• Dom DiSc (14/14) Feb 07 2023 Hello.
• Adam D Ruppe (5/7) Feb 07 2023 It is a preview switch in there (and one that should never be
• Nick Treleaven (3/5) Feb 07 2023 The spec PR was never merged:
• Dom DiSc (2/8) Feb 08 2023 Ok, thanks.
• Dmitry Ponyatov (5/7) Oct 08 2023 Is any chance that a few years later D will adopt packed
• Imperatorn (2/9) Oct 08 2023 Same
• Bruce Carneal (7/19) Oct 08 2023 Do you have issues with the earlier alternative?
• Imperatorn (3/22) Oct 08 2023 Sorry, I meant to reply to "I see D can be intensively used in

Dom DiSc <dominikus scherkl.de> writes:

Hello.

Bit fields were introduced in 101.0, but searching for them 
yields as first results:

"D does not support bit fields. If needed, they can be emulated 
with shift and mask operations, or use the std.bitmanip.bitfields 
library type."

"Bit fields are supported with the bitfields template."

and

"ImportC - There are many implementation defined aspects of C11 
bit fields. ImportC's behavior adjusts to match the behavior of 
the associated C compiler on the target platform."

But no mention that they are now available. Also the spec doesn't 
describe them.
Have they been removed again with 102.0 ?

Adam D Ruppe <destructionator gmail.com> writes:

On Tuesday, 7 February 2023 at 14:27:16 UTC, Dom DiSc wrote:
 Bit fields were introduced in 101.0, but searching for them 
 yields as first results:

It is a preview switch in there (and one that should never be 
allowed to remain, it is an absolutely awful and near totally 
useless design).

-preview=bitfields is still in the code right now.

Nick Treleaven <nick geany.org> writes:

On Tuesday, 7 February 2023 at 14:27:16 UTC, Dom DiSc wrote:
 But no mention that they are now available. Also the spec 
 doesn't describe them.

The spec PR was never merged:
https://github.com/dlang/dlang.org/pull/3190

Dom DiSc <dominikus scherkl.de> writes:

On Tuesday, 7 February 2023 at 16:21:44 UTC, Nick Treleaven wrote:
 On Tuesday, 7 February 2023 at 14:27:16 UTC, Dom DiSc wrote:
 But no mention that they are now available. Also the spec 
 doesn't describe them.

 The spec PR was never merged:
 https://github.com/dlang/dlang.org/pull/3190

Ok, thanks.

Dmitry Ponyatov <dponyatov gmail.com> writes:

 The spec PR was never merged:
 https://github.com/dlang/dlang.org/pull/3190

Is any chance that a few years later D will adopt packed 
union/structured with bit fields, slicing and formatting 
(Erlang-like maybe) ?

I see D can be intensively used in embedded and telecom 
application, where bit ops is definetely required.

Imperatorn <johan_forsberg_86 hotmail.com> writes:

On Sunday, 8 October 2023 at 08:22:20 UTC, Dmitry Ponyatov wrote:
 The spec PR was never merged:
 https://github.com/dlang/dlang.org/pull/3190

 Is any chance that a few years later D will adopt packed 
 union/structured with bit fields, slicing and formatting 
 (Erlang-like maybe) ?

 I see D can be intensively used in embedded and telecom 
 application, where bit ops is definetely required.

Same

Bruce Carneal <bcarneal gmail.com> writes:

On Sunday, 8 October 2023 at 15:52:34 UTC, Imperatorn wrote:
 On Sunday, 8 October 2023 at 08:22:20 UTC, Dmitry Ponyatov 
 wrote:
 The spec PR was never merged:
 https://github.com/dlang/dlang.org/pull/3190

 Is any chance that a few years later D will adopt packed 
 union/structured with bit fields, slicing and formatting 
 (Erlang-like maybe) ?

 I see D can be intensively used in embedded and telecom 
 application, where bit ops is definetely required.

 Same

Do you have issues with the earlier alternative?

https://dlang.org/phobos/std_bitmanip.html#bitfields


I use and like this one quite a bit.  The new bitfield variant 
*might* be important for C interop?  I'm not in embedded 
development at the moment but if I went back there I'd prefer the 
original D bitfields for any new development.

Imperatorn <johan_forsberg_86 hotmail.com> writes:

On Sunday, 8 October 2023 at 16:14:15 UTC, Bruce Carneal wrote:
 On Sunday, 8 October 2023 at 15:52:34 UTC, Imperatorn wrote:
 On Sunday, 8 October 2023 at 08:22:20 UTC, Dmitry Ponyatov 
 wrote:
 [...]

 Is any chance that a few years later D will adopt packed 
 union/structured with bit fields, slicing and formatting 
 (Erlang-like maybe) ?

 I see D can be intensively used in embedded and telecom 
 application, where bit ops is definetely required.

 Same

 Do you have issues with the earlier alternative?

 https://dlang.org/phobos/std_bitmanip.html#bitfields


 I use and like this one quite a bit.  The new bitfield variant 
 *might* be important for C interop?  I'm not in embedded 
 development at the moment but if I went back there I'd prefer 
 the original D bitfields for any new development.

Sorry, I meant to reply to "I see D can be intensively used in 
embedded and telecom application". That's what I wrote "Same" to