Fix #532: [Model] IntegerKnapsack

[isPANN]

Summary

• Add IntegerKnapsack model to src/models/set/ — the unbounded knapsack variant where items can be selected with non-negative integer multiplicities
• Value type Max<i64>, variable domains {0,...,floor(B/s_i)} per item, complexity (capacity+1)^num_items
• CLI create support (pred create IntegerKnapsack --sizes ... --values ... --capacity ...), canonical example-db entry, paper entry with example

close #532

Test plan

• 20 unit tests covering creation, evaluation (feasible, infeasible, empty, out-of-domain), brute force solving, serialization, deserialization validation, edge cases, and paper example verification
• make fmt-check passes
• make clippy passes
• make test passes (all workspace tests)
• make paper compiles
• Example-db tests pass

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.86%. Comparing base (6a4f5c7) to head (f8d785b).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff            @@
##             main     #815    +/-   ##
========================================
  Coverage   97.85%   97.86%            
========================================
  Files         633      635     +2     
  Lines       69346    69612   +266     
========================================
+ Hits        67859    68126   +267     
+ Misses       1487     1486     -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[GiggleLiu]

Agentic Review Report

Structural Check

Structural Review: model IntegerKnapsack

Structural Completeness

#	Check	Status
1	Model file exists	PASS — src/models/set/integer_knapsack.rs:1
2	inventory::submit! present	PASS — src/models/set/integer_knapsack.rs:11
3	Serialize / Deserialize derive on struct	PASS — src/models/set/integer_knapsack.rs:50
4	Problem trait impl	PASS — src/models/set/integer_knapsack.rs:112
5	Aggregate value present	PASS — type Value = Max<i64> at src/models/set/integer_knapsack.rs:114
6	#[cfg(test)] + #[path = ...] test link	PASS — src/models/set/integer_knapsack.rs:208
7	Test file exists	PASS — src/unit_tests/models/set/integer_knapsack.rs:1
8	Test file has >= 3 test functions	PASS — 19 test_ functions in src/unit_tests/models/set/integer_knapsack.rs:6
9	Registered in set/mod.rs	PASS — src/models/set/mod.rs:17, src/models/set/mod.rs:30
10	Re-exported in models/mod.rs	PASS — src/models/mod.rs:51
11	Variant registration exists	PASS — src/models/set/integer_knapsack.rs:152
12	CLI resolve_alias entry	PASS — alias resolution is registry-driven in problemreductions-cli/src/problem_name.rs:16, backed by the schema entry in src/models/set/integer_knapsack.rs:11
13	CLI create support	PASS — example hint at problemreductions-cli/src/commands/create.rs:719 and handler at problemreductions-cli/src/commands/create.rs:2323
14	Canonical model example registered	PASS — example spec in src/models/set/integer_knapsack.rs:192, wired through src/models/set/mod.rs:40 and src/example_db/model_builders.rs:3
15	Paper display-name entry	PASS — docs/paper/reductions.typ:114
16	Paper problem-def block	PASS — docs/paper/reductions.typ:4035
17	No blacklisted autogenerated files committed	PASS — diff touches none of the blacklisted generated paths

Build Status

• make test: PASS
• make clippy: PASS

Semantic Review

• Serialization/validation expectations: ISSUE — Deserialize only validates fields individually at src/models/set/integer_knapsack.rs:50. A JSON payload with mismatched sizes / values lengths bypasses the constructor invariant in src/models/set/integer_knapsack.rs:66, and evaluate() later indexes self.values[i] by self.sizes.len() at src/models/set/integer_knapsack.rs:135, so malformed input can panic instead of being rejected.
• evaluate() correctness: OK — for well-formed instances it rejects wrong-length, out-of-domain, and overweight configs before returning the objective value at src/models/set/integer_knapsack.rs:127.
• dims() correctness: OK — one multiplicity variable per item with domain size floor(capacity / size_i) + 1 at src/models/set/integer_knapsack.rs:120.
• Size getter consistency: OK — capacity() and num_items() exist and match the complexity metadata used by src/models/set/integer_knapsack.rs:152; see src/models/set/integer_knapsack.rs:101 and src/models/set/integer_knapsack.rs:106.
• Weight handling: OK — not a graph/weight-parameterized model.
• Paper/model definition match: OK — schema fields in src/models/set/integer_knapsack.rs:19 and the paper definition in docs/paper/reductions.typ:4035 agree with the implemented model.

Issue Compliance

#	Check	Status
1	Problem name matches issue	OK — IntegerKnapsack
2	Mathematical definition matches	OK — maximize Σ c_i v_i subject to Σ c_i s_i ≤ B in src/models/set/integer_knapsack.rs:27 and docs/paper/reductions.typ:4035
3	Problem framing matches	OK — optimization model with Value = Max<i64> at src/models/set/integer_knapsack.rs:114
4	Type parameters match	OK — no extra type parameters or threshold/bound field added
5	Configuration space matches	OK — dims() encodes {0, ..., floor(B/s_i)} per item at src/models/set/integer_knapsack.rs:120
6	Feasibility check matches	OK — infeasible / out-of-domain configs return Max(None) at src/models/set/integer_knapsack.rs:127
7	Objective function matches	OK — evaluate() computes Σ c_i * values[i] at src/models/set/integer_knapsack.rs:143
8	Complexity matches	OK — declare_variants! uses (capacity + 1)^num_items at src/models/set/integer_knapsack.rs:152

Summary

• 17/17 structural checks passed
• 8/8 issue compliance checks passed
• ISSUE — malformed JSON with mismatched sizes / values lengths can deserialize successfully and later panic in evaluate() instead of failing validation cleanly; see src/models/set/integer_knapsack.rs:50, src/models/set/integer_knapsack.rs:66, and src/models/set/integer_knapsack.rs:135.

---

Quality Check

Quality Review

Design Principles

• DRY: ISSUE — validation is split between field-level serde hooks and IntegerKnapsack::new, but the two paths enforce different invariants. Deserialization checks positivity only, while constructor-only length checks mean malformed JSON can create an invalid model state that later panics during evaluation. src/models/set/integer_knapsack.rs:52, src/models/set/integer_knapsack.rs:66, src/models/set/integer_knapsack.rs:143
• KISS: ISSUE — the CLI create path reuses an assertion-based constructor for user input, so ordinary validation failures become panics instead of structured errors. A fallible validation path would be simpler and safer at the API boundary. problemreductions-cli/src/commands/create.rs:2323, problemreductions-cli/src/commands/create.rs:2341, src/models/set/integer_knapsack.rs:67
• HC/LC: OK

HCI (if CLI/MCP changed)

• Error messages: ISSUE — pred create IntegerKnapsack will panic on mismatched lengths, non-positive sizes/values, or negative capacity because the branch forwards parsed user input straight into IntegerKnapsack::new. That bypasses actionable CLI diagnostics for exactly the invalid inputs users are most likely to make. problemreductions-cli/src/commands/create.rs:2323, problemreductions-cli/src/commands/create.rs:2341, src/models/set/integer_knapsack.rs:67
• Discoverability: OK
• Consistency: OK
• Least surprise: ISSUE — the new model registers brute-force complexity as (capacity + 1)^num_items, while the paper entry in the same PR states the standard exact dynamic program is O(n B). Registry-backed CLI surfaces will therefore report a worse complexity than the project’s own documentation claims is known. src/models/set/integer_knapsack.rs:152, docs/paper/reductions.typ:4038
• Feedback: OK

Test Quality

• Naive test detection: ISSUE
  • test_integer_knapsack_deserialization_rejects_invalid_fields covers sign checks but misses mismatched sizes/values lengths, which is the serialized-state bug that currently bypasses constructor validation and can later panic in evaluate/solver code. src/unit_tests/models/set/integer_knapsack.rs:156, src/models/set/integer_knapsack.rs:52, src/models/set/integer_knapsack.rs:143

Issues

Critical (Must Fix)

• Deserializing IntegerKnapsack does not enforce sizes.len() == values.len(), so malformed JSON can construct an invalid instance that later panics when evaluate indexes self.values[i]. This is a real crash path for loaded instances and for any solver that evaluates them. src/models/set/integer_knapsack.rs:52, src/models/set/integer_knapsack.rs:66, src/models/set/integer_knapsack.rs:143

Important (Should Fix)

• The new pred create IntegerKnapsack branch can panic on invalid user input because it calls IntegerKnapsack::new directly. Bad CLI input should return Result errors, not abort the process via assert!. problemreductions-cli/src/commands/create.rs:2323, problemreductions-cli/src/commands/create.rs:2341, src/models/set/integer_knapsack.rs:67
• The registered best-known complexity is inconsistent with the same PR’s paper text. declare_variants! advertises brute-force (capacity + 1)^num_items, but the paper documents the standard exact O(n B) DP; registry-backed complexity output will therefore be misleading. src/models/set/integer_knapsack.rs:152, docs/paper/reductions.typ:4038

Minor (Nice to Have)

• None.

Summary

• Critical: deserialization can build an invalid IntegerKnapsack with mismatched vector lengths, leading to panics later. src/models/set/integer_knapsack.rs:52, src/models/set/integer_knapsack.rs:143
• Important: pred create IntegerKnapsack crashes on invalid input because it uses an assertion-based constructor as a CLI validation path. problemreductions-cli/src/commands/create.rs:2323, problemreductions-cli/src/commands/create.rs:2341
• Important: the model’s registered complexity metadata contradicts the paper entry and will mislead registry/CLI consumers. src/models/set/integer_knapsack.rs:152, docs/paper/reductions.typ:4038

---

Agentic Feature Tests

Agentic Feature Tests

Summary

I tested the pred list / pred show / pred create --example / pred solve flow in the current worktree. The CLI exposes Knapsack as the canonical problem name; IntegerKnapsack is not present in the catalog or example DB. The documented Knapsack path works end to end.

Test Results

• cargo run -q -p problemreductions-cli --bin pred -- list | rg -n "Knapsack|IntegerKnapsack"
  • Output included Knapsack * and PartiallyOrderedKnapsack *; there was no IntegerKnapsack entry.
• cargo run -q -p problemreductions-cli --bin pred -- show Knapsack
  • Succeeded and displayed the problem summary, fields, complexity, and reductions.
• cargo run -q -p problemreductions-cli --bin pred -- show IntegerKnapsack
  • Failed with Error: Unknown problem: IntegerKnapsack.
• cargo run -q -p problemreductions-cli --bin pred -- create --example Knapsack -o /tmp/knapsack.json
  • Succeeded with Wrote /tmp/knapsack.json.
• cargo run -q -p problemreductions-cli --bin pred -- solve /tmp/knapsack.json
  • Succeeded and returned a valid solution for Knapsack.
• cargo run -q -p problemreductions-cli --bin pred -- create --example IntegerKnapsack -o /tmp/integer-knapsack.json
  • Failed with Error: Unknown problem: IntegerKnapsack.

Issues Found

• High, confirmed: IntegerKnapsack is not a valid CLI/catalog/example name in this worktree. A downstream user following that name cannot discover the feature, show its details, or create the example instance. The working canonical name is Knapsack.

Reproduction / Classification

• pred list does not show IntegerKnapsack.
• pred show IntegerKnapsack fails with Unknown problem: IntegerKnapsack.
• pred create --example IntegerKnapsack fails with Unknown problem: IntegerKnapsack.
• Classification: confirmed.

Doc Suggestions

• Align user-facing docs and examples with the canonical name Knapsack, or add IntegerKnapsack as an explicit alias if that name is intended to be supported.
• If the PR or issue text uses IntegerKnapsack, add a note in the CLI docs that Knapsack is the canonical catalog name so users do not hit the unknown-problem error.

Verdict

fail

---

Generated by review-pipeline

[zazabap]

LGTM