You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Projects which are Win32-heavy are peppered with either bespoke RAII types or invocations of scopeguard-type constructs. Are there plans to improve this situation in windows-rs itself or is there room for a wil-type library in the ecosystem? If the former, should we start contributing these types here? There's also a scoping question - e.g. an RAII Handle type vs something higher-level like a SetThreadToken/restore RAII type, or even beyond that higher-level bindings like the wil win32 helpers.
Drawbacks
No response
Rationale and alternatives
A key advantage of having these in windows-rs itself is for ecosystem standardization. Additionally, the various helpers and types could key off of the same feature partitions as the underlying APIs.
Additional context
No response
The text was updated successfully, but these errors were encountered:
I encourage developers to build WIL-style helpers as discrete crates atop the windows or windows-sys crates and to treat the windows-rs family of crates as merely the generated bindings representing the underlying APIs. The line is a little fuzzy as the windows crate necessarily provides some programming aids to make APIs usable but for the most part I try to avoid adding that as much as possible to keep the line between generated bindings and hand-crafted libraries clear.
Motivation
Projects which are Win32-heavy are peppered with either bespoke RAII types or invocations of scopeguard-type constructs. Are there plans to improve this situation in windows-rs itself or is there room for a wil-type library in the ecosystem? If the former, should we start contributing these types here? There's also a scoping question - e.g. an RAII Handle type vs something higher-level like a SetThreadToken/restore RAII type, or even beyond that higher-level bindings like the wil win32 helpers.
Drawbacks
No response
Rationale and alternatives
A key advantage of having these in windows-rs itself is for ecosystem standardization. Additionally, the various helpers and types could key off of the same feature partitions as the underlying APIs.
Additional context
No response
The text was updated successfully, but these errors were encountered: