feat(rust-fuzzer): implement truncate and pad length mutators + fix broken test data

[Yinklekay]

Summary

Implements length-mutation fuzzer helpers (truncate / pad) for the
rust-address-fuzzer and fixes pre-existing broken test data in both
Rust crates. Closes #291.

Problem

Off-by-one and truncated inputs are classic parser killers. Feeding
progressively shorter and padded strings probes the length-handling
logic that base32 decoders often get wrong. The rust-address-fuzzer
had no dedicated mutators for length manipulation, leaving the
parser's edge-case behaviour around over-length and under-length
inputs unexplored.

Additionally, several tests across both Rust crates used a phantom
G address (GAHJJJKMOKYE4RVPZEWZTKH5FVI4PA3VL7GK2LFNUBSGBV3PR5T4Q)
that was only 53 characters long, while the parser declares
LEN_G = 56. These tests silently failed to validate anything
useful because the parse step always rejected the input before
reaching the logic being tested.

Changes

New: length mutators (src/mutators/length.rs)

truncate(addr, rng) — Randomly removes 1 to max(1, len/2)
trailing characters. Uses saturating_sub to guard against underflow
on empty / 1-char inputs. The shortened string is always too short
for any valid Stellar address.

pad(addr, rng) — Appends 1–16 random RFC 4648 base32 characters
(A–Z, 2–7). The extended string is always too long for any valid
Stellar address.

Module wiring in src/mutators/mod.rs with #![allow(dead_code)].

Bugfixes in existing test data

File	Test	Fix
prism-core/src/address.rs	valid_g_address_parses	53-char phantom → 56-char spec-vector address
prism-core/src/address.rs	lowercase_normalised_correctly	Same replacement for both lower/upper variants
prism-core/src/address.rs	invalid_base32_character	Too-short address that failed at length gate → 56-char address with invalid char 1 that reaches the base32 check
rust-address-fuzzer/src/parse.rs	parses_valid_g_address	53-char phantom → 56-char spec-vector address
rust-address-fuzzer/src/main.rs	(compilation)	saturating_add_signed expects isize, changed from i64 range to isize

All fixes use the valid G address GAYCUYT553C5LHVE2XPW5GMEJT4BXGM7AHMJWLAPZP53KJO7EIQADRSI
from the normative spec/vectors.json (module muxed_encode).

Test coverage

12 rust-address-fuzzer tests — all pass

Test	Description
base_addresses_are_valid	Setup guard: VALID_G and VALID_M from spec vectors are parseable
truncate_produces_shorter_string	Truncated G address is strictly shorter
truncate_preserves_prefix	Prefix character preserved after truncation
truncate_g_always_err_no_panic	200 seeds on G address → always Err, never Ok, never panics
truncate_m_always_err_no_panic	200 seeds on M address → always Err, never Ok, never panics
pad_produces_longer_string	Padded G address is strictly longer
pad_preserves_prefix	Prefix character preserved after padding
pad_g_always_err_no_panic	200 seeds on G address → always Err, never Ok, never panics
pad_m_always_err_no_panic	200 seeds on M address → always Err, never Ok, never panics
parses_valid_g_address	(fixed) Valid spec-vector address parses correctly
rejects_garbage	Random string rejected
rejects_empty_string	Empty input rejected

Each parse call in the four "always Err" tests is wrapped in
std::panic::catch_unwind to catch both partial-parse Ok results
and genuine panics. 200 seeds × 4 test cases = 800 distinct
parse attempts, all producing errors with zero panics.

7 prism-core tests — all pass

All original tests now pass with the corrected addresses.

CI validation

Check	Result
cargo test (rust-address-fuzzer)	12 passed, 0 failed
cargo test (prism-core)	7 passed, 0 failed
cargo clippy --all-targets (rust-address-fuzzer)	No warnings
cargo clippy --all-targets (prism-core)	No warnings

Design decisions

• truncation range (1 to len/2): Removing more than half leaves
  strings rejected at the prefix check; keeping the range ensures
  the parser consistently reaches length-validation logic.
• pad range (1–16): Stellar addresses have fixed lengths (56/69).
  Even +1 guarantees InvalidLength. The 1–16 range adds variety.
• Seed-based determinism: StdRng::seed_from_u64 makes every
  test run reproducible for debugging.
• Spec-vector addresses: Cross-language consistency — the same
  addresses pass TypeScript, Go, Dart, and now Rust.

Summary by CodeRabbit

• Tests
  • Updated address validation test cases with refreshed lowercase and uppercase examples.
  • Expanded malformed-address coverage by testing truncated and overlong addresses.
  • Confirmed invalid address inputs are rejected safely without causing application crashes.
  • Added deterministic checks to improve confidence in address parsing across varied malformed inputs.

[drips-wave]

@Yinklekay Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 1e36b830-427a-4929-bf65-15c34cc98dcf

📥 Commits

Reviewing files that changed from the base of the PR and between d35f610 and 41a661d.

📒 Files selected for processing (2)

• examples/prism-core/src/address.rs
• examples/rust-address-fuzzer/src/main.rs

---

📝 Walkthrough

Walkthrough

Adds truncation and padding mutators for Stellar addresses, integrates them into the fuzzer module tree, validates malformed inputs, and updates lowercase/uppercase parser test vectors.

Changes

Address length mutators

Layer / File(s)	Summary
Length mutators and validation examples/rust-address-fuzzer/src/mutators/length.rs	Adds random truncation and base32 padding, with deterministic tests confirming length changes, parse errors, and panic-free behavior.
Fuzzer module wiring examples/rust-address-fuzzer/src/mutators/mod.rs, examples/rust-address-fuzzer/src/main.rs	Exports the length mutator module and registers the mutator module in the fuzzer binary.

Parser test vectors

Layer / File(s)	Summary
Lowercase normalization test vectors examples/prism-core/src/address.rs	Replaces the lowercase and uppercase address strings while retaining successful parse assertions.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related PRs

• Boxkit-Labs/stellar-address-kit#301: Updates the same lowercase normalization test vectors in address.rs.

Suggested reviewers: codeze-us

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the new length mutators and the test-data fix.
Linked Issues check	✅ Passed	The PR implements truncate/pad mutators and tests that mutated inputs fail safely without panics or partial parses.
Out of Scope Changes check	✅ Passed	The visible changes stay aligned with the fuzzer mutators and supporting test fixtures, with no clear unrelated additions.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@examples/rust-address-fuzzer/src/main.rs`:
- Line 1: Update the fuzz loop in run_random to add a length-mutation campaign
that uses valid G/M seeds, invokes both mutators::truncate and mutators::pad,
and passes each mutated result to fuzz_one. Keep the existing random_string
campaign intact and remove any unused-code suppression in mutators/mod.rs that
is no longer needed once these helpers are exercised.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 81a17a06-4f4a-4524-afc3-c74cbba308e2

📥 Commits

Reviewing files that changed from the base of the PR and between d2898d4 and d35f610.

📒 Files selected for processing (5)

• examples/prism-core/src/address.rs
• examples/rust-address-fuzzer/src/main.rs
• examples/rust-address-fuzzer/src/mutators/length.rs
• examples/rust-address-fuzzer/src/mutators/mod.rs
• examples/rust-address-fuzzer/src/parse.rs

[coderabbitai]

🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

Actually feed length mutations into the fuzz loop.

Declaring the module does not register either helper: run_random still fuzzes only random_string, while mutators/mod.rs suppresses their unused-code warning. Add a campaign that calls both truncate and pad on valid G/M seeds and passes each result to fuzz_one; otherwise this binary never exercises the new mutators.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@examples/rust-address-fuzzer/src/main.rs` at line 1, Update the fuzz loop in
run_random to add a length-mutation campaign that uses valid G/M seeds, invokes
both mutators::truncate and mutators::pad, and passes each mutated result to
fuzz_one. Keep the existing random_string campaign intact and remove any
unused-code suppression in mutators/mod.rs that is no longer needed once these
helpers are exercised.

[codeZe-us]

@Yinklekay fix conflicts