Next Gen STEP AP242 Native-ish BREP Geometric Modeling Kernel
Firmament V2 casing
Firmament generally uses PascalCase for semantic constructs and camelCase for ordinary data slots. Casing is a readability convention, not a semantic ceremony. The parser may accept legacy casing where it is unambiguous, and no style-only warnings are emitted. New V2 language vocabulary does not use snake_case; imported names, user identifiers, and engineering designations such as 5052_H32 retain their source spelling.
The bounded FEA-PROD-X1 audit used three decisions: canonicalize first-party output, allow both for low-cost migration aliases, and doesn't matter for identifiers whose spelling carries no language meaning.
| Token/family | Category | Existing spellings | Canonical spelling | Decision | Parser policy |
|---|---|---|---|---|---|
Model, Template, Record, Static, Profile, Selection |
declaration | primarily PascalCase | PascalCase | canonicalize | CanonicalOnly in the central V2 grammar |
Analysis |
semantic declaration | analysis, Analysis |
Analysis |
allow both | CanonicalPreferredAcceptAliases |
LinearElastic |
analysis/type value | LinearElastic |
LinearElastic |
canonicalize | CasePermissive inside analysis declarations |
Fixed, Force, Traction, Pressure |
analysis semantic construct | lowercase and PascalCase | PascalCase | allow both | CanonicalPreferredAcceptAliases |
InlineStep declaration |
imported-body declaration | InlineStep |
InlineStep |
canonicalize | CanonicalOnly in the central V2 grammar |
inlineSTEP(path) analysis expression |
built-in compatibility function | inlineSTEP, InlineStep |
inlineSTEP |
allow both | CanonicalPreferredAcceptAliases |
Recognize, Replace, Modify, Match, Require, Assert, Pmi |
semantic/DSL construct | PascalCase in canonical grammar; some legacy lowercase routes | PascalCase | canonicalize | CanonicalOnly where ambiguity or route selection matters |
body, bodyResource, material, region, components, vector, results, lattice |
ordinary property | camelCase | camelCase | canonicalize documentation | CasePermissive in the analysis block |
Displacement, Strain, Stress, ReactionForce; X, Y, Z |
enum-like value | PascalCase/uppercase | shown spelling | allow both | CasePermissive in analysis lists |
face(...), user function/member names |
member/function-like | camelCase or authored identifier | camelCase for first-party members | doesn't matter for user members | identifier semantics apply |
| model, body, region, constraint, load names | user identifier | authored spelling, including underscores | author choice | doesn't matter | case-sensitive identity |
| STEP entity references, recognized external names, material designation tokens | external/imported identifier | source spelling such as #141, 5052_H32 |
source spelling | doesn't matter | exact external identity |
This is deliberately not a global case-insensitive rewrite. The central V2 grammar retains canonical-only spellings where token case participates in deterministic route selection. The analysis compiler accepts the harmless legacy aliases listed above so existing M5 files remain viable while first-party FEA fixtures use the preferred form.