home *** CD-ROM | disk | FTP | other *** search
- Path: nnrp.info.ucla.edu!usenet
- From: thomasj@seas.ucla.edu (T.K. Jun)
- Newsgroups: comp.lang.c++
- Subject: Re: BC++ 5.0 IDE woes
- Date: Wed, 17 Apr 1996 08:11:06 GMT
- Organization: University of California, Los Angeles
- Message-ID: <4l293l$11p6@saba.info.ucla.edu>
- References: <4k9hjs$fl9@arl-news-svc-5.compuserve.com> <marnoldDpIvtA.E9G@netcom.com>
- NNTP-Posting-Host: micro2.seas.ucla.edu
- X-Newsreader: Forte Free Agent 1.0.82
-
- marnold@netcom.com (Matt Arnold) wrote:
-
- >Philippe Verdy <100105.3120@compuserve.com> writes:
-
- >>marnold@netcom.com (Matt Arnold) s'Θcrit :
- >>> pete@borland.com (Pete Becker) writes:
- >>>
- >>> >In article <dpayne.828477660@rainbow>, dpayne@rainbow.rmii.com says...
- >>> >>
- >>> >>One huge problem I am experiencing with BC++ 5.0 (with patch) that I
- >>> >>haven't seen posted is the terminal error while simply editing a source
- >>> >>file in the IDE. I have experienced numerous (very frustrating) abends
- >>> >>of the IDE while editing the source. At first I thought maybe there
- >>> >>was a problem with the Win95 patch (from Microsoft), but the same error
- >>> >>occurs in NT3.51 as well. Occasionally, the compile process will just hang
- >>> >>as well.
- >>>
- >>> >Nope, haven't heard of this broad a problem. Does this happen on a "clean"
- >>> >system, with no other applications running? If not, there's some interaction
- >>> >with some other application, and it would be a great help if you can track down
- >>> >which application it is.
- >>>
- >>> Borland C++ has been randomly causing GPFs for me while doing routine
- >>> editor operations since version 4.0, on a wide variety of system
- >>> configurations. I currently have version 4.52 and most of the time, I
- >>> run only the IDE (and stock Windows applets).
- >>>
- >>> These GPFs usually seem to occur in conjunction with selecting text or
- >>> using the clipboard. The exact same problem seems to have existed in
- >>> all versions I've ever used since 4.0. I can't give an exact
- >>> reproduction scenario, since it doesn't happen all the time, and it's
- >>> difficult to remember the exact keys that we had just pressed when it
- >>> happens. The one consistent theme that I do remember, is that it
- >>> always happens when working near the end of text in an editor window.
- >>>
- >>> I finally just accepted it as a bug that wasn't going away. If it *is*
- >>> some basic bug in the Borland C++ editor code, I'm not really surprised
- >>> it made it into 5.0, since apparently no one at Borland caught it going
- >>> from 4.0 all the way 4.52. However, since 5.0 is supposed to be such a
- >>> rework of the IDE, it seems unlikely the bug could have persisted.
- >>>
- >>> Still, I can't imagine, if the programmers at Borland use Borland C++
- >>> themselves, why someone at Borland in the IDE group hasn't come across
- >>> this bug themselves while programming during the last two years.
- >>>
- >>> Maybe they all use Brief or something.
- >>>
- >>You should also note that there are some problems with some
- >>keyboards, which do not report the correct codes when several
- >>keys are pressed simultaneously.
- >>There are also problems with some key macro utilities which
- >>sometime do not handle the keyboard state correctly, leaving
- >>the shift codes (Shift, Alt, AltGr, Control, NumLock, CapsLock
- >>and so on) in a bad state. Thse tools simply grab the system
- >>keyboard messages, and sometimes do not generate correct
- >>sequences to the other applications. Tools known to do that
- >>are those on Compaq machines, and those coming with some
- >>mouse emulator drivers...
-
- >That's nice, but I've never used a Compaq PC, nor anything but
- >whatever Microsoft mouse driver comes with whatever version of
- >Windows I've ever used, and this IDE crash has always happened
- >for me at some point on all machines I have ever run Borland
- >C++ 4.x on. It certainly seems like an IDE bug to me. I just
- >hope when I get 5.0, it stops happening.
-
- >I first noticed the problem using 4.0 on Windows 3.1. It
- >continues to happen, on occasion, with 4.52 on a stock Windows
- >95 installation.
-
- >Regards,
- >-------------------------------------------------------------------------
- >Matt Arnold | | ||| | |||| | | | || ||
- >marnold@netcom.com | | ||| | |||| | | | || ||
- >Boston, MA | 0 | ||| | |||| | | | || ||
- >617.389.7384 (h) 617.576.2760 (w) | | ||| | |||| | | | || ||
- >C++, MIDI, Win32/95 developer | | ||| 4 3 1 0 8 3 || ||
- >-------------------------------------------------------------------------
-
- I am afraid that the problem is even worse in BC 5.0. I can't believe
- that BC 5.0 crashes while doing a simple edit of the source code.
- When you think about it all it is a text editor but some how when you
- delete a line or character you get a complete crash of the complier.
- From the looks of how buggy version 5.0 is, Borland is not going to
- be around for long. I'm using NT 3.51 with 32MB of ram. On a very
- clean system (clean install of the operating system and programs).
-
-
- thomasj@seas.ucla.edu
-
-