# Changelog

## Score

- CAI 46 → 47 (+1.3)
- Rubric changed (rubric-2026.10.5 → rubric-2026.11.2) — scores are not directly comparable.

## Lenses

- Code Health 47 → 47 (+0.0)
- Architecture 85 → 94 (+9.0)
- Maturity 46 → 49 (+2.8)
- Readiness 33 → 33 (+0.0)
- Security 98 → 98 (+0.0)

## Resolved (7)

- Duplicate intent with inconsistent naming. 'checkMobile' and 'checkTelphone' (typo for telephone) appear to perform similar validation on phone numbers. The parameter name 'mobile' is reused for both, suggesting they might be interchangeable or one is a legacy alias.
- Duplicate intent. Both methods likely return the current time in microseconds or a float representation of microtime. The naming convention is inconsistent ('getMicrosecond' vs 'getFloatMicrotime').
- Fragmented type-checking API. There are five distinct methods for checking numeric types, with subtle variations (int vs float, zero allowance). This creates a confusing surface for users who must guess which specific check matches their validation needs.
- Inconsistent access patterns for the same data. `getClientIp` takes a `layer` argument, while `getClientIpFor4Layer` and `getClientIpFor7Layer` are separate methods. This forces users to choose between a parameterized API and specific methods for the same underlying logic.
- Inconsistent document creation API. There is a generic `newDocument` method alongside specific `newDocumentHTML`, `newDocumentXML`, etc. The parameters also vary (`contentType` vs `charset`), creating confusion about which method to use for a given string.
- Projects may be oversized for their cohesion
- Same inconsistency as `newDocument` but for file-based loading. Redundant specific methods alongside a generic one with inconsistent parameter names (`contentType` vs `charset`).
