home bbs files messages ]

Just a sample of the Echomail archive

Cooperative anarchy at its finest, still active today. Darkrealms is the Zone 1 Hub.

   GOLDED      GoldED Public Release discussion.      2,690 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 2,010 of 2,690   
   Vitaliy Aksyonov to andrew clarke   
   Re: Compilers/systems   
   07 Feb 23 14:09:10   
   
   REPLY: 3:633/267 63e2abf5   
   MSGID: 1:104/117 63e2c0a4   
   CHRS: CP866 2   
   TZUTC: -0700   
   TID: hpt/lnx 1.9 2022-07-03   
   Привет, andrew!   
      
   08 Feb 23 06:52, ты писал(а) мне:   
      
    ac>>> From memory the very recent versions of MSVC no longer produce   
    ac>>> binaries that will run under Windows XP. I've found a good   
    ac>>> "middle ground" is something like Visual Studio 2012, which   
    ac>>> still runs well under Windows 10 & 11.   
    VA>> Visual Studio 2012 is not available anymore. Oldest one is 2013.   
    VA>> Would it produce binaries for Windows XP?   
    ac> I'm not sure, but if your C++ code builds with a modern version of   
    ac> MSVC then it will probably build with VS2012, so anyone with that   
    ac> compiler should still be able to build a version that will run in XP   
    ac> unless you're using a new C++ feature from C++11 or C++20 that VS2012   
    ac> doesn't support.   
      
   That's my plan. Switch to STL where possible and stay compatible with older   
   compilers.   
      
    ac> But for GoldED it shouldn't really be necessary to refactor the code   
    ac> using C++'s increasingly estoric features. Instead, just using   
    ac> features from the STL would be a big improvement.   
      
   Totally agree. Code will be much safer and easier to support.   
      
    ac>>> OTOH it would be preferable if a free compiler could be used in   
    ac>>> Windows. Recently I learned that in 2020 Embarcadero released a   
    ac>>> fork of Dev-C++ that provides GCC 9.2 and supports C++11:   
    VA>> MS provides free VS community edition. I don't see any reasons   
    VA>> why can't it be used.   
    ac> I'd forgotten about the community edition.   
      
   But it's used for windows builds. Need to be supported too. Windows is not   
   very developer-friendly system. I like Unix-style systems much more for   
   development.   
      
    ac> I suppose you could use that, though the modern versions of VS are   
    ac> kind of scary. I mean a 10+ GB download just to compile "Hello world"   
    ac> in C is a bit excessive. The download is bigger than any regular Linux   
    ac> distro. Just for the IDE and C/C++ compiler.   
      
   That is true. I don't know what do they put there. Maybe crypto miner? ;)   
      
    ac> It's strange Microsoft never released a free version of MSVC with just   
    ac> the compiler, header files and libraries but without the IDE. Borland   
    ac> did that in 2000, 23 years ago! I think it even had STL support.   
      
   Who knows, why? Probably there is some secret optimizations. Or maybe code is   
   so ugly, that they shy to show it to wide public. :D BTW, their STL   
   implementation was very bad performance wise when they introduced VS 2015.   
   They focused on following standard even with performance penalty. Our company   
   stayed on STLPort for long time because of that.   
      
    ac>>> I was going to suggest DOS/DPMI support should be dropped, but I   
    ac>>> see GCC 10.2.0 was ported to DJGPP in 2020, so that's probably   
    ac>>> useable. Though I doubt many people would complain if DOS   
    ac>>> support was removed from future GoldED versions.   
      
   There are people who may build it on DOS. I'll take that help.   
      
    VA>> That's a good question. Main concern here is that FidoNet is   
    VA>> mostly retro hobby and people may want to run it on old computers   
    VA>> and old OSes.   
    ac> Well they can still run the older retro versions.   
      
   That is true. But what if they want to use new features or fix some nasty   
   bugs? The more master evolves - the harder would be to backport such changes.   
   From other side - that decision will be made sooner or later. Then nice   
   features like smart pointers or other may be used.   
      
    ac> The question really comes down to whether you and other devs are   
    ac> hamstrung by having to still support DOS. Of course if it turns out to   
    ac> be easy to keep supporting it, eg. by cross-compiling with OpenWatcom,   
    ac> then that's good news.   
      
   I'll try. I don't want to break it yet. First touches will be just to make   
   safe changes and learn the code. It's big!   
      
    ac>>> Another option for Windows, OS/2 & DOS may be to build with a   
    ac>>> recent version of OpenWatcom 2.0, though I don't know how well   
    ac>>> it supports STL or C++11. It may be good enough. The great thing   
    ac>>> about OpenWatcom 2.0 is can run under Linux, and it's also a   
    ac>>> cross-compiler, so you can build DOS, Windows & OS/2 apps from   
    ac>>> Linux.   
    VA>> I like the idea of using cross-compiling. In this case there is   
    VA>> no need to setup many different systems. Even if they work on   
    VA>> virtual machine.   
    ac> It does simplify things. It's also really fast.   
    ac> Though another option for cross-compiling (Linux to Windows, and also   
    ac> MacOS to Windows) is mingw-w64.   
      
   Maybe some day... I'm not ready to setup that evnironment yet.   
      
    VA>> From what I've read about OpenWatcom - they don't really conform   
    VA>> fully to even C++98 and I'm not sure about C++11 support either.   
    VA>> Need to try that.   
      
    VA>> Even refactor code to use C++98 would be a huge improvement. It's   
    VA>> full of old pure C approaches.   
      
    ac> Yep, modern C/C++ in OW2.0 is a bit hit-and-miss, though there is at   
    ac> least some STL support available, ie. strings, vectors, stacks, etc.   
    ac> It's worth experimenting with it just to learn what its limits are,   
    ac> with respect to which C++ features you'd like to use. It may be good   
    ac> enough.   
      
   Totally argee!   
      
   Best regards,   
   Vitaliy Aksyonov.   
      
   --- GoldED+/LNX 1.1.5-b20220504   
    * Origin: Aurora, Colorado (1:104/117)   
   SEEN-BY: 1/123 15/0 18/200 90/1 104/117 105/81 106/201 120/340 123/131   
   SEEN-BY: 129/305 153/7715 218/700 226/30 227/114 229/110 111 112 113   
   SEEN-BY: 229/114 206 307 317 400 424 426 428 452 470 664 700 266/512   
   SEEN-BY: 282/1038 292/854 317/3 320/219 322/757 342/200 396/45 460/58   
   SEEN-BY: 633/280 712/848   
   PATH: 104/117 229/426   
      

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca