Hello, On http://asio.sourceforge.net/ among the listed supported systems/compilers is Win32 using Visual C++ 7.1 and Visual C++ 8.0. Are there plans to support Visual C++ 6SP{5,6} also? What are the difficulties/obstacles for this to happen? Greetings Franz
"Dr. Franz Fehringer" <fehrin@t-online.de> wrote in message news:061801c6976d$3ff973e0$0101010a@franz...
Hello,
On http://asio.sourceforge.net/ among the listed supported systems/compilers is Win32 using Visual C++ 7.1 and Visual C++ 8.0. Are there plans to support Visual C++ 6SP{5,6} also? What are the difficulties/obstacles for this to happen?
IMO Chris should not waste even a single moment of his time on VC++ 6.5. I'd really like to see asio progress to the point where Chris can write a full proposal for TR2. That is a lot of work to do between now and the October deadline. That work is a far more useful expenditure of Chis' effort, IMO. Download VC++ 8.0 Express edition. It is free. It is highly C++ standards compliant. The IDE is also much improved compared to 6.5. Time to move on. --Beman
Beman Dawes wrote:
"Dr. Franz Fehringer" <fehrin@t-online.de> wrote in message news:061801c6976d$3ff973e0$0101010a@franz...
Hello,
On http://asio.sourceforge.net/ among the listed supported systems/compilers is Win32 using Visual C++ 7.1 and Visual C++ 8.0. Are there plans to support Visual C++ 6SP{5,6} also? What are the difficulties/obstacles for this to happen?
IMO Chris should not waste even a single moment of his time on VC++ 6.5.
I'd really like to see asio progress to the point where Chris can write a full proposal for TR2. That is a lot of work to do between now and the October deadline. That work is a far more useful expenditure of Chis' effort, IMO.
Download VC++ 8.0 Express edition. It is free. It is highly C++ standards compliant. The IDE is also much improved compared to 6.5. Time to move on.
I'm in total agreement with this. Even if someone else did the the work to port asio to vc6 Chris would still have to spend time looking at / integrating the changes. Jeff
On Sat, 24 Jun 2006 06:34:10 -0400, "Beman Dawes" <bdawes@acm.org> wrote:
I'd really like to see asio progress to the point where Chris can write a full proposal for TR2. That is a lot of work to do between now and the October deadline.
Hi Beman, I haven't had time to do a serious review of Asio yet, so this question may be totally dumb: will it be submitted *monolitically*? It seems to me there are some parts, such as the ipv4, ipv6 classes which could well live on their own and be used separately from the rest. I guess many of the thread related parts could be insulated as well, though again I haven't checked throughly. Sorry if this has already been addressed. --Gennaro.
"Gennaro Prota" <gennaro_prota@yahoo.com> wrote in message news:3q2t92lvg668p42r9maq1a9unjpep8sgdq@4ax.com...
On Sat, 24 Jun 2006 06:34:10 -0400, "Beman Dawes" <bdawes@acm.org> wrote:
I'd really like to see asio progress to the point where Chris can write a full proposal for TR2. That is a lot of work to do between now and the October deadline.
Hi Beman,
I haven't had time to do a serious review of Asio yet, so this question may be totally dumb: will it be submitted *monolitically*? It seems to me there are some parts, such as the ipv4, ipv6 classes which could well live on their own and be used separately from the rest. I guess many of the thread related parts could be insulated as well, though again I haven't checked throughly.
You really need to address this question to Chris. It is up to him how he wants to structure his proposal to the committee. But the usual practice is to put all interdependent components into the same proposal, even if the components are logically separated. Otherwise in a committee environment, you run the risk of the infrastructure items getting approved but the stuff actually useful to end-uses not getting approved, or visa versa. That's why the filesystem proposal, for example, included some changes to the diagnostics clause. --Beman
Beman Dawes <bdawes@acm.org> wrote:
"Gennaro Prota" <gennaro_prota@yahoo.com> wrote in message news:3q2t92lvg668p42r9maq1a9unjpep8sgdq@4ax.com...
I haven't had time to do a serious review of Asio yet, so this question may be totally dumb: will it be submitted *monolitically*? It seems to me there are some parts, such as the ipv4, ipv6 classes which could well live on their own and be used separately from the rest. I guess many of the thread related parts could be insulated as well, though again I haven't checked throughly.
You really need to address this question to Chris. It is up to him how he wants to structure his proposal to the committee.
It is my plan to submit it as a single proposal. In the nearish future I hope to publish a very rough draft that will give some idea of what I think should be included.
But the usual practice is to put all interdependent components into the same proposal, even if the components are logically separated. Otherwise in a committee environment, you run the risk of the infrastructure items getting approved but the stuff actually useful to end-uses not getting approved, or visa versa.
Yeah, I wouldn't want to end up like this guy: http://news.independent.co.uk/uk/this_britain/article1018524.ece ;) Cheers, Chris
On Sun, 25 Jun 2006 22:23:07 -0700 (PDT), Christopher Kohlhoff <chris@kohlhoff.com> wrote:
Beman Dawes <bdawes@acm.org> wrote:
[...] You really need to address this question to Chris. It is up to him how he wants to structure his proposal to the committee.
It is my plan to submit it as a single proposal. In the nearish future I hope to publish a very rough draft that will give some idea of what I think should be included.
It would be very nice if you took some extra care for separating headers which can be used independently from the rest however (e.g. <ip_address>)
[...]
Yeah, I wouldn't want to end up like this guy:
http://news.independent.co.uk/uk/this_britain/article1018524.ece
Very amusing! Yeah, nobody would want that :) --Gennaro.
Hi Gennaro, Gennaro Prota <gennaro_prota@yahoo.com> wrote:
It would be very nice if you took some extra care for separating headers which can be used independently from the rest however (e.g. <ip_address>)
I have coarse-grained headers in my draft so far, but I'm not entirely happy with it. However, my preferred style -- one header per class or group of related functions -- doesn't seem to be idiomatic in the standard. I think what i'll do is gloss over the header arrangement for now. It is something that can be rearranged relatively easily anyway. Once I've got a readable draft together we can have a discussion on what ought to go where. Cheers, Chris
Hi Franz, Dr. Franz Fehringer <fehrin@t-online.de> wrote:
Are there plans to support Visual C++ 6SP{5,6} also?
Not at this time.
What are the difficulties/obstacles for this to happen?
Other than lack of time, lack of access to the compiler, and a general lack of enthusiasm for VC6? ;) Ancient versions of asio did support VC6, but since dropping it as a target platform I have added code using C++ template features that I am fairly certain will not work on it. Cheers, Chris
participants (6)
-
Beman Dawes -
Christopher Kohlhoff -
Dr. Franz Fehringer -
Gennaro Prota -
Jeff Garland -
loufoque