Pay attention to how Delphi910 - Analysis of the wonderful blog of Danny Thorpe (Delphi compiler designer after Andrers)!

zhaozj2021-02-12  153

///

Thursday, April 08, 2004

Technology Goals for 2004

# 12:09 pm

Here's The Synopsis of Compiler Work On The Boards for Calendar Year 2004:

Win32 compiler enhancements, including codegen improvements and language enhancements to improve parity with the .NET side of the house. IL compiler enhancements to broaden the .NET push started with Delphi 8 for .NET. This will include more language enhancements, codegen improvements, and reorganized metadata output to better align with informal conventions found elsewhere in CLR. Prepare for "Whidbey" (.NET 2.0). Since the release of Whidbey has slipped out to 2005, we can not deliver a commercial product to target Whidbey this year. nevertheless, compiler work to fully support Whidbey's new IL features has to start now. The biggest item we're currently exploring is Delphi syntax for parameterized types (generics). We've had general syntax ideas cooking for several years now, so fleshing out The details in the spec and catching the caveats Has Moved Very Quickly. .NET Compact Framework. make Sure The Compact CF Compatible IL. CF IS QUITE A DIFFERENT PLAT Form Target Than Desktop .net.

While there's been vocal interest in creating a new Delphi for AMD64 project, it's just not in the cards for this year. We can not afford to drop everything to build yet another platform up from scratch, particularly when native AMD64 codegen will be coming under Extreme Pressure By .NET WITHIN THEXT YEAR OR SO. Delphi 8 for .NET Code Already Runs As Native Machine Code on AMD64 and Itanium Hardware.

More specific details will be posted here as we get things firm enough to publically commit to Subscribe to the ATOM or RSS feeds so you will not miss anything;...> Plans are forward-looking speculation subject to continuous change As these plans evolve And Take Shape, You Will Read About It here.

This is "Technology Goals for 2004" (2004 Technical Objective) from Danny Thorpe's Blog.

Delphi.net is now facing a Microsoft .NET and C # groups led by Andre to create a .NET strategy, of course, the sword of .NET is to point to the Java platform, but the C # language may swank all other program languages, C # The C has not fallen under the large pressure :), but even if there is no .NET, with the MFC library, C can also dominate the WINDOWS platform general application development. (However, it is said that SCOTT recommends B.S to modify C to allow C to adapt to the virtual platform, the B.s seems to be attached to this.).

I am most concerned about how the most important language Delphi in RAD will deal with such a situation. Because most of the program language has been integrated into the OS core (CLR is also one of the cores), so the general program language will lose the selling point! Delphi.net, how to add additional value to the feature provided by the system? ? This is related to the important survival development of Delphi.net!

Background After I finished, we came to Danny Thorpe's Blag to see how Delphi added its selling point:

Danny Thorpe lists some of the 2004 technical objectives, we have to sniff out from the development direction of D9 and D10!

1 :: win32 compiler enhancements (Enhanced Win32 Compiler)

Let Delphi support Win32 and IL compilers, which can be said to be Delphi's important selling points for a while!

2: IL Compiler Enhancements (Enhanced IL Compiler)

This said that the Delphi.Net language itself is enhanced, that is, let Delphi.net get the status of the equivalent of C #. This is indeed an indispensable part of the competition with C #.

3: Prepare for "whidbey" (.NET 2.0). (Prepare .Net 2.0)

Includes the function of adding a model for delphi.net,

4: .NET Compact Framework (.NET Framework Simplia version)

Let Delphi.net produce portable IL, support CF!

From the 2004 plan, it is quite conservative in the plan of Danny Thorpe. It is mainly to provide the features that must be provided under .NET. This is why these functions are still insufficient to beat C #, but must admit that by defeating C # in a 1 year of 1 year is undoubtedly coming. What's more, C # leaders are Andrers. And see Delphi performance in the next few years.

Next, let's take a look, what is expected to Danny Thorpe and / D9, D10 in Danny Thorpe and / D9, To Be Continued

--- by Littleroy 2004-7-3

转载请注明原文地址:https://www.9cbs.com/read-7650.html

New Post(0)