Switch IAVL Store query to use ics proofs #6324

AdityaSripal · 2020-06-02T19:40:31Z

Description

closes: #6324

Before we can merge this PR, please make sure that all the following items have been
checked off. If any of the checklist items are not applicable, please leave them but
write a little note why.

For contributor use:

Targeted PR against correct branch (see CONTRIBUTING.md)
Linked to Github issue with discussion and accepted design OR link to spec that describes this work.
Code follows the module structure standards.
Wrote unit and integration tests
Updated relevant documentation (docs/) or specification (x/<module>/spec/)
Added relevant godoc comments.
Added a relevant changelog entry to the Unreleased section in CHANGELOG.md
Re-reviewed Files changed in the Github PR explorer

codecov · 2020-06-02T19:43:42Z

Codecov Report

Merging #6324 into master will decrease coverage by 0.14%.
The diff coverage is 17.30%.

@@            Coverage Diff             @@
##           master    #6324      +/-   ##
==========================================
- Coverage   55.68%   55.54%   -0.15%     
==========================================
  Files         451      452       +1     
  Lines       27254    27340      +86     
==========================================
+ Hits        15177    15185       +8     
- Misses      10985    11064      +79     
+ Partials     1092     1091       -1

ethanfrey · 2020-06-02T21:22:24Z

Please ping me when ready for review. Happy to have a look

ethanfrey · 2020-06-03T09:11:57Z

Hey Aditya, I had an idea how to make this a bit more generic so that 95% of the code can be reused to parse the simple merkle proofs as well.

I will give another review, then make a small PR on top of this one, that you can chose to merge in (or not)

ethanfrey

Looks good. Only real issue is the logic for calculating root of the non-existence proog

A few cleanup comments and then I think this is gold.
More than anything this points out other apis that need updates.

ethanfrey · 2020-06-03T10:06:30Z

go.mod

@@ -5,6 +5,8 @@ require (
 	github.com/bgentry/speakeasy v0.1.0
 	github.com/btcsuite/btcd v0.20.1-beta
 	github.com/btcsuite/btcutil v1.0.2
+	github.com/confio/ics23-iavl v0.6.0


As a follow up to this (sometime before tagging the 0.39 sdk), it would be nice to merge this ics23-iavl code into tendermint/iavl, so this can be optimized and integrated in the tendermint codebase.

Do you want to make an issue for that if you agree (I feel odd doing so and cannot label/milestone it)

ethanfrey · 2020-06-03T10:07:10Z

store/iavl/proof.go

+	"github.com/tendermint/tendermint/crypto/merkle"
+)
+
+const ProofOpIAVLCommitment = "iavlstore"


I would use a name like ics23:iavl as this is more clear and as we will be adding other ics23:* proofs that differ only in spec.

ethanfrey · 2020-06-03T10:08:31Z

store/iavl/proof.go

+	return op.Key
+}
+
+func (op CommitmentOp) Run(args [][]byte) ([][]byte, error) {


Please link to the ProofOperator definition in this go-doc.
Which really needs to be much more clear about expected input and output.
But that is another issue (a core team member could gladly open and link)

ethanfrey · 2020-06-03T10:09:55Z

store/iavl/proof.go

+		// and at least one proof must be non-nil
+		root, err := nonexistProof.Nonexist.Left.Calculate()
+		if err != nil {
+			// Left proof may be nil, check right proof


Uh, you should probably check if nonexistProof.Nonexist.Left != nil above rather than calling Calculate on it (which would likely panic) and treating any error as missing

store/iavl/proof.go

ethanfrey · 2020-06-03T10:14:27Z

store/iavl/store.go

+		var commitmentProof *ics23.CommitmentProof
+		var err error
+
+		// Must convert store.Tree to iavl.MutableTree with given version to use in CreateProof


Can you take a look at ics23-iavl to see if I can change this requirement (and accept iavl.ImmutableTree instead). I was just gettng it to work, but in this use case referring to an immutable tree seems better.

Yea since you aren't using any Versioned function, you should be able to just swap out MutableTree for ImmutableTree in the function signatures and everything else should still work.

This will make the code here cleaner

store/rootmulti/proof.go

ethanfrey · 2020-06-03T12:06:03Z

@AdityaSripal please look at #6331 which extends this a bit. You may also want to move the CommitmentOp to a new package, but that is your call... I just wanted to show how to reuse the same code for all ics23.CommitmentProofs

* Make the CommitmentOp generic over all ics23 ProofSpecs, using Type to distinguish * Register SimpleMerkle ics23 proof op as well * Addressed linter issues

colin-axner

nice work! I mostly just reviewed for code hygiene stuff, see suggestions

store/types/proof.go

store/iavl/store.go

Co-authored-by: colin axner <25233464+colin-axner@users.noreply.github.com>

…into aditya/iavl-ics-query

fedekunze

Changes LGTM in general. Can we move the contents of the case "/key": to a private function? 🙏

store/iavl/store.go

store/types/proof.go

Co-authored-by: Federico Kunze <31522760+fedekunze@users.noreply.github.com>

…into aditya/iavl-ics-query

alexanderbez · 2020-06-04T17:54:35Z

store/types/proof.go

+// ProofOp implements ProofOperator interface and converts a CommitmentOp
+// into a merkle.ProofOp format that can later be decoded by CommitmentOpDecoder
+// back into a CommitmentOp for proof verification
+func (op CommitmentOp) ProofOp() merkle.ProofOp {


nit: I would opt to return an error here and allow the caller to decide to panic or not.

We cannot return an error here, this method is satisfying the ProofOperator interface

Yikes, I would consider revising that interface if possible. Not a good UX IMHO. Ideally, you should always be able to bubble up errors as much as possible and allow callers to handle it.

Hmm... yea that would definitely require a separate discussion with TM team and a separate PR since this is a tendermint interface. Don't think a panic is the worst idea here since it should be an invariant that any well-constructed ProofOperator should successfully return a merkle.ProofOp here

store/types/proof.go

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

tac0turtle · 2020-06-05T17:02:03Z

@AdityaSripal if this is good to go, could you add the automerge label

switch iavl store to use ics proof

5f0a378

AdityaSripal requested review from aaronc, alessio, alexanderbez, clevinson, fedekunze and jgimeno as code owners June 2, 2020 19:40

AdityaSripal mentioned this pull request Jun 2, 2020

Make store.Query use ics23-proofs #6315

Closed

8 tasks

AdityaSripal added 2 commits June 2, 2020 15:47

fix proofs to return for correct height

3f38699

appease linter

6eae708

Register commitment op correctly

aad1ac7

AdityaSripal added R4R C:Store labels Jun 2, 2020

ethanfrey reviewed Jun 3, 2020

View reviewed changes

ethanfrey mentioned this pull request Jun 3, 2020

Make CommitmentOp generic over all ics23 specs #6331

Merged

8 tasks

ethanfrey and others added 3 commits June 3, 2020 13:18

Make CommitmentOp generic over all ics23 specs (#6331)

09519aa

* Make the CommitmentOp generic over all ics23 ProofSpecs, using Type to distinguish * Register SimpleMerkle ics23 proof op as well * Addressed linter issues

move commitment proof to types

17b6f21

register commitment op

1dae554

colin-axner reviewed Jun 3, 2020

View reviewed changes

store/types/proof.go Outdated Show resolved Hide resolved

store/types/proof.go Outdated Show resolved Hide resolved

store/types/proof.go Outdated Show resolved Hide resolved

store/types/proof.go Show resolved Hide resolved

store/iavl/store.go Outdated Show resolved Hide resolved

AdityaSripal and others added 3 commits June 3, 2020 19:08

Apply suggestions from code review

056fcfd

Co-authored-by: colin axner <25233464+colin-axner@users.noreply.github.com>

address review comments:

1e13c9f

Merge branch 'aditya/iavl-ics-query' of github.com:cosmos/cosmos-sdk …

cc9aa0d

…into aditya/iavl-ics-query

fedekunze suggested changes Jun 4, 2020

View reviewed changes

store/iavl/store.go Outdated Show resolved Hide resolved

store/types/proof.go Outdated Show resolved Hide resolved

store/types/proof.go Outdated Show resolved Hide resolved

store/types/proof.go Show resolved Hide resolved

AdityaSripal and others added 2 commits June 4, 2020 12:02

Apply suggestions from code review

28c344c

Co-authored-by: Federico Kunze <31522760+fedekunze@users.noreply.github.com>

allow proofs against empty store

cf6a061

AdityaSripal added 3 commits June 4, 2020 13:23

Merge branch 'aditya/iavl-ics-query' of github.com:cosmos/cosmos-sdk …

cf3778d

…into aditya/iavl-ics-query

address review comments

c715058

Merge branch 'master' into aditya/iavl-ics-query

9f39846

alexanderbez approved these changes Jun 4, 2020

View reviewed changes

Update store/types/proof.go

f59df00

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

fedekunze approved these changes Jun 4, 2020

View reviewed changes

fedekunze and others added 3 commits June 4, 2020 14:59

Merge branch 'master' into aditya/iavl-ics-query

4e7d95e

add changelog

2a471ec

Merge branch 'master' into aditya/iavl-ics-query

9d0292a

Merge branch 'master' into aditya/iavl-ics-query

71664bc

AdityaSripal added the A:automerge Automatically merge PR once all prerequisites pass. label Jun 8, 2020

Merge branch 'master' into aditya/iavl-ics-query

52128a8

fedekunze merged commit d9e1497 into master Jun 8, 2020

fedekunze deleted the aditya/iavl-ics-query branch June 8, 2020 19:02

AdityaSripal mentioned this pull request Jun 15, 2020

WIP: Fix ICS23-Proofs #6178

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Switch IAVL Store query to use ics proofs #6324

Switch IAVL Store query to use ics proofs #6324

AdityaSripal commented Jun 2, 2020 •

edited

Loading

codecov bot commented Jun 2, 2020 •

edited

Loading

ethanfrey commented Jun 2, 2020

ethanfrey commented Jun 3, 2020

ethanfrey left a comment

ethanfrey Jun 3, 2020

ethanfrey Jun 3, 2020

ethanfrey Jun 3, 2020

ethanfrey Jun 3, 2020

ethanfrey Jun 3, 2020

AdityaSripal Jun 3, 2020

ethanfrey commented Jun 3, 2020

colin-axner left a comment

fedekunze left a comment

alexanderbez Jun 4, 2020

AdityaSripal Jun 4, 2020

alexanderbez Jun 4, 2020 •

edited

Loading

AdityaSripal Jun 4, 2020

tac0turtle commented Jun 5, 2020

Switch IAVL Store query to use ics proofs #6324

Switch IAVL Store query to use ics proofs #6324

Conversation

AdityaSripal commented Jun 2, 2020 • edited Loading

Description

codecov bot commented Jun 2, 2020 • edited Loading

Codecov Report

ethanfrey commented Jun 2, 2020

ethanfrey commented Jun 3, 2020

ethanfrey left a 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

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ethanfrey commented Jun 3, 2020

colin-axner left a comment

Choose a reason for hiding this comment

fedekunze left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

alexanderbez Jun 4, 2020 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

tac0turtle commented Jun 5, 2020

AdityaSripal commented Jun 2, 2020 •

edited

Loading

codecov bot commented Jun 2, 2020 •

edited

Loading

alexanderbez Jun 4, 2020 •

edited

Loading