...and a way less annoying completion behaviour, as there's a completion request time out now that just cancels any too long attempt to get completion info.

http://mono-d.alexanderbothe.com/again-internal-refactorings-less-completion-bugs-v1-7-2/



Cheers mates,
Alex

Am 17.03.2014 23:08, schrieb Alexander Bothe:
> ...and a way less annoying completion behaviour, as there's a completion
> request time out now that just cancels any too long attempt to get
> completion info.
>
> http://mono-d.alexanderbothe.com/again-internal-refactorings-less-completion-bugs-v1-7-2/
>
> Cheers mates,
> Alex

I've given Mono-D the first serious try recently (I'm using a simple Sublime Text setup usually) and it was a really pleasant experience. Apart from the code completion and the DUB integration, which is of course awesome to have, some features like the integrated compare/blame/merge views are also nice productivity boosters.

Good to see progress on the auto completion, as that was the only thing that still was a little annoying - annoying just because it usually works so well that you quickly come to rely on it and thus also immediately notice when it fails. There is probably some kind of "uncanny valley" there, where it either has to be perfect, or very sketchy (based on text buffers) to be comfortable to use.

Thanks for pushing this project forward!

On Tuesday, 18 March 2014 at 10:04:27 UTC, Sönke Ludwig wrote:
> I've given Mono-D the first serious try recently (I'm using a simple Sublime Text setup usually) and it was a really pleasant experience. Apart from the code completion and the DUB integration, which is of course awesome to have, some features like the integrated compare/blame/merge views are also nice productivity boosters.
>
> Good to see progress on the auto completion, as that was the only thing that still was a little annoying - annoying just because it usually works so well that you quickly come to rely on it and thus also immediately notice when it fails. There is probably some kind of "uncanny valley" there, where it either has to be perfect, or very sketchy (based on text buffers) to be comfortable to use.

This is basically a very situation-dependent thing, as it sometimes is just not able to deduce the currrently initialized type, although I've already built-in recognition pass-throughs if template parameters couldn't be satisfied.
I've also made hard-coded Tuple completion available - and although this might not be the best idea, it's actually a big leap forward even without having to spend too much time on deducing each template (or handling CTFE - this is still a thing I don't want to touch atm).

On Monday, 17 March 2014 at 22:08:02 UTC, Alexander Bothe wrote:
> ...and a way less annoying completion behaviour, as there's a completion request time out now that just cancels any too long attempt to get completion info.
>
> http://mono-d.alexanderbothe.com/again-internal-refactorings-less-completion-bugs-v1-7-2/
>
>
>
> Cheers mates,
> Alex

Could you please port your add-in to SharpDevelop?

On Tuesday, 18 March 2014 at 13:16:45 UTC, Suliman wrote:
> Could you please port your add-in to SharpDevelop?

No, because it's Windows-only, because there's XS and VisualD for Windows already, because I'm not paid for doing this ;-)