starknet_committer: compute and store accessed keys digest

[ArielElp]

No description provided.

[reviewable-StarkWare]

This change is 

[ArielElp]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• apollo_committer: add new request handler #14003
• starknet_committer: delete patricia paths on reverts #14002
• starknet_committer,apollo_committer: add timers for fetch patricia paths #14189
• starknet_committer: underlying logic of the new read witnesses and commit endpoint #14001
• starknet_committer: extract metadata construction to a function #14000
• starknet_committer: extract commit plan to commit_or_load #13999
• starknet_committer: read and write patricia paths to index db #13998
• starknet_committer: db keys for proofs and accessed keys digest #13997
• apollo_committer_types: add patricia proofs request and response types #13996
• starknet_committer: add patricia paths forest reader/writer traits #13995
• starknet_committer: define StarknetForestProofs serde #13994
• starknet_committer: compute and store accessed keys digest #13993 👈 (View in Graphite)
• starknet_committer: add utility that returns sorted leaf indices #13992
• starknet_committer: abstract fetch_all_patricia_paths #13991
• starknet_patricia_storage: two layer storage #13990
• starknet_committer: move fetch_patricia_paths tests #13960
• starknet_committer: fetch patricia paths tests #13959
• starknet_committer: support both layouts in fetch_patricia_paths #13958
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[cursor]

PR Summary

Low Risk
Additive, feature-gated hashing helper with no changes to default builds or existing serialization paths.

Overview
Adds an os_input Cargo feature on starknet_committer that pulls in optional blake2 / digest and exposes accessed_keys_digest in serde_db_utils.

That function builds a canonical, length-prefixed byte encoding of a SortedLeavesRequest (classes trie, contracts trie, then per-contract storage leaves with sorted contract keys) and returns a BLAKE2s-256 32-byte digest. This gives a stable fingerprint of accessed Patricia leaf indices for OS-input / witness-related flows without changing default builds (everything is behind cfg(feature = "os_input")).

^{Reviewed by Cursor Bugbot for commit 5a5ff22. Bugbot is set up for automated code reviews on this repo. Configure here.}

[yoavGrs]

@yoavGrs reviewed 2 files and all commit messages, and made 1 comment.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on ArielElp).

---

crates/starknet_committer/src/db/serde_db_utils.rs line 60 at r1 (raw file):

) -> [u8; 32] {
    let mut class_hashes: Vec<ClassHash> = class_hashes.iter().copied().collect();
    class_hashes.sort();

In the previous PR you sort it. Define a "sorted" struct?

Code quote:

class_hashes.sort();

[ArielElp]

@ArielElp partially reviewed 2 files and made 2 comments.
Reviewable status: 0 of 3 files reviewed, 2 unresolved discussions (waiting on yoavGrs).

---

crates/starknet_committer/src/db/serde_db_utils.rs line 39 at r2 (raw file):

Previously, yoavGrs wrote…

Document the structure of the payload

Done.

---

crates/starknet_committer/src/db/serde_db_utils.rs line 72 at r2 (raw file):

Previously, yoavGrs wrote…

Avoid using as.

Done.

[yoavGrs]

@yoavGrs reviewed 2 files and all commit messages, made 1 comment, and resolved 2 discussions.
Reviewable status: 2 of 3 files reviewed, all discussions resolved (waiting on ArielElp).

[dorimedini-starkware]

@dorimedini-starkware reviewed 3 files and all commit messages, and made 4 comments.
Reviewable status: all files reviewed, 4 unresolved discussions (waiting on ArielElp).

---

crates/starknet_committer/Cargo.toml line 15 at r4 (raw file):

[dependencies]
blake2 = { workspace = true, optional = true }
digest = { workspace = true, optional = true }

excuse me ser we sort dependencies here

Code quote:

blake2 = { workspace = true, optional = true }
digest = { workspace = true, optional = true }

---

crates/starknet_committer/src/db/serde_db_utils.rs line 47 at r4 (raw file):

///    (already sorted within the contract).
#[cfg(feature = "os_input")]
pub fn accessed_keys_digest(sorted: &SortedLeavesRequest<'_>) -> [u8; 32] {

consider making this a method of SortedLeavesRequest

Code quote:

pub fn accessed_keys_digest(sorted: &SortedLeavesRequest<'_>) -> [u8; 32] {

---

crates/starknet_committer/src/db/serde_db_utils.rs line 63 at r4 (raw file):

    let mut contract_indices: Vec<NodeIndex> = sorted.storage_sorted.keys().copied().collect();
    contract_indices.sort_unstable();

unstable? why?
also, why do you need to sort at all? isn't that what SortedLeavesRequest is for?
if it's because of the HashMap field, consider using a BTreeMap field to get well-defined key order in the struct itself

Code quote:

contract_indices.sort_unstable();

---

crates/starknet_committer/src/db/serde_db_utils.rs line 81 at r4 (raw file):

#[cfg(feature = "os_input")]
fn encode_usize(n: usize) -> [u8; 4] {
    u32::try_from(n).expect("accessed leaf count exceeds u32::MAX").to_be_bytes()

mmmmmm why not u64? and return [u8; 8]? it doesn't matter much, and we won't have to answer the question "will we ever read 4B leaves"?

Code quote:

u32::try_from(n).expect("accessed leaf count exceeds u32::MAX").to_be_bytes()

[ArielElp]

@ArielElp reviewed all commit messages and made 4 comments.
Reviewable status: all files reviewed, 4 unresolved discussions (waiting on dorimedini-starkware).

---

crates/starknet_committer/Cargo.toml line 15 at r4 (raw file):

Previously, dorimedini-starkware wrote…

excuse me ser we sort dependencies here

Done.

---

crates/starknet_committer/src/db/serde_db_utils.rs line 47 at r4 (raw file):

Previously, dorimedini-starkware wrote…

consider making this a method of SortedLeavesRequest

IMO given that it's db-serialization related, it's slightly better placed here, but won't insist if you prefer moving.

---

crates/starknet_committer/src/db/serde_db_utils.rs line 63 at r4 (raw file):

Previously, dorimedini-starkware wrote…

unstable? why?
also, why do you need to sort at all? isn't that what SortedLeavesRequest is for?
if it's because of the HashMap field, consider using a BTreeMap field to get well-defined key order in the struct itself

AI decision, it's supposed to be marginally faster, and you know there are no duplications here (it's the keys from a HashMap).

---

crates/starknet_committer/src/db/serde_db_utils.rs line 81 at r4 (raw file):

Previously, dorimedini-starkware wrote…

mmmmmm why not u64? and return [u8; 8]? it doesn't matter much, and we won't have to answer the question "will we ever read 4B leaves"?

Done.

[dorimedini-starkware]

@dorimedini-starkware reviewed 2 files and all commit messages, made 1 comment, and resolved 3 discussions.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on ArielElp).

---

crates/starknet_committer/src/db/serde_db_utils.rs line 63 at r4 (raw file):

Previously, ArielElp wrote…

AI decision, it's supposed to be marginally faster, and you know there are no duplications here (it's the keys from a HashMap).

but with a HashMap, order is not well defined, and you could get different digests for the same input. right?

[ArielElp]

@ArielElp made 1 comment.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on dorimedini-starkware).

---

crates/starknet_committer/src/db/serde_db_utils.rs line 63 at r4 (raw file):

Previously, dorimedini-starkware wrote…

but with a HashMap, order is not well defined, and you could get different digests for the same input. right?

Hence the sort, sort unstable is just about ordering of equal elements