DSharp project status

[Former user]

Hi Stefan,

I just wondered what were your intentions concerning the DSharp library.

I saw that you're starting to work on it and use it for some samples of the new Spring4d persistence framework in 1.2

Couple of months ago I forked your library because I ported the C# Prism library to delphi. I wanted to use your Databinding library for it. I'd like to make my Prism4d library open source, but for that I'd need to know if you're planning on developing DSharp or not in the future.

Thanks a lot

[Stefan Glienke]

Hi,

I just looked through the most obvious and easy to fix bugs. Also there is still a rather big task of merging back many of our internal changes (we are basically using a private fork of DSharp at work) that we made over the past months/years. I don't want to dump them right into develop or master without reviewing in a single big commit which is why I said rather big task.

In contrast to Spring4D where we aim at only releasing rather polished and tested features (there are quite a lot non finished, experimental, to be tested things internally) DSharp contains many experimental things that might or not work - that can be useful or fall apart - or that are already superseded by a new more mature version (Interception/Mocks is a good example - both work fine but the Spring4D implementation is way simpler and less convoluted)

As I mentioned in another ticket most things contained in any of the packages can be considered stable or at least not experimental (which are also the things we are using in production).

There is also still quite a bit of cleanup to be done to remove redundant things from DSharp that are in Spring4D (mostly related to DSharp.Core.Reflection and DSharp.Interception).

As for the bindings core and the underlying expression engine I still want to polish that feature and move it to Spring4D. But that also requires some rework as DSharp.Bindings.pas already has some dependencies to other DSharp features. I think a first step would be to move and refactor the expression engine but as I also have other things with higher priority on my list that might not happen anytime soon (I might look into it to measure how much work it would be though).

Currently DSharp (the aforementioned stable/non experimental features) should work in all versions from XE to 10.2 (we used XE up until a few weeks ago and now are on 10.1). I currently have no plans to develop new features for DSharp (those that I develop are for Spring4D). I once had the idea to look a bit more into the entire MVVM thing but kinda lost the interest because it requires way too much UI stuff which is not my most favourite area and because many concepts I would like to implement are impossible or at least feel very alien in Delphi.