- Decided to remove boost due to extensive dependencies
- Huge refactor on Wayland loading to support retrieval of Protocol headers - Setup EGL to create surfaces for Wayland windows
This commit is contained in:
@@ -53,15 +53,6 @@ be achieved using events at different stages of those engines that are on-demand
|
||||
* Walking the stack and retrieving information about the code is *very* platform
|
||||
dependent, and it makes little sense spending days writing a library that will
|
||||
be less effective.
|
||||
* `boost-atomic` `boost-thread`
|
||||
* Concurrency support is *extremely* platform dependent. Each OS has its own
|
||||
implementations with its own idiosyncrasies. We are not writing an operating system,
|
||||
we do not want to spend days writing a library that will be less performant and
|
||||
much more bugprone. Comparing this with the containers library, the containers
|
||||
library is worth reimplementing since it is not platform dependent and we can
|
||||
make an implementation with consistent behavior. Boost is slightly better for our
|
||||
purposes since its implementation is more consistent across implementations than
|
||||
the C++ stdlib.
|
||||
|
||||
|
||||
## Definitions
|
||||
|
||||
Reference in New Issue
Block a user