Hi Guys,

today I have fixed a bug which has been haunting me since February, which caused vtbl's to be too short under certain circumstances.

It turned out to be an order-dependent off-by-one error in the vtbl length computation.

With this puzzler fixed I can now concentrate on getting the phobos unit-tests to compile again.

A good day to you all,

Stefan

On Mon, May 13, 2019 at 02:05:49PM +0000, Stefan Koch via Digitalmars-d wrote:
> Hi Guys,
> 
> today I have fixed a bug which has been haunting me since February, which caused vtbl's to be too short under certain circumstances.
> 
> It turned out to be an order-dependent off-by-one error in the vtbl length computation.
> 
> With this puzzler fixed I can now concentrate on getting the phobos unit-tests to compile again.
[...]

Still can't wait for newCTFE to land in master.  Any updates as to the expected timeframe?


T

-- 
By understanding a machine-oriented language, the programmer will tend to use a much more efficient method; it is much closer to reality. -- D. Knuth

On Monday, 13 May 2019 at 14:05:49 UTC, Stefan Koch wrote:
> Hi Guys,
>
> today I have fixed a bug which has been haunting me since February, which caused vtbl's to be too short under certain circumstances.

That’s great to hear Stefan, way to go! That sure looked like a puzzling issue you showed me. I hope you find the time to bring newCTFE to perfection.

Bastiaan.

On Monday, 13 May 2019 at 14:05:49 UTC, Stefan Koch wrote:
> Hi Guys,
>
> today I have fixed a bug which has been haunting me since February, which caused vtbl's to be too short under certain circumstances.
>
> It turned out to be an order-dependent off-by-one error in the vtbl length computation.
>
> With this puzzler fixed I can now concentrate on getting the phobos unit-tests to compile again.
>
> A good day to you all,
>
> Stefan

Like others I can't wait for newCTFE! It may well turn the language from "I have to avoid meta-prog for build times" to "I can do as much CTFE that I want"