You can subscribe to this list here.
2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(52) |
Jun
(30) |
Jul
(17) |
Aug
(9) |
Sep
(4) |
Oct
(7) |
Nov
(11) |
Dec
(19) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2008 |
Jan
|
Feb
(1) |
Mar
(37) |
Apr
(28) |
May
(15) |
Jun
(28) |
Jul
(7) |
Aug
(125) |
Sep
(116) |
Oct
(85) |
Nov
(14) |
Dec
(6) |
2009 |
Jan
(11) |
Feb
(4) |
Mar
(5) |
Apr
|
May
(9) |
Jun
(5) |
Jul
(4) |
Aug
(40) |
Sep
(1) |
Oct
(19) |
Nov
(43) |
Dec
(45) |
2010 |
Jan
(76) |
Feb
(95) |
Mar
(3) |
Apr
(23) |
May
(39) |
Jun
(54) |
Jul
(6) |
Aug
(13) |
Sep
(12) |
Oct
(59) |
Nov
(53) |
Dec
(43) |
2011 |
Jan
(43) |
Feb
(44) |
Mar
(25) |
Apr
(23) |
May
|
Jun
|
Jul
|
Aug
|
Sep
(5) |
Oct
(1) |
Nov
(2) |
Dec
|
2013 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(6) |
Oct
|
Nov
|
Dec
|
From: Nelson, E. - 2 <eri...@ba...> - 2010-02-04 19:50:06
|
Glyn Matthews wrote > I fixed this this evening in my fork. In future, please use the issue tracker otherwise these will simply get lost. okay, thanks |
From: Jeroen H. <vex...@gm...> - 2010-02-04 19:40:01
|
Hi, On Thu, Feb 4, 2010 at 20:15, Nelson, Erik - 2 <eri...@ba...> wrote: > I noticed that basic_uri::port always returns the same value for a given > scheme... Is that by design? I see lots of web servers running on ports > other than 80. > It shouldn't, this is the current implementation: return parts_.port ? *(parts_.port) : (boost::iequals(parts_.scheme, string_type("https")) ? 443u : 80u); As it's a boost::optional<boost::uint16_t> it first checks whether a port is available in the uri_parts, this is the boolean operator of the optional, reflected by "parts_.port", and if it is it will return it, "*(parts_.port)". Otherwise it will check the scheme and return 443 if it is 'https' (case insensitve), and only if that's not the case it will return 80. > Thanks > > Erik > Jeroen |
From: Glyn M. <gly...@gm...> - 2010-02-04 19:39:26
|
Hi Erik, On 4 February 2010 19:54, Nelson, Erik - 2 <eri...@ba...>wrote: > Erik Nelson wrote > >2) the error comes from > > request_methods.ipp, on line > > static char const * const DELETE = "DELETE"; Winnt.h #defines > DELETE. If I change the offending line in request_methods.ipp to > 'DELETE_', that seems to fix it > > I think this didn't make it into v0.5, either. > I fixed this this evening in my fork. In future, please use the issue tracker otherwise these will simply get lost. http://github.com/cpp-netlib/cpp-netlib/issues Thanks, Glyn |
From: Nelson, E. - 2 <eri...@ba...> - 2010-02-04 19:15:13
|
I noticed that basic_uri::port always returns the same value for a given scheme... Is that by design? I see lots of web servers running on ports other than 80. Thanks Erik |
From: Nelson, E. - 2 <eri...@ba...> - 2010-02-04 18:55:04
|
Erik Nelson wrote >2) the error comes from > request_methods.ipp, on line > static char const * const DELETE = "DELETE"; Winnt.h #defines DELETE. If I change the offending line in request_methods.ipp to 'DELETE_', that seems to fix it I think this didn't make it into v0.5, either. Thanks Erik |
From: Nelson, E. - 2 <eri...@ba...> - 2010-02-04 16:13:28
|
After looking at it a little more, the request/response typedefs in 'basic_server' are coupled to the request/response types in class 'connection', and couldn't possibly be anything other than 'basic_request' and 'basic_response'. It seems to me that if the 'Tag' template arg is exclusively used to drive the request and response types. If the 'Tag' template arg was replaced with 'Request' and 'Response' template args in both the basic_server and connection classes, the type coupling could be factored out. I know this is all under development, and the design is a work in progress... it could be that this doesn't work. Erik |
From: Nelson, E. - 2 <eri...@ba...> - 2010-02-04 15:22:52
|
In server.hpp, there are some typedefs inside class 'basic_server' typedef basic_request<Tag> request; typedef basic_response<Tag> response; And corresponding typedefs referring to these inside class 'server' typedef typename basic_server<tags::http_server,Handler>::request request; typedef typename basic_server<tags::http_server,Handler>::response response; All this is inside the namespace 'http' I don't pretend to understand the overall design, but since the class 'basic_server' is templated on type 'Handler' as well as type 'Tag', it seems to me that the types 'request' & 'response' are unnecessarily coupled to the 'Handler' type. An example may help explain what I'm trying to say: In hello_world_server.cpp, we have something like struct hello_world; typedef http::server<hello_world> server; struct hello_world { void operator() (server::request const &request, server::response &response) {} }; This hello_world handler can *only* be used with a server of type 'http::server<hello_world>', because the request and response types are defined in terms of the server. However, the type 'http::server::request' can *only* be type 'http::basic_request<tags::http_server>'. Note that the request type is not at all dependent on the type of the handler, 'hello_world' The result is that my handler can only be used with one certain server, even though the request type is always http::basic_request<tags::http_server>. I know this could be solved with templated typedefs if they were available, but given that all this is in the namespace 'http', is there some reason that there couldn't be a typedef along the lines of namespace boost { namespace network { namespace http { typedef basic_request<tags::http_server> request; } } } So I could write struct hello_world { void operator() (http::request const &request, http::response &response) {} }; And thereby use my handler more generically? I know I could write something like this struct hello_world { void operator() (http::basic_request<tags::http_server> const &request, http::basic_response<tags::http_server> &response) {} }; but that seems a bit excessively clever, seeing as how it's stealing a typedef from both the parent and grandparent. Any comment? Erik |
From: Glyn M. <gly...@gm...> - 2010-02-04 08:22:01
|
Hi Erik, On 3 February 2010 20:44, Nelson, Erik - 2 <eri...@ba...>wrote: > Sorry to top-post, but I just wanted to note that this *wasn't* fixed in > the recent 0.5 release, and adding > > #include <ciso646> > > To parse_specific.hpp still solves the problem. > > This has been missed. I've added an issue so it won't get lost: http://github.com/cpp-netlib/cpp-netlib/issues/issue/7 I don't have access to the repository from work but I will fix this asap. Apologies and thanks for the heads up, G |
From: Nelson, E. - 2 <eri...@ba...> - 2010-02-03 19:47:02
|
Sorry to top-post, but I just wanted to note that this *wasn't* fixed in the recent 0.5 release, and adding #include <ciso646> To parse_specific.hpp still solves the problem. Thanks Erik -----Original Message----- From: Nelson, Erik - 2 [mailto:eri...@ba...] Sent: Wednesday, January 20, 2010 6:25 PM To: C++ Networking Library Developers Mailing List Subject: Re: [cpp-netlib-devel] Compiling server sample with VS2008 & boost 1.41 Dean Michael Berris wrote: > I think it's because vendors aren't quick to support them natively in > the compiler. GCC takes this in stride just fine as does other vendors > (Intel's compiler is another one I know that works pretty well). I > would quote the appropriate parts of the C++ standard if I had a copy > (which I promise I should buy someday soon) but anyway it's something > really annoying to me for non-conformant compilers > like MSVC. > > I also like them better but unfortunately to support more compilers > we're going to have to stick with !, ||, && et al. I'd say just put #include <ciso646> up high and continue to use them. They are standard C++, and they're much more descriptive than the !&&#|| stuff. Don't cave in to the noncompliant compiler vendors! Erik |
From: Glyn M. <gly...@gm...> - 2010-02-01 20:20:56
|
On 1 February 2010 19:00, Nelson, Erik - 2 <eri...@ba...>wrote: > Glyn Matthews wrote > > > Especially someone with access to MSVC (do we really not have anyone > who uses Windows?) > > I can run on windows, using msvc 9 > > Erik > > > Thanks Erik |
From: Nelson, E. - 2 <eri...@ba...> - 2010-02-01 18:00:24
|
Glyn Matthews wrote > Especially someone with access to MSVC (do we really not have anyone who uses Windows?) I can run on windows, using msvc 9 Erik |
From: Glyn M. <gly...@gm...> - 2010-02-01 17:17:09
|
Hello, I added a wiki page on testing platforms that we have available for cpp-netlib: http://wiki.github.com/cpp-netlib/cpp-netlib/developers-vs-testing-platforms If you are willing to help and be a tester, please let us know. Thanks, Glyn |
From: Kim G. <kim...@gm...> - 2010-02-01 15:50:56
|
Hi there, On Mon, Feb 1, 2010 at 16:00, Darren Garvey <dar...@gm...> wrote: > > On 1 February 2010 14:20, Glyn Matthews <gly...@gm...> wrote: >> >> >> Especially someone with access to MSVC (do we really not have anyone who >> uses Windows?) >> G > > I can set up testing on MSVC 9.0 if you are looking for testers. I run MSVC 9 as my primary platform, so I can definitely run the tests at some point. I don't have Boost properly set up on this machine at this time, so I can't provide results just yet. I seem to recall I had trouble configuring my environment when I last tried to, but managed to sort it out with some help from Dean, maybe it'd be nice to jot down some notes on prerequisites for running cpp-netlib tests on the respective platforms? Off the top of my head; python 2.6 (?), Boost (which version)?, what parts of Boost need to be built? etc. - Kim |
From: Glyn M. <gly...@gm...> - 2010-02-01 15:43:32
|
Hello Darren, On 1 February 2010 16:00, Darren Garvey <dar...@gm...> wrote: > Hey Guys, > > Glad to see the progress on the library. > > On 1 February 2010 14:20, Glyn Matthews <gly...@gm...> wrote: > >> Thanks guys, >> >> On 1 February 2010 15:02, Dean Michael Berris <mik...@gm...>wrote: >> >>> On Mon, Feb 1, 2010 at 8:14 PM, Glyn Matthews <gly...@gm...> >>> wrote: >>> > If you are willing to help and be a tester, please let us know. >>> > >>> >>> Yes please, this would be greatly appreciated. >>> >> >> Especially someone with access to MSVC (do we really not have anyone who >> uses Windows?) >> G >> > > I can set up testing on MSVC 9.0 if you are looking for testers. Are unit > tests usually run on a scheduled basis? Is there a way the test-runner > script for Boost could be utilised to automate this? > Great, thanks for getting involved. We don't yet run scheduled tests, each developer runs tests locally and reports problems on the list. Of course, this isn't going to take us very far so I'm taking this opporunity to organise ourselves a bit better. I wasn't aware of how boost automated it's tests, I'm going to take a closer look at this. How tightly coupled is it to subversion? G |
From: Darren G. <dar...@gm...> - 2010-02-01 15:00:46
|
Hey Guys, Glad to see the progress on the library. On 1 February 2010 14:20, Glyn Matthews <gly...@gm...> wrote: > Thanks guys, > > On 1 February 2010 15:02, Dean Michael Berris <mik...@gm...>wrote: > >> On Mon, Feb 1, 2010 at 8:14 PM, Glyn Matthews <gly...@gm...> >> wrote: >> > If you are willing to help and be a tester, please let us know. >> > >> >> Yes please, this would be greatly appreciated. >> > > Especially someone with access to MSVC (do we really not have anyone who > uses Windows?) > G > I can set up testing on MSVC 9.0 if you are looking for testers. Are unit tests usually run on a scheduled basis? Is there a way the test-runner script for Boost could be utilised to automate this? Regards, Darren |
From: Glyn M. <gly...@gm...> - 2010-02-01 14:57:12
|
Hi Marshall, On 1 February 2010 15:45, Marshall Clow <mcl...@gm...> wrote: > On Feb 1, 2010, at 6:02 AM, Dean Michael Berris wrote: > > > On Mon, Feb 1, 2010 at 8:14 PM, Glyn Matthews <gly...@gm...> > wrote: > >> Hello, > >> > >> > >> I added a wiki page on testing platforms that we have available for > >> cpp-netlib: > >> > >> > http://wiki.github.com/cpp-netlib/cpp-netlib/developers-vs-testing-platforms > >> > > > > Added my system too. > > > > Hopefully I can get my hands on a Mac OS X machine by asking some > > friends to help out testing. > > I can do Mac OS X testing. > That's great! Thank you very much for this. Do you have github account? Regards, Glyn |
From: Marshall C. <mcl...@gm...> - 2010-02-01 14:45:44
|
On Feb 1, 2010, at 6:02 AM, Dean Michael Berris wrote: > On Mon, Feb 1, 2010 at 8:14 PM, Glyn Matthews <gly...@gm...> wrote: >> Hello, >> >> >> I added a wiki page on testing platforms that we have available for >> cpp-netlib: >> >> http://wiki.github.com/cpp-netlib/cpp-netlib/developers-vs-testing-platforms >> > > Added my system too. > > Hopefully I can get my hands on a Mac OS X machine by asking some > friends to help out testing. I can do Mac OS X testing. -- Marshall |
From: Glyn M. <gly...@gm...> - 2010-02-01 14:20:13
|
Thanks guys, On 1 February 2010 15:02, Dean Michael Berris <mik...@gm...>wrote: > On Mon, Feb 1, 2010 at 8:14 PM, Glyn Matthews <gly...@gm...> > wrote: > > Hello, > > > > > > I added a wiki page on testing platforms that we have available for > > cpp-netlib: > > > > > http://wiki.github.com/cpp-netlib/cpp-netlib/developers-vs-testing-platforms > > > > Added my system too. > > Hopefully I can get my hands on a Mac OS X machine by asking some > friends to help out testing. > > > If you are willing to help and be a tester, please let us know. > > > > Yes please, this would be greatly appreciated. > Especially someone with access to MSVC (do we really not have anyone who uses Windows?) G |
From: Dean M. B. <mik...@gm...> - 2010-02-01 14:02:54
|
On Mon, Feb 1, 2010 at 8:14 PM, Glyn Matthews <gly...@gm...> wrote: > Hello, > > > I added a wiki page on testing platforms that we have available for > cpp-netlib: > > http://wiki.github.com/cpp-netlib/cpp-netlib/developers-vs-testing-platforms > Added my system too. Hopefully I can get my hands on a Mac OS X machine by asking some friends to help out testing. > If you are willing to help and be a tester, please let us know. > Yes please, this would be greatly appreciated. Thanks Glyn for taking this on! -- Dean Michael Berris cplusplus-soup.com | twitter.com/deanberris linkedin.com/in/mikhailberis | facebook.com/dean.berris | deanberris.com |
From: Jeroen H. <vex...@gm...> - 2010-02-01 13:56:13
|
Hi, On Mon, Feb 1, 2010 at 13:14, Glyn Matthews <gly...@gm...> wrote: > Hello, > > I added a wiki page on testing platforms that we have available for > cpp-netlib: > > http://wiki.github.com/cpp-netlib/cpp-netlib/developers-vs-testing-platforms > > If you are willing to help and be a tester, please let us know. > > Thanks, > Glyn > > I've added my system. Jeroen |
From: Glyn M. <gly...@gm...> - 2010-02-01 12:46:01
|
Hello, I added a wiki page on testing platforms that we have available for cpp-netlib: http://wiki.github.com/cpp-netlib/cpp-netlib/developers-vs-testing-platforms If you are willing to help and be a tester, please let us know. Thanks, Glyn |
From: Dean M. B. <mik...@gm...> - 2010-02-01 01:56:20
|
On Sat, Jan 30, 2010 at 4:01 PM, Glyn Matthews <gly...@gm...> wrote: > And massive thanks of course go to Dean for the development of the > http::server and for managing the move to github which has helped enormously > with co-ordinating the project. > You're welcome, it's my pleasure. :) > On 30 January 2010 00:47, Dean Michael Berris <mik...@gm...> > wrote: >> >> Hi Everyone, >> >> I've just bundled and released cpp-netlib v0.5 on Sourceforge and >> Github. You can check the wonderful work of Glyn over at >> http://cpp-netlib.github.com/ as well as download the release from >> http://github.com/cpp-netlib/cpp-netlib/downloads -- please file >> issues found over at http://github.com/cpp-netlib/cpp-netlib/issues . >> >> We should be starting conversations around what we should be releasing >> by the end of February (0.6). I'm personally looking into a more >> robust and extensible HTTP Server template implementation as well as >> some helper utility libraries for server-side REST service >> development. >> > > Support for other protocols too. If anyone wants to take up SMTP, FTP, ICMP > etc. in a new fork I encourage them to do so. Perhaps not in time for 0.6 > if the end of february is a hard deadline, but maybe in time for 0.7 or 0.8 > in march or april. > I'm not setting the end of Feb as a hard deadline, because I think it really depends on what we get done within Feb. :) Although I do look forward to putting more stuff related to the HTTP Server, and testing the consistency of the Message Concept across a broader swath of message types/specializations. Particularly this is for making sure that all the message types comply to the Message Concept and appropriate refinements of it. >> >> If you have feature requests and or suggestions and comments they >> would very much be welcome. >> >> Thanks to everyone who contributed to the effort and making this 0.5 >> release on target and successful. Have a great weekend everyone and I >> hope this helps! >> > > Thanks again, You're welcome! :) I'm personally looking forward to more contributions from more people through February too. :) -- Dean Michael Berris cplusplus-soup.com | twitter.com/deanberris linkedin.com/in/mikhailberis | facebook.com/dean.berris | deanberris.com |
From: Glyn M. <gly...@gm...> - 2010-01-30 08:01:48
|
And massive thanks of course go to Dean for the development of the http::server and for managing the move to github which has helped enormously with co-ordinating the project. On 30 January 2010 00:47, Dean Michael Berris <mik...@gm...>wrote: > Hi Everyone, > > I've just bundled and released cpp-netlib v0.5 on Sourceforge and > Github. You can check the wonderful work of Glyn over at > http://cpp-netlib.github.com/ as well as download the release from > http://github.com/cpp-netlib/cpp-netlib/downloads -- please file > issues found over at http://github.com/cpp-netlib/cpp-netlib/issues . > > We should be starting conversations around what we should be releasing > by the end of February (0.6). I'm personally looking into a more > robust and extensible HTTP Server template implementation as well as > some helper utility libraries for server-side REST service > development. > > Support for other protocols too. If anyone wants to take up SMTP, FTP, ICMP etc. in a new fork I encourage them to do so. Perhaps not in time for 0.6 if the end of february is a hard deadline, but maybe in time for 0.7 or 0.8 in march or april. > If you have feature requests and or suggestions and comments they > would very much be welcome. > > Thanks to everyone who contributed to the effort and making this 0.5 > release on target and successful. Have a great weekend everyone and I > hope this helps! > > Thanks again, Glyn |
From: Dean M. B. <mik...@gm...> - 2010-01-30 01:10:57
|
On Sat, Jan 30, 2010 at 8:16 AM, Nelson, Erik - 2 <eri...@ba...> wrote: > Dean Michael Berris wrote: >> Hi Everyone, >> >> I've just bundled and released cpp-netlib v0.5 on Sourceforge and >> Github. You can check the wonderful work of Glyn over at > > Thanks for the hard work of all who contributed! I'll be checking it > out on Monday! > You're welcome Erik -- looking forward to your most valuable feedback! :) -- Dean Michael Berris cplusplus-soup.com | twitter.com/deanberris linkedin.com/in/mikhailberis | facebook.com/dean.berris | deanberris.com |
From: Nelson, E. - 2 <eri...@ba...> - 2010-01-30 01:00:11
|
Dean Michael Berris wrote: > Hi Everyone, > > I've just bundled and released cpp-netlib v0.5 on Sourceforge and > Github. You can check the wonderful work of Glyn over at Thanks for the hard work of all who contributed! I'll be checking it out on Monday! Erik |