This is an archive of CompuServe forum messages from 1985 to 1995, as saved in transcripts by John Foust.
Search Results (296 messages)
#8786-Announcing Symantec C++
Message #8838
John, Symantec C++ 6.0 is compatible with the following DOS 386 extenders: Pharlap's 386 extender The same DosX extender that ships with Zortech Everything needed to compile, link and distribute programs with the DosX extender is included with the package. I don't know what the "official" word on which versions…
Reply to message #8786
#8786-Announcing Symantec C++
Message #8815
I was under the impression that this forum is for the MAC Symantec C++. Shouldn't this (and the other similar messages) be in the Zortech C++ areas? Isn't that where the people who would be interested in this look for this?
Reply to message #8786
#8709-#Announcing Symantec C++
Message #8786
Oh, wonderful – exactly six weeks since I plunked down $1000 for the Metaware / Pharlap system, and two weeks past my 30-day money-back guarantee. 🙂 Are there more details on the 32-bit DOS extender that's part of the package? Like whose is it, and what is it compatible with?
Reply to message #8709
#Announcing Symantec C++
Message #8709
SYMANTEC ANNOUNCES SYMANTEC C++ Three New Development Tools Allow Programmers to Visually and Rapidly Generate Superior Object-Oriented Code SYMANTEC C++ PROFESSIONAL 6.0 KEY FEATURES AND BENEFITS OVERVIEW o Breakthrough integrated debugging and development environment (IDDE), for complete DOS, Windows, and Win32s support o Contains Visual Programmer for MFC, developed by…
#8381-Quotes in replies
Message #8501
>> If you're not sure what someone's talking about, you can always read the earlier posting… it's online, I use the 'ri' command. Not when the scan is being done offline using Nav. I sometimes get off-the-wall comments to me that, out of context are meaningless. A one liner quoted…
Reply to message #8381
#8347-Size_t existence
Message #8486
–> There's no need to quote… what do you think this is, Internet? 🙂
Reply to message #8347
#8381-Quotes in replies
Message #8417
John: >>> I use the 'ri' command. This is NOT Internet. People pay for their time.
Reply to message #8381
#8381-#Quotes in replies
Message #8384
Sure, this is not UseNet, and people pay for their time. People who want to pay as little as possible use Navigator. But that means they cannot easily read the predecessors of a message, if they were read in a previous Nav session.
Reply to message #8381
#Quotes in replies
Message #8357
John: >>> There's no need to quote… what do you think this is, Internet? 🙂
#8311-Size_t existence
Message #8338
–> Pragmas may be "ANSI compliant", but they're certainly implementation dependent.
Reply to message #8311
Size_t existence
Message #7998
–> Presto, by depending on 'pragma once', Think C isn't ANSI.
#7695-Size_t existence
Message #7782
John, I think part of your message got chopped off. It ends with, "But if someone needs to include 'stdio.h'," … and stops there. –> And the answer is (was?) that it decides to run the game "Rogue", no matter what the pragma was. Pragmas, by definition, are not portable.…
Reply to message #7695
#7604-C++ Compile stages…
Message #7781
–> As for writing, I tend to stick to more consumer-oriented magazines, since they pay better than the tech magazines, where you get paid less to know more. 🙂
Reply to message #7604
#7605-Size_t existence
Message #7680
"Every other compiler I've seen, that wants to insure that an include file is only included once, uses a simple system of #ifndef / #define / #endif instead of resorting to an entirely non-portable pragma." Saying that one is simpler than the other is a matter of taste. From the…
Reply to message #7605
#7410-C++ Compile stages…
Message #7587
John, I know on the IBM Mainframe, under something like CMS or MVS or VM or something like that, #include "file" is SIGNIFICANTLY different from #include . Of course, they don't have directories, so you HAVE to specify whether the include file is local or not. Joe
Reply to message #7410
#7347-C++ Compile stages…
Message #7586
John, I'm not clear what the difference between INC2 and INC3 are. What if you have more than 3 levels of directory structure you need to go through? (Yeah, you can pass the results of an INC2 back into INC2, but… 🙂 ) –> And if any magazine editors are…
Reply to message #7347
#7466-Size_t existence
Message #7580
–> Why does Think C use this custom pragma to do the job, when a more useful alternative exists?
Reply to message #7466
#7348-MPW vs Think vs Unix?
Message #7579
John, Part of the problem here is THINK C does things slightly differently. A "Build Application" will take care of all of what a makefile would do, including verifying changed files (without actually checking the file dates, of course, unless you tell it to, either directly or indirectly by turning…
Reply to message #7348
#7466-#Size_t existence
Message #7553
I agree with John. The #pragma once directive always struck me as superfluous. The #ifndef/#define/#endif trio _is_ a de-facto standard among most C programmers. BTW, I have also been hit by the size_t problem and just added the trio myself, but having it included would have been nice. Cheers, Rob
Reply to message #7466
#7410-C++ Compile stages…
Message #7492
> Some files we use have want the system path, others expect to find files locally. Just trying to adapt to standard conventions. The less I have to change the better. I've got about 20 programmers to keep happy with these 'rules' on about 4 type computers. > We are…
Reply to message #7410
#7434-#Size_t existence
Message #7466
I understand there's no "portable", standardized way to do this. It just "happens" that Microsoft, Borland, Manx, several Unix compilers, the Metaware compiler, etc. "happened" to use this #define method. Why does Think C use this custom pragma to do the job, when a more useful alternative exists?
Reply to message #7434
#7348-MPW vs Think vs Unix?
Message #7465
"I've heard that future E.T.O discs will include Symantec Think C 6.0 instead of the MPW system." Not instead of, in addition to. Bob
Reply to message #7348
#7349-Size_t existence
Message #7434
"Is there some way I can portably check if size_t is around, short of re-editing your standard #include files to act like the other compilers?" By definition, there no "portable" way to check if size_t has been defined. Instead, you should just #include the header file that defines size_t, which…
Reply to message #7349
#7347-C++ Compile stages…
Message #7399
Thanks for the message on #include macro portability. The two I needed (Unix, Mac) I wrote basicly as you did. Of course our standards police made me use our naming convention. We also had to deal with quotes v.s. greater/less symbol inludes. #include "Wally.h" #include Thanks again! Cheers, —– Andrew
Reply to message #7347
#MPW vs Think vs Unix?
Message #7348
I've heard that future E.T.O discs will include Symantec Think C 6.0 instead of the MPW system. I've embarked on a Mac development project, but it stalled when I couldn't find an automatic way to build a series of applications. On Unix and DOS environments, I can write a script…
Internal Error (ZREF)
Message #7345
I received the Error "Internal error (ZREF)" while trying to compile this simple program using THINK C 6.0 on my Macintosh Centris 610. And according to the User Manual it states to Report this error to you. I am running under System 7.1, with several INIT's loaded, but have never…
#7272-C++ Compile stages…
Message #7344
Andrew, "I gather that the connection between the preprocceser and the compiler in Symantec C++ is completely closed and private?" I don't know the internals of the Symantec compiler, but there is no separate pre-processor as in an AT&T style C++ pre-processor which spits out C code. Obviously, there's some…
Reply to message #7272
#7286-Is this a bug?
Message #7343
Lee, C++ compilers seem to use *lots* of memory. You probably need to allocate more memory to the TPM application. (I'm up to 5MB. 4MB won't hack it, even for a fairly small project.) Regards, Robin
Reply to message #7286
#7314-Complete C++?
Message #7341
Michael, >>constructors may not have arguments Not true. I've got lots of constructors with arguments. SC++ appears to follow the proposed ANSI C++ spec, with about the only exception being "exceptions". If you have constructs from another C++ that don't compile in SC++, then either there's a bug in SC++…
Reply to message #7314
#7330-Oracle & THINK 6
Message #7340
I'm emailing you a project file from a project that works with Oracle. No sources, but you can look at the options I used to see what worked for me. – Burt
Reply to message #7330
#7324-Using C from C++
Message #7339
>> C++ lets you assign any pointer type to variable of type void *, but NOT the other way around; C++ considers such things "unsafe." Typecasting, however, works Yes, I knew that typecasting worked, and was what I did to make that code work. I was just surprised that going…
Reply to message #7324
#7287-AM TCL Lists
Message #7338
I finally called a friend today and went over with my program in hand and asked him to help me past this hurdle. As I expected, it turned out to be pretty easy. The key was that when I created the instance of my class (CCard, which is the class…
Reply to message #7287
#7325-C++ Conversion Ops
Message #7337
Your assignments make good sense, I'll do that from now on. Thanks. Bob
Reply to message #7325
#7305-porting file i/o
Message #7335
I'm back. I solved my own problem, by determining that I had "infer prototypes" set, and there were no actual prototypes for my file i/o functions. As soon as I added unix.h to my code, all was well. Prehaps, this can help someone else. Have fun, Rick
Reply to message #7305
#7296-NonTCL Ported to TCL
Message #7334
Great. Some possibilities that pop into mind where confusion might occur: 1. Does popping up a dialog box (more importantly a progress dialog) via direct toolbox calls confuse TCL? Do some object in TCL expect to know about every window in order to operate properly? 2. My error recovery mechanism,…
Reply to message #7296
#7329-Upgrading Zortech C++
Message #7333
John- Please see my first message (#4449) for upgrade instructions. There are many new features in version 3.1. Please contact the customer service office in Australia and they will be happy to provide you with a complete list. Thanks, Mike Rhone Sysop Customer Service
Reply to message #7329
Upgrade for ZTC++
Message #7332
Richard- We received your message, but since you didn't send it as private, you may want to closely monitor your visa charges. I will be happy to process the order, but I will also need a street address for shipping and a daytime phone number. Please send that information and…
#7231-OGL for Borland C++
Message #7331
Kurt- Unfortunately, we don't process international orders here. I will forward your request to our international office in Leiden, Holland. I will provide you with their phone and fax numbers. Please contact them if you don't receive the update within the next three weeks. Symantec Ltd 31 71 353 111…
Reply to message #7231
#7273-#Oracle & THINK 6
Message #7330
Burt… Yes, I have the docs and have installed Oracle. In fact, I have both HyperCard and MacApp-written apps using Oracle, but they were written with MPW. What I was hoping to find was some compiler option in Think that needed to be set a certain way, etc. I probably…
Reply to message #7273
#Upgrading Zortech C++
Message #7329
What differences are there between 3.0 & 3.1 Zortech C++ ? How do I upgrade from 3.0 to 3.1 here in Australia ?
#7210-asm ( ) vs asm { }
Message #7328
Suggestion noted. Kevin Irlen Symantec Languages Support
Reply to message #7210
#7267-Pane enabling
Message #7327
Your suggestions regarding color and increased modularity in the TCL have been noted. Kevin Irlen Symantec Languages Support
Reply to message #7267
#7157-new Symantac C++ book
Message #7326
phil, There was no Symantec C++ 5.0 compiler. Thus, any new books targeted for Symantec C++ would, no doubt, be limited to 6.0. There are some books dealing with THINK C 5.0, which was a C compiler with object-oriented extensions similar to C++. Rich Parker has one, for example. Some…
Reply to message #7157
#7100-#C++ Conversion Ops
Message #7325
–> By the way, is it OK to compile the TCL with Symantec C++?
Reply to message #7100
#7191-#Using C from C++
Message #7324
The reason changing the extension from .c to .cp made it work is the functions are now C++ functions, complete with mangled names. That's where you were running into trouble before. To inhibit the name mangling, put 'extern "C"' in front of the prototype, like so: extern "C" void *myFunc…
Reply to message #7191
#7190-Think C 5.0 -> 6.0?
Message #7320
Matthew, It's quite possible that you got replies (I'm fairly sure I replied to you, for example), but you didn't come back into the forum before the replies scrolled off. Don't forget, too, that you might want to fetch this entire thread, not just look at messages waiting for you.…
Reply to message #7190
#7169-Code Segment Mgmt …
Message #7319
Sounds neat, Joe. Not quite as good as having segments unloaded automatically (when specified), but significantly better than what we have now. Thanks!!! (How's that for commentary payment? 🙂 ) Another Joe
Reply to message #7169
#7294-4 byte int problem
Message #7318
I'm with you, Michael. Where I used to work, we even went as far as to define types that indicated the size, and used THEM, not the underlying C-provided types. (We were dealing with code that had to be as portable as possible to VAX/VMS, MeSs-Windows, and the IBM Mainframe.)…
Reply to message #7294
#7180-C++ Conversion Ops
Message #7317
Aha! I didn't think of that; when I initially built AppMaker's starter projects, I went ahead and changed the extension mapping immediately. (I was making some other starter projects customized for THINK C & C++, so I was in the habit of doing that already. 🙂 ) Joe
Reply to message #7180
#7176-Monthly??????
Message #7316
–> OK, fine, they've done all that. Now why wait (possibly) thirty days? Instead, send the new bug doc to the bug librarian. The librarian retrieves the cumulative bug report from source-code control, adds the new bug, replaces the bug-report in the library and uploads the doc to the forum.…
Reply to message #7176
Page 1 of 6