Implement Rust crate version of EQBSL library with docs, examples, and tests by Steake · Pull Request #17 · Steake/EQBSL

Steake · 2026-02-25T00:04:18Z

Implement a rust crate version of the EQBSL library. Include docs, examples and TDD/BDD tests.

Superconductor Ticket Implementation | App Preview

Initial release of the Evidence-Based Subjective Logic (EQBSL) crate. Provides core logic for trust modeling, including evidence-to-opinion mapping, temporal decay, and multi-hop trust propagation. - Implement Subjective Logic operators (Consensus and Discounting). - Add EQBSL pipeline for tensor evidence and hyperedge attribution. - Include deterministic node-level trust embeddings. - Provide comprehensive documentation, examples, and TDD/BDD tests. Co-authored-by: Superconductor <noreply@superconductor.com>

netlify · 2026-02-25T00:04:24Z

✅ Deploy Preview for eqbsl-demo ready!

Name	Link
🔨 Latest commit	`3c1036b`
🔍 Latest deploy log	https://app.netlify.com/projects/eqbsl-demo/deploys/69b3a6382f18ec0008d3cba9
😎 Deploy Preview	https://deploy-preview-17--eqbsl-demo.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify project configuration.

Copilot

Pull request overview

Adds a new Rust crate (rust/) implementing core EQBSL primitives (opinions, evidence→opinion mapping, state/pipeline ops) along with examples and automated tests, to provide a Rust-native EQBSL library.

Changes:

Introduces Rust implementations for Subjective Logic opinions (fusion/discounting) and EBSL evidence→opinion mapping.
Adds EQBSL state model utilities (decay, hyperedge attribution, depth-1 propagation) plus a basic node embedding.
Adds crate packaging (Cargo.toml/lock), docs, example usage, and integration/BDD-style tests.

Reviewed changes

Copilot reviewed 10 out of 12 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
rust/src/lib.rs	Defines module structure and re-exports for the crate’s public API.
rust/src/opinion.rs	Implements the `Opinion` type with expectation/fusion/discount operators.
rust/src/ebsl.rs	Implements evidence→opinion mapping (`calculate_opinion`) and unit test.
rust/src/model.rs	Adds EQBSL state model, parameter validation, decay, hyperedge attribution, and propagation.
rust/src/embedding.rs	Adds a deterministic baseline node embedding from edge opinions.
rust/examples/basic_usage.rs	Demonstrates basic usage of Params/State/opinions and opinion operators.
rust/tests/integration_tests.rs	Exercises the end-to-end pipeline pieces (opinions, propagation, decay, hyperedges).
rust/tests/bdd_tests.rs	Adds scenario-style tests for fusion and discounting behavior.
rust/Cargo.toml	Declares the new Rust crate and dependencies (serde/ndarray).
rust/Cargo.lock	Locks dependency graph for the Rust crate.
rust/README.md	Crate-level documentation and usage instructions.
.gitignore	Ignores Rust build artifacts (`rust/target/`).

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-25T00:11:09Z

rust/src/ebsl.rs

+pub fn calculate_opinion(r: f64, s: f64, k: f64, a: f64) -> Opinion {
+    let denominator = r + s + k;
+    Opinion::new(
+        r / denominator,
+        s / denominator,
+        k / denominator,
+        a
+    )


calculate_opinion divides by denominator = r + s + k without guarding against denominator == 0.0. Since this is a public API (callers can pass k = 0 and r = s = 0), this can produce NaNs that later break sorting/propagation. Consider validating inputs (e.g., assert/return Result when k <= 0 or denominator <= 0) or defining a safe fallback opinion for the zero-denominator case.

Copilot · 2026-02-25T00:11:09Z

rust/src/model.rs

+        let n = h.nodes.len();
+        if n < 2 { continue; }
+
+        let alpha = 1.0 / (n * (n - 1)) as f64;


alpha is computed as 1.0 / (n * (n - 1)) as f64. The multiplication happens in usize first and can overflow for large hyperedges (panic in debug, wrap in release). Consider computing in f64 (e.g., (n as f64) * ((n - 1) as f64)) or using checked/saturating arithmetic before converting.

Suggested change

let alpha = 1.0 / (n * (n - 1)) as f64;

let denom = (n as f64) * ((n - 1) as f64);

let alpha = 1.0 / denom;

Copilot · 2026-02-25T00:11:09Z

rust/src/model.rs

+}
+
+/// Projects an m-dimensional evidence vector into scalar (r, s) space.
+pub fn rs_from_vec(e: &Array1<f64>, w_pos: &Array1<f64>, w_neg: &Array1<f64>) -> (f64, f64) {


rs_from_vec calls e.dot(w_pos) / e.dot(w_neg), which will panic at runtime if the evidence vector length doesn’t match the weight vector length. Since State.edges / Hyperedge.e can be populated externally, consider validating evidence dimensionality (e.g., in Params::validate, a State::validate, or by returning Result from rs_from_vec/compute_opinions) to avoid panics on malformed inputs.

Suggested change

pub fn rs_from_vec(e: &Array1<f64>, w_pos: &Array1<f64>, w_neg: &Array1<f64>) -> (f64, f64) {

pub fn rs_from_vec(e: &Array1<f64>, w_pos: &Array1<f64>, w_neg: &Array1<f64>) -> (f64, f64) {

// Ensure all vectors have matching dimensionality to avoid panics from `dot`.

if e.len() != w_pos.len() || e.len() != w_neg.len() {

// Malformed input: return a safe default instead of panicking.

return (0.0, 0.0);

}

Copilot · 2026-02-25T00:11:09Z

rust/src/model.rs

+        return;
+    }
+
+    let beta_dt = params.decay_beta.mapv(|x| x.powi(dt_steps as i32));


dt_steps is a u32 but is cast to i32 for powi. For dt_steps > i32::MAX, this wraps to a negative exponent and produces incorrect decay (or unexpected growth). Use a checked conversion (e.g., i32::try_from(dt_steps)), change the parameter type, or compute decay in a way that supports u32 without lossy casting.

Suggested change

let beta_dt = params.decay_beta.mapv(|x| x.powi(dt_steps as i32));

let beta_dt = params.decay_beta.mapv(|x| x.powf(dt_steps as f64));

Copilot · 2026-02-25T00:11:10Z

rust/src/model.rs

+                lst.push((ok.expectation(), k.clone()));
+            }
+        }
+        lst.sort_by(|a, b| b.0.partial_cmp(&a.0).unwrap());


lst.sort_by(|a, b| b.0.partial_cmp(&a.0).unwrap()) can panic if any expectation is NaN (e.g., if base_rate is NaN, or if an Opinion was constructed from invalid inputs). Consider filtering out NaNs, using total_cmp, or providing an ordering fallback instead of unwrap() to keep propagation robust.

Suggested change

lst.sort_by(|a, b| b.0.partial_cmp(&a.0).unwrap());

lst.sort_by(|a, b| b.0.total_cmp(&a.0));

Steake

@copilot Please include mermaid diagrams. Please make read me considerably more useful and include examples Inc. real world examples

Steake · 2026-02-25T01:23:58Z

@copilot Please include mermaid diagrams. Please make read me considerably more useful and include examples Inc. real world examples

@copilot

Copilot · 2026-02-25T01:24:06Z

@Steake I've opened a new pull request, #18, to work on those changes. Once the pull request is ready, I'll request review from you.

Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

…ents Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

…Logic Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

docs: Expand EQBSL Rust crate README and update root README with crate details

Steake · 2026-03-13T12:25:27Z

@copilot

Resolve merge conflicts to main from this branch, while taking care to preserve the architecture of the cathexis system and the rust implementation of EQBSL, ensure the two codebases are preserved in functionality and do not duplicate implementation details. Additionally, they should be carefully integrated and the rust crates should form a unified and cohesive whole.

Copilot · 2026-03-13T12:25:34Z

@Steake I've opened a new pull request, #22, to work on those changes. Once the pull request is ready, I'll request review from you.

Copilot AI review requested due to automatic review settings February 25, 2026 00:04

Copilot started reviewing on behalf of Steake February 25, 2026 00:04 View session

Copilot AI reviewed Feb 25, 2026

View reviewed changes

Steake commented Feb 25, 2026

View reviewed changes

Steake self-assigned this Feb 25, 2026

Initial plan

2f4f852

Copilot AI mentioned this pull request Feb 25, 2026

docs: Expand EQBSL Rust crate README and update root README with crate details #18

Merged

Copilot AI and others added 8 commits February 25, 2026 01:28

docs(rust): enhance README with mermaid diagrams and real-world examples

5645793

Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

docs(rust): fix mermaid diagram styles for readability in all themes

c57e95a

Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

docs(rust): fix diagram/code value mismatches and address review comm…

883d409

…ents Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

docs(rust): correct EQBSL full name to Evidence-Qualified Subjective …

959854f

…Logic Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

docs(rust): fix review comments in README code examples and API docs

8aae3e0

Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

docs: add Rust crate section to root README and fix EQBSL terminology

b3ca5de

Co-authored-by: Steake <530040+Steake@users.noreply.github.com>

Add reputation-gated airdrop example and align EQBSL terminology

740eb87

Merge pull request #18 from Steake/copilot/sub-pr-17

3c1036b

docs: Expand EQBSL Rust crate README and update root README with crate details

Steake closed this Mar 13, 2026

Steake reopened this Mar 13, 2026

Copilot AI mentioned this pull request Mar 13, 2026

Unify EQBSL and CATHEXIS Rust crates into a shared workspace #22

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Implement Rust crate version of EQBSL library with docs, examples, and tests#17

Implement Rust crate version of EQBSL library with docs, examples, and tests#17
Steake wants to merge 10 commits intomainfrom
implement-rust-crate-version-of-eqbsl-library-with-docs-examples-and-tests-Wn9PR9JC77fh

Steake commented Feb 25, 2026

Uh oh!

netlify bot commented Feb 25, 2026 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Feb 25, 2026

Uh oh!

Copilot AI Feb 25, 2026

Uh oh!

Copilot AI Feb 25, 2026

Uh oh!

Copilot AI Feb 25, 2026

Uh oh!

Copilot AI Feb 25, 2026

Uh oh!

Steake left a comment •

edited

Loading

Uh oh!

Steake commented Feb 25, 2026

Uh oh!

Copilot AI commented Feb 25, 2026

Uh oh!

Steake commented Mar 13, 2026

Uh oh!

Copilot AI commented Mar 13, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

	let alpha = 1.0 / (n * (n - 1)) as f64;
	let denom = (n as f64) * ((n - 1) as f64);
	let alpha = 1.0 / denom;

-pub fn rs_from_vec(e: &Array1<f64>, w_pos: &Array1<f64>, w_neg: &Array1<f64>) -> (f64, f64) {
+pub fn rs_from_vec(e: &Array1<f64>, w_pos: &Array1<f64>, w_neg: &Array1<f64>) -> (f64, f64) {
+    // Ensure all vectors have matching dimensionality to avoid panics from `dot`.
+    if e.len() != w_pos.len() || e.len() != w_neg.len() {
+        // Malformed input: return a safe default instead of panicking.
+        return (0.0, 0.0);
+    }

	let beta_dt = params.decay_beta.mapv(\|x\| x.powi(dt_steps as i32));
	let beta_dt = params.decay_beta.mapv(\|x\| x.powf(dt_steps as f64));

	lst.sort_by(\|a, b\| b.0.partial_cmp(&a.0).unwrap());
	lst.sort_by(\|a, b\| b.0.total_cmp(&a.0));

Conversation

Steake commented Feb 25, 2026

Uh oh!

netlify bot commented Feb 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ Deploy Preview for eqbsl-demo ready!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 25, 2026

Choose a reason for hiding this comment

Uh oh!

Steake left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Steake commented Feb 25, 2026

Uh oh!

Copilot AI commented Feb 25, 2026

Uh oh!

Steake commented Mar 13, 2026

Uh oh!

Copilot AI commented Mar 13, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

netlify bot commented Feb 25, 2026 •

edited

Loading

Steake left a comment •

edited

Loading