C++: Updated constructor generation and allocator handling for unions#383
Conversation
448aaf9 to
aa2849f
Compare
|
How can we enable pmr-flavored language checks for this PR? |
In |
|
|
||
| {%- if composite_type.inner_type is not UnionType %} | ||
| {% if composite_type.fields_except_padding %} | ||
| // Initializing constructor |
There was a problem hiding this comment.
At the moment tagged in_place_index constructors are only defined when the language options have constructor_convention != default, FYI
aa2849f to
318a314
Compare
|
Interesting failure that popped up all of a sudden during the lint test. Could it be from me updating |
|
My mistake, I had meant .github/workflows/test.yml |
318a314 to
4428bd5
Compare
|
Bugfix for CETL VariableLengthArray required to resolve failing checks: OpenCyphal/CETL#188 |
|
Waiting for the release of CETL v.1.4.2 to push submodule update |
|
- Allocators are stored and propagated to union composite type alternative values during construction and emplacement - `in_place_index` constructor is generated for union composite types - `DependencyBuilder` sets `uses_union` flag appropriately
4428bd5 to
2f193f3
Compare
|
|
Strong work. Thank you for your contribution citizen. 🇺🇸🏈🥧 |



in_place_indexconstructor is generated for union composite typesDependencyBuildersetsuses_unionflag appropriately