VisualD regressions are severe; what do we do about critical infrastructure?

[Richard (Rikki) Andrew Cattermole]

On 15/10/2024 10:42 PM, Manu wrote:
> On Tue, 15 Oct 2024 at 19:16, Richard (Rikki) Andrew Cattermole via 
> Digitalmars-d <digitalmars-d at puremagic.com <mailto:digitalmars- 
> d at puremagic.com>> wrote:
> 
>     On 15/10/2024 10:07 PM, Manu wrote:
>      > ...to make this worse; Rainer has effectively checked-out too. We've
>      > lost another one of our finest.
> 
>     He implemented the negative annotation exportation switches for dmd at
>     end of last year.
> 
>     (dllimport and visibility)
> 
>     Druntime/phobos are not currently offered as a shared library though.
> 
>     That was a MAJOR contribution to dmd.
> 
> 
> Oh I know it... I mean, I have always said Rainer is one of the 
> greatest, most competent, most high impact, and most quiet and humble 
> contributors we've ever had... but I'm not sure what your point is. He 
> seems to have checked out... that's fine, he's welcome to do so, but it 
> reinforces a serious problem.
> Key-person risk is a HUGE practical problem for D :/

More like too busy. Which is not a good thing at all.

>      > I now recognise a really major conundrum; I've recently returned
>     to D to
>      > start a company with a greenfields project. VisualD failing is
>      > essentially terminal. I'm not sure what to do. I don't have time
>      > available to try and pick up the project and work it myself, but the
>      > current state is really pushing at the border of forcing me to
>      > completely rewrite all my code in C++ on account of ecosystem
>     reliability.
> 
>     Over the past year, VisualD has broken a couple of times.
> 
>     The fix appears to update & recompile it.
> 
> 
> To be clear, it's broken in almost every single semantic-related task 
> that it performs.
> The way you describe here is because MS have changed their VS release 
> model so make major revision increments every couple of months rather 
> than every couple of years... and so yes, it needs to be regularly 
> recompiled, but that's not the cause of any functionality regression.

But yes, that is basically what happened around a year ago.

Debugging didn't work. I had to disable the plugin it was so bad.