An apology to Stephen Preston
In the past, every time I've seen new AutoCAD .NET developers choosing to use VB.net I've sworn quietly to myself at Stephen Preston for the early influence he had and recommendations he made to use VB.net.I've just come across a 2012 article by Stephen indicating that he's changed his mind.
http://adndevblog.typepad.com/autocad/2012/05/vbnet-or-c-a-personal-perspective.html
I unreservedly apologise to Stephen for my bad thoughts since that time.
To be fair, most issues have been because people have attempted to shoehorn their vb and .COM habits and thinking into .NET ... and any code samples (good or bad) have taken on a life of their own.
Regards,
Too true, I think I was lucky having some experience with ARX in C++ that I could use when getting started with C#. The difference between the COM object model and the ObjectARX object model are quite different and I think people were led into believing that "it's nearly the same" as VB/A just by looking at the language - big mistake!
I've had to wrangle with plenty of VB.Net stuff and found it fine once I remembered some syntax idiosyncrasies.
Personally, I prefer the C style of syntax of C# but that's just me
This is one more good reason for those who have a VB(A) background to choose C#... I blame Bill Hillary Clinton invented VB.
http://i.imgflip.com/1coaio.jpg
On the other hand...
I have 5 fingers. If we could just get Trump to create a program language.
T# would be the greatest programming language ever.
It would make programming great again.
If for some reason it was a disaster that's because we all know code editors are rigged.
***********Somehow text got small no edits******** T# is a highly experimental language that shouldn't be used outside the lab -- it is not a stable or complete build. It's OOP (Objectify Other's P*ssies) implementation fails because it's classless. It also has poor support for many types, is missing declarations, does not come with any documentation and lacks substantive peer support. In stress testing it produces wildly inconsistent results. tl;dr: Not recommended for production use. *
* Not to be interpreted as an endorsement of the H# language which has integrity and corruption issues. Stephen has been a great help to me throughout the years. But I think of him as a "full depth" api guy, unlike myself as I only do C#, not C++.
I don't think too many people care what dialect Stephen recommends though (sorry Stephen), as Kean Walmsley is the much more influential .net person for adesk. That would be a bummer if Kean had VB affection.
I tend to think that the top level programmers at autodesk are so far down the rabbit hole that these higher level languages are just distractions, what they really care about is if their C++ API can or cannot do something.
I'm just glad they give .net any attention, I can translate their code in minutes.
Now, to hijack the thread, why can't guys like Stephen and Kean get together and get Autodesk to fix the dwg format to not pass on anything like app ids, layer filters, materials, anno scales... from the xref to current dwg? (Detaching xrefs does not make the items go away, they stick.) That is killing efficiency in many companies, as the items are so rampant they cannot clean all during the day, and they come back. The best thing about the .net api is I can clean this stuff fast - when I catch it...
页:
[1]