hide resolved

aaronc requested changes Dec 5, 2022

View reviewed changes

robert-zaremba added 2 commits December 5, 2022 20:10

review: typos

bcd66d1

Merge remote-tracking branch 'origin/main' into robert/update-adr28

24dc097

alexanderbez reviewed Dec 6, 2022

View reviewed changes

docs/architecture/adr-028-public-key-addresses.md Outdated Show resolved Hide resolved

docs/architecture/adr-028-public-key-addresses.md Outdated Show resolved Hide resolved

x/auth/types/credentials.go Outdated Show resolved Hide resolved

robert-zaremba and others added 4 commits December 8, 2022 22:22

Apply suggestions from code review

50580f0

Co-authored-by: Aleksandr Bezobchuk <alexanderbez@users.noreply.github.com>

Merge remote-tracking branch 'origin/main' into robert/update-adr28

cf22468

update NewModuleCredentialto return an error

8e9ef45

changed NewModuleCredential interface

8882546

robert-zaremba added 3 commits January 4, 2023 19:46

linter

8354fc5

update to support NewModuleCredential for Root Module

34768e4

Merge remote-tracking branch 'origin/main' into robert/update-adr28

d46bf3c

robert-zaremba requested a review from aaronc January 4, 2023 18:55

Merge branch 'main' into robert/update-adr28

a2c6398

aaronc reviewed Jan 9, 2023

View reviewed changes

x/auth/types/credentials.go Outdated Show resolved Hide resolved

aaronc approved these changes Jan 10, 2023

View reviewed changes

Update x/auth/types/credentials.go

c6542a7

Co-authored-by: Aaron Craelius <aaron@regen.network>

alexanderbez approved these changes Jan 11, 2023

View reviewed changes

Merge branch 'main' into robert/update-adr28

ae14791

julienrbrt added the A:automerge Automatically merge PR once all prerequisites pass. label Jan 12, 2023

adu-crypto reviewed Jan 13, 2023

View reviewed changes

docs/architecture/adr-028-public-key-addresses.md Outdated Show resolved Hide resolved

tac0turtle and others added 2 commits January 16, 2023 14:04

Merge branch 'main' into robert/update-adr28

8d769f9

add changelog and comment

7f095f1

julienrbrt enabled auto-merge (squash) January 16, 2023 13:27

Merge branch 'main' into robert/update-adr28

7bea48a

julienrbrt merged commit fd3bac0 into main Jan 16, 2023

julienrbrt deleted the robert/update-adr28 branch January 16, 2023 13:49

mergify bot mentioned this pull request Jan 16, 2023

chore: simplify ADR-028 and address.Module (backport #14017) #14632

Merged

julienrbrt mentioned this pull request Jan 16, 2023

ModuleAccount addresses don't follow ADR-028 #13782

Closed

alexanderbez pushed a commit that referenced this pull request Jan 16, 2023

chore: simplify ADR-028 and address.Module (backport #14017) (#14632)

8df686e

Co-authored-by: Robert Zaremba <robert@zaremba.ch> Co-authored-by: Julien Robert <julien@rbrt.fr>

0xstepit mentioned this pull request Jul 1, 2024

chore(cachemulti): add Copy method evmos/cosmos-sdk#52

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

chore: simplify ADR-028 and address.Module #14017

chore: simplify ADR-028 and address.Module #14017

robert-zaremba commented Nov 24, 2022 •

edited

Loading

robert-zaremba Nov 24, 2022

odeke-em Nov 26, 2022

aaronc Dec 5, 2022

robert-zaremba Dec 5, 2022

aaronc Dec 5, 2022

amaury1093 Dec 6, 2022 •

edited

Loading

robert-zaremba Dec 8, 2022

robert-zaremba Dec 16, 2022

robert-zaremba Dec 19, 2022

amaury1093 Jan 3, 2023

julienrbrt left a comment

robert-zaremba commented Nov 25, 2022

odeke-em left a comment

odeke-em Nov 26, 2022

aaronc Dec 5, 2022

aaronc commented Dec 5, 2022

robert-zaremba commented Jan 4, 2023

aaronc commented Jan 4, 2023

robert-zaremba commented Jan 4, 2023

sonarqubecloud bot commented Jan 9, 2023

aaronc left a comment

julienrbrt commented Jan 12, 2023

julienrbrt commented Jan 12, 2023

adu-crypto left a comment

adu-crypto Jan 13, 2023

julienrbrt Jan 16, 2023 •

edited

Loading

	func Module(moduleName string, derivationKey []byte, derivationKeys ...[]byte) []byte {
	func Module(moduleName string, derivationKeys ...[]byte) []byte {

chore: simplify ADR-028 and address.Module #14017

chore: simplify ADR-028 and address.Module #14017

Conversation

robert-zaremba commented Nov 24, 2022 • edited Loading

Description

Author Checklist

Reviewers Checklist

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

amaury1093 Dec 6, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

julienrbrt left a comment

Choose a reason for hiding this comment

robert-zaremba commented Nov 25, 2022

odeke-em left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

aaronc commented Dec 5, 2022

robert-zaremba commented Jan 4, 2023

aaronc commented Jan 4, 2023

robert-zaremba commented Jan 4, 2023

sonarqubecloud bot commented Jan 9, 2023

aaronc left a comment

Choose a reason for hiding this comment

julienrbrt commented Jan 12, 2023

julienrbrt commented Jan 12, 2023

adu-crypto left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

julienrbrt Jan 16, 2023 • edited Loading

Choose a reason for hiding this comment

robert-zaremba commented Nov 24, 2022 •

edited

Loading

amaury1093 Dec 6, 2022 •

edited

Loading

julienrbrt Jan 16, 2023 •

edited

Loading