Board index » cppbuilder » Re: BorCon 2004: A New Hope for C++Builder?

Re: BorCon 2004: A New Hope for C++Builder?


2004-09-30 03:45:29 AM
cppbuilder46
Remy Lebeau (TeamB) wrote:
Quote
"Atmapuri" < XXXX@XXXXX.COM >wrote in message
news:415150ba$ XXXX@XXXXX.COM ...



>C++Builder users would have to "port" C++ code to .NET,
>
>

Actually, they wouldn't. Herb Sutter, a Microsoft employee, had a session
on what the new Managed C++ stuff would look like. *All* of the current
"unmanaged" features of C++ are being preserved! That means pointers,
new/delete, destructors, etc. So the next C++ standard will have all of the
existing C++ features, just with Managed C++ keywords and such added to it.


Gambit




Yeah, but there's ZERO guarantee that VS.NET 2005 support for C++ will
have all that in it. They're
"hoping" to, but there's ZERO guarantee of that. Especially if the C++
Express version is any indication.
Which, by the way, now comes with the added feature of NOT co-existing
with PREVIOUS versions of
VS.NET, like 2002 and 2003 did.
 
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Randall Parker wrote:
Quote
If one is running on processor X and talks to processor Y where a
de{*word*81} back-end is running and the de{*word*81} back-end sends bytes
over you in response to an examine request then on the client side the
app has to know what the byte sex is for those bytes.

Also, if that other processor does not use the same floating point
format or integer format (e.g. a DSP that has 48 bit variables) then,
again, you have to know on the client side what data representations
the target side is using.

Keep in mind that CBX has its own GUI front-end into back-end
de{*word*81}s. A fair amount of the formatting of the data and the logic
for how the data is represented is coded for on the front-end side.
The de{*word*81} back-end expects the front-end to know a fair amount
about the target architecture.

Leroy Casterline wrote:

>This isn't an area with which I'm familiar, but if CBX interfaces with
>GDB wouldn't that take care of it? Or perhaps GDB doesn't support these
>processors...or perhaps I'm not familiar enough with the issues<g>.

And no one's thought of designing XML schema's for this mapping ?
Hmmm....perhaps there's a need there....
PXML...ProcessorXML. Anyone want to have a stab at it ( I, however,
solely request mention as being the original coiner of the term )
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

"Marcelo R. Lopez, Jr." < XXXX@XXXXX.COM >wrote in message
Quote
Yeah, but there's ZERO guarantee that VS.NET
2005 support for C++ will have all that in it.
Why do you thing Microsoft would get involved in standardizing all of that
stuff if they don't plan on actually implementing it as well?
Gambit
 

{smallsort}

Re:Re: BorCon 2004: A New Hope for C++Builder?

Mark Jacobs wrote:
Quote
Why aren't you using TDateTime?


Because TDate time doesn't let you do conversions to localtime() in a
cross-platform manner. But that's just why I use it in certain places.
You know what would be nice ? If I could create a TDateTime, and do
"ToUTC(LocalTimezone)", and "ToLocalTime(LocalTimezone)" on it.
I had to write my own routines using API to correct the seconds offsets,
so later on I could call "TryEncodeDateTime(.....etc...etc....etc.. )"
Not HAPPY.
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Duane Hebert wrote:
Quote
"Andrue Cope [TeamB]" < XXXX@XXXXX.COM >wrote in message news:4157ca92$ XXXX@XXXXX.COM ...


>The only time we've known it be wrong is occasionally with strcpy().
>Sometimes for no obvious reason it starts giving out spurious errors.
>You can tell they are wrong because the range information is blatantly
>valid. Thankfully it doesn't do it often and we usually find that a
>complete ground-up (deleting OBJs /and/ EXE/DLL) build fixes it.
>
>

Haven't seen that one much but I do often get errors like
std::numeric_limits::max() returns std::numeric_limits::min()
when codeguard is enabled. I'm not sure that it's CG itself
though as I have to enable dynamic rtl (to prevent resource
from a different rtl errors) and "use debug
libraries" (to catch leaks using VCL code) for CG.
These are not normally enabled in
my projects. I found memory sleuth less of a problem
though CG does bounds checking much better.




CodeGuard is completely useless in helping you when you've got a lot of
TComInterface action going on.
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Marcelo R. Lopez, Jr. wrote:
Quote
Yeah, but there's ZERO guarantee that VS.NET 2005 support for C++ will
have all that in it. They're
"hoping" to, but there's ZERO guarantee of that. Especially if the C++
Express version is any indication.
Which, by the way, now comes with the added feature of NOT co-existing
with PREVIOUS versions of
VS.NET, like 2002 and 2003 did.
I've got 2005 running along side 2003 without any problems. As for
standards, haven't hit a problem with it yet, although I haven't put any
big projects in to it yet. And the improved debugging of STL types is a
huge bonus, and an indication that at this stage, they are looking to
support existing standards.
Russell
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

"Marcelo R. Lopez, Jr." < XXXX@XXXXX.COM >wrote in message
Quote
Mark Jacobs wrote:

>Why aren't you using TDateTime?
>
>
Because TDate time doesn't let you do conversions to localtime() in a
cross-platform manner. But that's just why I use it in certain places.
You know what would be nice ? If I could create a TDateTime, and do
"ToUTC(LocalTimezone)", and "ToLocalTime(LocalTimezone)" on it.

I had to write my own routines using API to correct the seconds offsets,
so later on I could call "TryEncodeDateTime(.....etc...etc....etc.. )"
Not HAPPY.
If you're looking for cross platform date/time stuff maybe Boost::date_time
could help. We're using it successfully here.
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

mr organic wrote:
Quote
It's far bigger than you think. Non-commercial-software entities produce
a *huge* amount of in-house code: insurance companies, banks, brokerages,
lawyers offices, hospitals, and state and federal government agencies.
There's a huge amount of legacy C and C++ code out there, especially on
Unix platforms.

Bear that in mind: Unix is the growth market, not windows. In-house Unix
programmers tend to be badly-served by IDEs. There are a few good ones
available, but Unix programmers have tended to eschew fancy graphical
tools for the usual GNU toolchain. There's a market opportunity here for
a graphical IDE.

This is at any rate what I think Borland's salesforce is telling
Management. Time will tell if people are willing to *pay* for a C/C++
IDE -- especially since Eclipse is targeting the same market.

mr_organic


I'm GLAD you mentioned that, because unless everyone missed it, the most
recent issue of SD Times mentioned
this massive migration to "Mega IDE's" where everything is a "plugin" (
Gee, I wonder where that term was coined ).
IntelliJ today....Eclipse Plugin tomorrow. Wither JBuilder ?
Well...that's the same thing for CDT, once there is a C++ IDE
plugin worth a damn ( and CDT is getting there ) for cross-platform
development, tell me again why someone's going to
pay 2K+ for the "ALM" toolset ? The folks in all those banks, insurance
companies, hospitals, state and federal government
agencies. I mean GEEZE, for all the government waste you hear about, it
sure as heck isn't being wasted in their development
departments. You just mentioned several industries that I'll tackle one
at a time:
1. Banks......They'll pay for hardware...Software ? Uh, doesn't it come
with the box ?
2. Insurance companies........You mean we have to buy software to write
the software we need to pay your medical claims ? ( and this, I can
state from personal experience, as I wrote a medical claims adjustment
system prior to coming on board for my present employer. And yes, it
was written in BCB with a nice UI, so the nice people in the Claims
Processing Center could modify what they were going to pay the Doctor
for that
ingrown toenail they just took out for you. ). I remember having to use
DIAL UP to upload 10 (ten) 5 (five) Mb sized files of processed claims
individually to a certain carriers servers because their "gateway"
server only had capacity to temporarily store 5 Mb at a time. Yes I
automated this process ( aren't scriptable terminal programs FUN boys
and girls ?!? ), but the answer I got when I asked why there wasn't a
secure FTP site available via the internet was, "No funding".
3. Hospitals ? You're kidding right ? My S.O. is the Administrator of a
small facility. Do you think they brandish about money for development
tools for their custom software needs, just like that ? Some of the
hospital chsains have rather deep pockets, but they sure as heck don't
spend it on I.T.
4. Brokerages .....you've got to be kidding, right ? They're more
tightwadded than the rest of them combined.
5. Lawyers offices.....that's gotta be SOME law firm that they have some
guy in a closet clickety clacking away writing C++ code with EMacs.
I'm just curious to see exactly WHAT "Large Companies" these sales folks
spoke to, that convinced THEM, Borland Sales Force, that it was a "GOTTA
HAVE". I mean, there are some large companies out there, but I'd be
surprised if their developers had NO UI tools at their disposal. And if
they don't, they had to have been locked up underground and fed through
a door for over a decade. I'm sorry, but the excuse of "Our sales
people made us do it ! Mean ol' sales people. ", just doesn't cut it. (
I'm speaking of Borland Management, not the Developers, of course. )
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Chris Uzdavinis (TeamB) wrote:
Quote
"Dave Jewell" < XXXX@XXXXX.COM >writes:



>The C++ team abandoned the project because of feedback from the
>*SALES* team? Yikes. Suddenly I begin to understand the reasons
>for Borland's corporate madness over the last couple of years....
>Presumably some guy on the sales team got chatting to some schoolkid
>programmer who just happened to be in love with wxWindows....
>
>

I guess it's better than having the lawyers running the show. :)



You mean they're not ?
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Harold Howe wrote:
Quote

Let me elaborate. Sales reps do not determine product release
schedules. When CBX 1.0 was initially released, the sales reps had
never seen it. They didn't play a role in determining when it was
released.

Excuse me ? You're kidding, right ? Sales have ALWAYS driven production
in the software industry.....I already have a course laid out for what
the next version of our product HAS to have because of what Sales gives
as input. And PART of that, is timeframe. A time frame the customer has
for this or that.
What the heck do you think the Futures markets are BASED ON. Gee, if I
forsee X demand and Y supply...Hmmm....I can SELL it for Z.
I'm going to disagree with you there, if you expect any of us to believe
that, you definitely have to be dreaming.
Quote
Now that it has been released and they have been selling it, they are
pushing to get an update because the quality of CBX 1.0 is hindering
their ability to sell it. The reps want Borland to place a high
priority on getting a new version of CBX ready. However, once the beta
happens, the sales reps will not have a say (I hope) in deciding when
the field test is complete. That would just be stupid.

Specifically because it WAS pushed to market. It's a CLASSIC example of
"See? See ? We've got something. We've got something you should buy." or
"See ? See ? We told you we could make this for you. Please buy it"
Quote
So yes, the sales reps have been pushing Borland to place a higher
priority on CBX than BCB. But it is not their fault that the build
options explorer is broke in cbx 1.0.

H^2
No it's not. The developer's obviously had some bugs to address. But WHY
didn't those bugs get addressed in the product ? Because it was rushed
to market at the urging of the Sales FORCE. C'mon Harold.....let's all
do the Homer......DOH !
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Quote
Yeah, but there's ZERO guarantee that VS.NET 2005 support for C++ will
have all that in it.
It already has C++/CLI. Maybe not 100% of it, but a significant portion
of the standard is already implemented, and it's integrated with the
form designer. Obviously the product can't be 100% C++/CLI compliant
yet, as the standard is not finalized at this moment.
It does co-exist with VS2003 on my system. You just be careful not to
install 2003 after 2005. The 2005 beta may have problems, yes, as it's a
beta version, but it's very stable I think.
There's no question in my mind that Microsoft will continue supporting
C++. You should be able to include standard C++ code into a C++/CLI
module. Even assigning strings between managed/unmanaged code is not
that hard anymore as it used to be with MC++ in VS2003. There's such a
thing as STL.NET, which adds an STL-like framework around the .NET
containers.
Tom
www.tweakbits.com/tfm/
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Remy Lebeau (TeamB) wrote:
Quote
"Marcelo R. Lopez, Jr." < XXXX@XXXXX.COM >wrote in message
news:415b123a$ XXXX@XXXXX.COM ...



>Yeah, but there's ZERO guarantee that VS.NET
>2005 support for C++ will have all that in it.
>
>

Why do you thing Microsoft would get involved in standardizing all of that
stuff if they don't plan on actually implementing it as well?


Gambit




I didn't say they weren't. I said they weren't going to put it all in
VS.NET 2005. At least there's no stated guarantee
in anything that I've read from M$ that said that they will. Of course
they're going to the effort of doing it, it's in their
best interest that they do. Just the same, there's not guarantee from
them that it'll all make it into "Whidbey".
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Russell Hind wrote:
Quote
Marcelo R. Lopez, Jr. wrote:

>Yeah, but there's ZERO guarantee that VS.NET 2005 support for C++
>will have all that in it. They're
>"hoping" to, but there's ZERO guarantee of that. Especially if the
>C++ Express version is any indication.
>Which, by the way, now comes with the added feature of NOT
>co-existing with PREVIOUS versions of
>VS.NET, like 2002 and 2003 did.


I've got 2005 running along side 2003 without any problems. As for
standards, haven't hit a problem with it yet, although I haven't put
any big projects in to it yet. And the improved debugging of STL
types is a huge bonus, and an indication that at this stage, they are
looking to support existing standards.

Russell
The Express ? or the Beta 1 ?
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Tamas Demjen wrote:
Quote
>Yeah, but there's ZERO guarantee that VS.NET 2005 support for C++
>will have all that in it.


It already has C++/CLI. Maybe not 100% of it, but a significant
portion of the standard is already implemented, and it's integrated
with the form designer. Obviously the product can't be 100% C++/CLI
compliant yet, as the standard is not finalized at this moment.

Uh, I think that they already submitted it back in the summer. I'd have
to check on the date, but I'm pretty sure I saw that announcement. Then
again, it could've been an announcement that they were GOING to, in
which case it adds more to my point that we can't necessarily WAIT to
see what MS is going to do, before forging ahead.
Quote
It does co-exist with VS2003 on my system. You just be careful not to
install 2003 after 2005. The 2005 beta may have problems, yes, as it's
a beta version, but it's very stable I think.

Apparently you installed Beta 1, I'm talking about the Express packages
( and I think the same applies to Beta 2 ).
Quote
There's no question in my mind that Microsoft will continue supporting
C++. You should be able to include standard C++ code into a C++/CLI
module. Even assigning strings between managed/unmanaged code is not
that hard anymore as it used to be with MC++ in VS2003. There's such a
thing as STL.NET, which adds an STL-like framework around the .NET
containers.

Tom
www.tweakbits.com/tfm/
Yeah, that copy/copy back step really stinks.
 

Re:Re: BorCon 2004: A New Hope for C++Builder?

Duane Hebert wrote:
Quote
If you're looking for cross platform date/time stuff maybe Boost::date_time
could help. We're using it successfully here.




Wrote that stuff a little over a year ago, hasn't given me a speck of
problems since.