Menu
▾
▴
cpp-netlib-devel
|
From: Dean M. B. <mik...@gm...> - 2010-09-02 17:07:19
|
Hi Guys, I've got some good news and some better news. The good news is that the current 0.7-devel branch after the introduction of the asynchronous HTTP connection stuff and the HTTP Request/Response and enhanced Message concepts, is now stable enough to continue working on. I intend to push more work in progress code so if you're interested in helping out with the asynchronous connection implementation, now is a good time to jump in. The better news is that I just got the green light to work on cpp-netlib while I'm in between projects. So hopefully you'll see more progress from me in the next couple of weeks. That's all for now and hopefully everyone can start helping with testing the bug fixes and the CMake build in Windows/Linux on different compilers now. Have a good one guys and I hope this helps! -- Dean Michael Berris deanberris.com |
|
From: Glyn M. <gly...@gm...> - 2010-09-02 21:38:07
|
Hi Dean. Its great to hear you're making progress. On 2 September 2010 19:06, Dean Michael Berris <mik...@gm...>wrote: > Hi Guys, > > I've got some good news and some better news. > > The good news is that the current 0.7-devel branch after the > introduction of the asynchronous HTTP connection stuff and the HTTP > Request/Response and enhanced Message concepts, is now stable enough > to continue working on. I intend to push more work in progress code so > if you're interested in helping out with the asynchronous connection > implementation, now is a good time to jump in. > > The better news is that I just got the green light to work on > cpp-netlib while I'm in between projects. So hopefully you'll see more > progress from me in the next couple of weeks. > That's all for now and hopefully everyone can start helping with > testing the bug fixes and the CMake build in Windows/Linux on > different compilers now. > > I tried to build using GCC 4.4.3 on Ubuntu and I get tons and tons of errors. Most of them it appears to be simply missing headers or misplaced typenames. I got so far but couldn't continue because boost::network::http::uri is being defined in more than place (boost/network/protocol/http/message/directives/uri.hpp and boost/network/uri/http/uri.hpp). I committed my changes in the 0.7-devel branch of my own repository. > Have a good one guys and I hope this helps! > > G |
|
From: Dean M. B. <mik...@gm...> - 2010-09-03 04:07:03
|
On Fri, Sep 3, 2010 at 5:38 AM, Glyn Matthews <gly...@gm...> wrote: > Hi Dean. > Its great to hear you're making progress. > I'm just happy I get to do C++ now. :D > On 2 September 2010 19:06, Dean Michael Berris <mik...@gm...> > wrote: >> >> Hi Guys, >> >> I've got some good news and some better news. >> >> The good news is that the current 0.7-devel branch after the >> introduction of the asynchronous HTTP connection stuff and the HTTP >> Request/Response and enhanced Message concepts, is now stable enough >> to continue working on. I intend to push more work in progress code so >> if you're interested in helping out with the asynchronous connection >> implementation, now is a good time to jump in. >> >> The better news is that I just got the green light to work on >> cpp-netlib while I'm in between projects. So hopefully you'll see more >> progress from me in the next couple of weeks. >> >> That's all for now and hopefully everyone can start helping with >> testing the bug fixes and the CMake build in Windows/Linux on >> different compilers now. >> > > I tried to build using GCC 4.4.3 on Ubuntu and I get tons and tons of > errors. Most of them it appears to be simply missing headers or misplaced > typenames. I got so far but couldn't continue because > boost::network::http::uri is being defined in more than place > (boost/network/protocol/http/message/directives/uri.hpp and > boost/network/uri/http/uri.hpp). I committed my changes in the 0.7-devel > branch of my own repository. Ah, right. I haven't considered that -- no tests cover the uri directive at this time (except for the concept checks). MSVC 2010 doesn't complain about the URI function and the URI type having the same name though, so I can't say what the fix should be. I'll try building it in my Ubuntu desktop to see if I can make it build correctly with GCC soon. It would be great if you can send me a pull request so that I can test on Windows with MSVC 2010 as well on my repo. >> >> >> Have a good one guys and I hope this helps! >> > > G > Thanks Glyn! -- Dean Michael Berris deanberris.com |
|
From: Dean M. B. <mik...@gm...> - 2010-09-03 17:46:10
|
On Fri, Sep 3, 2010 at 12:06 PM, Dean Michael Berris <mik...@gm...> wrote: > > Ah, right. I haven't considered that -- no tests cover the uri > directive at this time (except for the concept checks). MSVC 2010 > doesn't complain about the URI function and the URI type having the > same name though, so I can't say what the fix should be. I'll try > building it in my Ubuntu desktop to see if I can make it build > correctly with GCC soon. > > It would be great if you can send me a pull request so that I can test > on Windows with MSVC 2010 as well on my repo. > I tried working with clang and I can say that it's way better in pointing out errors. And it's also lightning fast compared to GCC -- it compiles everything in 30% less time. I highly recommend developing with clang instead of GCC on Linux -- even for just the error messages. I'm going to keep whacking at this as I go along, so please expect things to be in a flux on the 0.7-devel branch. Sorry about the breakage, but some of the refactorings were a little hairy and is getting very involved. At some point a real cleanup of the implementation has to be done (by me). Help would most definitely be appreciated. Have a great weekend guys! -- Dean Michael Berris deanberris.com |
|
From: Divye K. <div...@gm...> - 2010-09-03 20:16:44
|
I too can confirm a huge no. of error messages on ubuntu lucid. % g++ --version g++ (Ubuntu 4.4.3-4ubuntu5) 4.4.3 ... most of them are indicating lack of << operator implementations. Any ideas on where to start? --- Thanks, Divye On Fri, Sep 3, 2010 at 11:15 PM, Dean Michael Berris <mik...@gm... > wrote: > On Fri, Sep 3, 2010 at 12:06 PM, Dean Michael Berris > <mik...@gm...> wrote: > > > > Ah, right. I haven't considered that -- no tests cover the uri > > directive at this time (except for the concept checks). MSVC 2010 > > doesn't complain about the URI function and the URI type having the > > same name though, so I can't say what the fix should be. I'll try > > building it in my Ubuntu desktop to see if I can make it build > > correctly with GCC soon. > > > > It would be great if you can send me a pull request so that I can test > > on Windows with MSVC 2010 as well on my repo. > > > > I tried working with clang and I can say that it's way better in > pointing out errors. And it's also lightning fast compared to GCC -- > it compiles everything in 30% less time. I highly recommend developing > with clang instead of GCC on Linux -- even for just the error > messages. > > I'm going to keep whacking at this as I go along, so please expect > things to be in a flux on the 0.7-devel branch. > > Sorry about the breakage, but some of the refactorings were a little > hairy and is getting very involved. At some point a real cleanup of > the implementation has to be done (by me). > > Help would most definitely be appreciated. > > Have a great weekend guys! > > -- > Dean Michael Berris > deanberris.com > > > ------------------------------------------------------------------------------ > This SF.net Dev2Dev email is sponsored by: > > Show off your parallel programming skills. > Enter the Intel(R) Threading Challenge 2010. > http://p.sf.net/sfu/intel-thread-sfd > _______________________________________________ > Cpp-netlib-devel mailing list > Cpp...@li... > https://lists.sourceforge.net/lists/listinfo/cpp-netlib-devel > |
|
From: Dean M. B. <mik...@gm...> - 2010-09-04 03:01:17
|
On Sat, Sep 4, 2010 at 4:16 AM, Divye Kapoor <div...@gm...> wrote: > I too can confirm a huge no. of error messages on ubuntu lucid. > % g++ --version > g++ (Ubuntu 4.4.3-4ubuntu5) 4.4.3 > ... > most of them are indicating lack of << operator implementations. Any ideas > on where to start? I think the header-ordering issue is biting us in the ass now. I'm spending quite a lot of time inserting includes and checking the order of the includes to fix these things. I stopped last night when I got the hello_world server test building, and now the http_1_0_test is failing with the overloaded wrapper names are not being properly resolved by GCC and clang, but MSVC2010 had no problem figuring out. I'd say it would be better if all the wrappers, modifiers, and directives were included appropriately. Hopefully that's not too much work to do and I intend to work on that part in the next few days. I recommend Divye that you try out clang in Ubuntu to get a better idea on what the errors really are and some sort of indication on how to fix them. ;) This has helped me a lot in decoding what the problem is and what the possible solutions are. Clang is at http://clang.llvm.org/ -- I can say that it's more standards-compliant, and whatever build with clang seems to build with GCC and MSVC, but not the other way around. If we use clang in the development of things it'd be better because we can have a higher confidence that it will work in GCC and MSVC. HTH -- Dean Michael Berris deanberris.com |
|
From: Dean M. B. <mik...@gm...> - 2010-09-05 12:49:18
|
On Sat, Sep 4, 2010 at 11:00 AM, Dean Michael Berris <mik...@gm...> wrote: > On Sat, Sep 4, 2010 at 4:16 AM, Divye Kapoor <div...@gm...> wrote: >> I too can confirm a huge no. of error messages on ubuntu lucid. >> % g++ --version >> g++ (Ubuntu 4.4.3-4ubuntu5) 4.4.3 >> ... >> most of them are indicating lack of << operator implementations. Any ideas >> on where to start? > > I think the header-ordering issue is biting us in the ass now. > And *finally* I have the cpp-netlib related tests building in clang. Somehow the MIME tests are still a little needing attention, and I am having some hiccups encountered with clang and Boost.MIME -- Marshall, are you there? BTW, I just pushed and have the header include issues fixed. So now at least we can go on stabilizing the branch, and for me to actually finish the asynchronous client implementation. Lots for me to do but hopefully I can get a lot of work in this coming week. Hang in there guys I'll get this futures-based basic_response<> thingy working soon enough. Once that's done I'm moving on to the server to support asynchronous handlers and the beginnings of a web framework that works on top of cpp-netlib. Fingers crossed I'll have that done by the end of this coming week. -- Dean Michael Berris deanberris.com |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X