add new parameters for AWG 2.0

[chgchi]

No description provided.

[MelvinTo]

PR Review Summary

✅ What looks good

• Change is tightly scoped to AWG obfuscation option propagation, which keeps risk contained.
• Expanding supported keys (s3/s4, i1..i5) aligns with broader AmneziaWG parameter sets and improves compatibility with newer client configs.
• No behavioral changes outside _addObfuscationOptions, so regression surface is small.

⚠️ Issues found

• Medium – falsy-value drop bug remains and affects new fields too.
  The current gate if (networkConfig[key]) skips valid values like 0 (or "0" depending normalization). For protocol/tuning knobs, zero can be meaningful. This can silently omit intended config.
• Low – no validation/range checks for newly accepted keys.
  Adding passthrough keys without validation may allow malformed values to reach downstream command/config generation.
• Issue requirement coverage: No linked issue found.

💡 Suggestions

• Change presence check to explicit existence instead of truthiness, e.g.:
• if (networkConfig[key] !== undefined && networkConfig[key] !== null).
• Add lightweight validation for numeric obfuscation fields (type/range), at least for newly introduced i*/s* keys.
• Add/extend unit tests:
• includes 0 value cases,
• confirms all new keys are emitted when present,
• confirms invalid values are rejected or normalized.

Verdict

COMMENT

---

Repo: firewalla/firerouter
PR: #1807
Head SHA: 8a8735af9b69c06104e08ff3b4e96c1b57c8115e
Checked at: 2026-03-27 20:01:06 CST

[j-sallyjin]

PR Review Summary

✅ What looks good

• Extends obfuscation option coverage in a focused way by adding:
• s3, s4
• i1–i5
• Change is minimal and localized to key whitelist logic, which keeps regression surface small.
• Backward compatibility looks preserved for existing keys (jc/jmin/jmax/s1/s2/h1..h4).

⚠️ Issues found

• Medium – potential correctness gap for numeric zero values remains.
  The condition if (networkConfig[key]) excludes valid falsy values (e.g., 0). If any obfuscation field allows zero, it will be silently dropped from generated entries.
• Low – style inconsistency in key literals.
  "i5" uses double quotes while neighboring entries use single quotes; not functional, but reduces consistency/readability.
• Issue requirement coverage: No linked issue found.

💡 Suggestions

• Prefer explicit presence checks instead of truthiness:
• e.g., if (networkConfig[key] !== undefined && networkConfig[key] !== null) (or Object.hasOwn(networkConfig, key) depending on desired semantics).
• Keep key literal style consistent ('i5') to match project conventions.
• Optional: add a small unit test ensuring obfuscation entries include zero-valued fields when provided.

Verdict

COMMENT

---

Repo: firewalla/firerouter
PR: #1807
Head SHA: 8a8735af9b69c06104e08ff3b4e96c1b57c8115e
Checked at: 2026-03-28 00:11:14 CST

[chgchi]

AI feedback reviewed, no issues.