Board index » cppbuilder » They didn't promise continued CodeGuard support

They didn't promise continued CodeGuard support


2005-02-03 05:41:31 AM
cppbuilder17
I don't find this reassuring...
community.borland.com/article/0,1410,32958,00.html
jkaster: AlisdairM: "Will CodeGuard be supported?" - We're looking at it.
 
 

Re:They didn't promise continued CodeGuard support

It's better than "no", right?
Randall Parker wrote:
Quote
I don't find this reassuring...

community.borland.com/article/0,1410,32958,00.html

jkaster: AlisdairM: "Will CodeGuard be supported?" - We're looking at
it.
--
Anders Ohlsson - Borland Developer Relations - bdn.borland.com/
Borland Software Corporation - www.borland.com/ - Excellence Endures
Enabling our customers to move into the future without abandoning their past
blogs.borland.com/ao/
homepages.borland.com/aohlsson/blog_beta/
 

Re:They didn't promise continued CodeGuard support

"Randall Parker" < XXXX@XXXXX.COM >wrote in
message news: XXXX@XXXXX.COM ...
Quote
I don't find this reassuring...

community.borland.com/article/0,1410,32958,00.html

jkaster: AlisdairM: "Will CodeGuard be supported?" - We're looking at
it.
No, they didn't. There are certainly lots of issues with integrating
multiple personalities into an IDE, and getting something like CodeGuard
integrated too probably is not a top priority, though it might end up on the
to-do list for a future version.
However, having said that, CodeGuard is a long way from being the best
memory allocation de{*word*81} around, and you would do well to use something
more robust and comprehensive, like AQTime.
Borland seems to have a habit of developing tools that, while useful at the
time, do not receive proper attention and eventually get dropped because
much better tools become available that make their tools obsolete. Turbo
Profiler is a perfect example. AQTime is far better than TProf ever was,
and it includes a memory allocation de{*word*81} that is superior to CodeGuard.
That being the case, I would not be surprised to see Borland drop CodeGuard
altogether.
- Dennis
 

{smallsort}

Re:They didn't promise continued CodeGuard support

"Dennis Jones" < XXXX@XXXXX.COM >wrote in message
Quote

That being the case, I would not be surprised to see Borland drop
CodeGuard
altogether.
In fact, I am going to go a step further and say that I hope they actually
do drop CodeGuard! I'd prefer that Borland used the manpower to concentrate
their efforts on the C++ compiler, VCL, and the RAD IDE.
- Dennis
 

Re:They didn't promise continued CodeGuard support

Dennis Jones wrote:
Quote
In fact, I am going to go a step further and say that I hope they
actually do drop CodeGuard! I'd prefer that Borland used the
manpower to concentrate their efforts on the C++ compiler, VCL, and
the RAD IDE.
I really must speak up for CodeGuard, after the compiler itself and the
IDE it is probably the best tool in the box. I would hate to lose a
perfectly good error checker like this and be expected to go out, find
and pay for a 3rd party tool to do the same job. That hardly sounds
like an incentive to upgrade ;?)
AlisdairM(TeamB)
 

Re:They didn't promise continued CodeGuard support

"Alisdair Meredith [TeamB]" < XXXX@XXXXX.COM >writes:
Quote
I really must speak up for CodeGuard, after the compiler itself and the
IDE it is probably the best tool in the box. I would hate to lose a
perfectly good error checker like this and be expected to go out, find
and pay for a 3rd party tool to do the same job. That hardly sounds
like an incentive to upgrade ;?)
Maybe more ideal would be for Borland to work out a deal to include
the 3rd party version in their box...
--
Chris (TeamB);
 

Re:They didn't promise continued CodeGuard support

"Chris Uzdavinis (TeamB)" < XXXX@XXXXX.COM >wrote in message
Quote
"Alisdair Meredith [TeamB]"
< XXXX@XXXXX.COM >writes:

>I really must speak up for CodeGuard, after the compiler itself and the
>IDE it is probably the best tool in the box. I would hate to lose a
>perfectly good error checker like this and be expected to go out, find
>and pay for a 3rd party tool to do the same job. That hardly sounds
>like an incentive to upgrade ;?)

Maybe more ideal would be for Borland to work out a deal to include
the 3rd party version in their box...

Sure, which is essentially what they said they might decide to do with
regard to a profiler (specifically, VTune was mentioned).
- Dennis
 

Re:They didn't promise continued CodeGuard support

Well, one I can't speak for AQTime, or how it compares to Codeguard
because I've never used it (In fact, I'd never even heard of it prior to
now, but in my defense, up until recently, I'd taken a two year hiatus
from Borland products as my job has required me to be .NET). My
argument in favor of Codeguard is and always has been that it was
useful. There was once a product from a company called Numega (now part
of Compuware) called BoundsChecker (once small, lightweight, and
perfectly honed to its task. Now it's a small piece of a huge
monstrosity of a project that has a cazillion features, most of which I
could care less about. But of course, I have to pay for all that stuff
I don't use; but I digress). Anyway, Boundschecker had
object-code-insertion which worked with the Microsoft compilers, and
while they had a version that worked with C++ Builder, the Borland
version they sold didn't have the same degree of functionality. During
that time, as far as I know, the ONLY product that had similar (but not
quite as extensive) functionality was Codeguard.
Codeguard, as recently as yesterday, helped me find obscure structure
resizing issues (due to some header files which weren't very
well-behaved about changing structure sizes on the fly). It may be
flawed, it may not be perfect, but it was the only thing around for a
while, and let's not forget it shipped free with the product. Last I
checked Visual Studio 2K3 had no equivalent that shipped with theirs.
Remember, the compilers really aren't changing that much for this
upcoming release, so technically it shouldn't be that hard to make CG
work with what will come with Delphi.
Dennis Jones wrote:
Quote
"Randall Parker" < XXXX@XXXXX.COM >wrote in
message news: XXXX@XXXXX.COM ...

>I don't find this reassuring...
>
>community.borland.com/article/0,1410,32958,00.html
>
>jkaster: AlisdairM: "Will CodeGuard be supported?" - We're looking at

it.

No, they didn't. There are certainly lots of issues with integrating
multiple personalities into an IDE, and getting something like CodeGuard
integrated too probably is not a top priority, though it might end up on the
to-do list for a future version.

However, having said that, CodeGuard is a long way from being the best
memory allocation de{*word*81} around, and you would do well to use something
more robust and comprehensive, like AQTime.

Borland seems to have a habit of developing tools that, while useful at the
time, do not receive proper attention and eventually get dropped because
much better tools become available that make their tools obsolete. Turbo
Profiler is a perfect example. AQTime is far better than TProf ever was,
and it includes a memory allocation de{*word*81} that is superior to CodeGuard.
That being the case, I would not be surprised to see Borland drop CodeGuard
altogether.

- Dennis


 

Re:They didn't promise continued CodeGuard support

Anders:
"Anders Ohlsson (Borland)" < XXXX@XXXXX.COM >escribi?en el mensaje
Quote
It's better than "no", right?
I'm not sure this is true. Considering the whole C++Builder fiasco and the
uncertainty has led me to change programming language and compiler provider.
Sometimes a good 'NO' is better than a bad 'YES or MAYBE'.
At least people can make informed decisions rather than wait and see...
Regards,
Martin.
 

Re:They didn't promise continued CodeGuard support

Dennis Jones wrote:
Quote
However, having said that, CodeGuard is a long way from being the best
memory allocation de{*word*81} around, and you would do well to use
something more robust and comprehensive, like AQTime.
While I agree that AQTime is very good I must object to the way you are
portraying Code Guard. It is one of the best memory abuse detection
tools I've used. It's simple and effective. For most developers it is
probably all they'll ever need.
--
Andrue Cope [TeamB]
[Bicester, Uk]
info.borland.com/newsgroups/guide.html
 

Re:They didn't promise continued CodeGuard support

Since I had compatibility problems with 3rd party libraries I had to
turn off "Validate global and stack accesses", and to turn off tracking
of memory blocks and file handles.
Now Everytime Codeguard gives alarm I check up if it is not a false
alarm. I would be happy over a good third party tools that give no false
alarm
Pete
Quote
I don't find this reassuring...

community.borland.com/article/0,1410,32958,00.html

jkaster: AlisdairM: "Will CodeGuard be supported?" - We're looking at it.
 

Re:They didn't promise continued CodeGuard support

"Andrue Cope [TeamB]" < XXXX@XXXXX.COM >wrote in message
Quote
Dennis Jones wrote:

>However, having said that, CodeGuard is a long way from being the best
>memory allocation de{*word*81} around, and you would do well to use
>something more robust and comprehensive, like AQTime.

While I agree that AQTime is very good I must object to the way you are
portraying Code Guard. It is one of the best memory abuse detection
tools I've used. It's simple and effective. For most developers it is
probably all they'll ever need.
And I'm not saying CodeGuard is a bad, necessarily. I even use it on
occassion. All I'm saying is that there are better tools available, and if
Borland were to drop CodeGuard, in favor of course, of putting that same
manpower to use on the compilers and RAD tools, I wouldn't be terribly
disappointed.
- Dennis
 

Re:They didn't promise continued CodeGuard support

"Dennis Jones" < XXXX@XXXXX.COM >wrote:
Quote
Sure, which is essentially what they said they might decide to do with
regard to a profiler (specifically, VTune was mentioned).
Does VTune work with the BCB compiler?
 

Re:They didn't promise continued CodeGuard support

"Dennis Jones" < XXXX@XXXXX.COM >wrote:
Quote
Borland were to drop CodeGuard, in favor of course, of putting that same
manpower to use on the compilers and RAD tools, I wouldn't be terribly
disappointed.
And the de{*word*81}. That needs lots of fixing and extending. Not to
mention enchantment!
 

Re:They didn't promise continued CodeGuard support

"Leroy Casterline" < XXXX@XXXXX.COM >wrote in message
Quote
"Dennis Jones" < XXXX@XXXXX.COM >wrote:

>Sure, which is essentially what they said they might decide to do with
>regard to a profiler (specifically, VTune was mentioned).

Does VTune work with the BCB compiler?
Don't know...never tried it. Maybe someone else has?
- Dennis