s,cppnetlib,network, #372

ruslo · 2014-01-24T10:17:27Z

No description provided.

ruslo · 2014-01-24T10:18:56Z

Probably now variables need to be renamed from CPP-NETLIB_ to NETWORK_? (:

deanberris · 2014-01-24T11:18:28Z

I would hold off on the CMAKE variables -- I think there's value in the project name being there at least. ;)

deanberris · 2014-01-24T11:19:50Z

CMakeLists.txt

@@ -8,11 +8,11 @@
 cmake_minimum_required(VERSION 2.8)
 project(CPP-NETLIB)

-option( CPP-NETLIB_BUILD_SHARED_LIBS "Build cpp-netlib as shared libraries." OFF )
-option( CPP-NETLIB_BUILD_SINGLE_LIB "Build cpp-netlib into a single library" OFF )
+option( CPP-NETLIB_BUILD_SHARED_LIBS "Build network as shared libraries." OFF )


Haha -- I think you did too thorough a sed script here. ;)

You'd want to keep cpp-netlib in these comments (and other places where it's descriptive rather than "normative"). 😁

I did it on purpose. It depends on how to name main library.

Though it's better to quote this name. This looks weird: Build the examples using network. (:

I still would like to keep cpp-netlib here instead. These are descriptions, which really should be geared towards humans. :)

That said, I think having the actual name of the output binary is a good idea. In which case something like:

"Build cpp-netlib shared libraries (network-.so or network-.dll)."

glynos · 2014-01-24T12:21:15Z

In fact it isn't really necessary to change the prefix for what will be deprecated and removed anyway, as new stuff is developed. Eventually all the code that will replace these libraries ought to come with the network prefix. This is so we can have the old and new stuff being able to run side-by-side as we move to 1.0.

ruslo · 2014-01-24T15:33:55Z

I'm a little bit confused...

To summarize, expected:

project(CPP-NETLIB)
option(CPP-NETLIB_* "... cpp-netlib ...")
add_library(network-* ${CPP-NETLIB_*_SRCS})
add_library(cppnetlib ... ${CPP-NETLIB_*_SRCS})

If single library will build output: cppnetlib
If not single: network-uri, network-client, network-server
?

deanberris · 2014-01-25T13:05:55Z

I think you got the gist of what I'd like to happen @ruslo :)

Single library output: cppnetlib
Individual libraries: network-*

I also agree with @glynos but I'd rather be migrating the names now and cutting them out aggressively in the coming days/weeks. This means, all the deprecated stuff (message framework, etc) will fall by the wayside and won't be developed anymore -- and actually removed from the codebase.

ruslo · 2014-01-26T07:05:15Z

concurrency/test/CMakeLists.txt

@@ -8,14 +8,14 @@ include_directories(${CPP-NETLIB_SOURCE_DIR}/concurrency/src
  ${CPP-NETLIB_SOURCE_DIR}
 )
 if (CPP-NETLIB_BUILD_TESTS)
-  add_executable(cpp-netlib-thread_pool_test thread_pool_test.cpp)
-  target_link_libraries(cpp-netlib-thread_pool_test
+  add_executable(network-thread_pool_test thread_pool_test.cpp)


What about test executable? Revert it too?

Yes please, thanks @ruslo :)

ruslo · 2014-01-26T19:09:10Z

Libraries list:

network-constants
network-http-message
network-http-message-wrappers
network-http-server
network-http-v2-client
network-logging
network-message
network-message-directives
network-message-wrappers
network-uri
network_concurrency
network_error

ruslo · 2014-01-26T19:13:14Z

I can't find travis build. Are you sure cpp-netlib is activated?
See this link: http://docs.travis-ci.com/user/getting-started/

glynos · 2014-01-26T19:30:24Z

At least it's working here:

https://travis-ci.org/glynos/cpp-netlib

I'll see what I can do set it up for the cpp-netlib repo.

deanberris · 2014-01-28T03:34:42Z

@ruslo -- can you please rebase/merge from HEAD, so we can merge this soon? Thanks.

ruslo · 2014-01-28T11:09:40Z

@ruslo -- can you please rebase/merge from HEAD, so we can merge this soon? Thanks.

Ready to merge:
https://github.com/ruslo/cpp-netlib/commit/19cc4d09f992609c058b2e2f93694d3d92a15823

deanberris · 2014-01-30T01:54:59Z

Thanks @ruslo -- sorry for the delay.

LGTM

s,cppnetlib,network,

ruslo · 2014-01-30T07:41:16Z

Thanks, Dean!

Libraries list's current state:

network-constants
network-http-message
network-http-message-wrappers
network-http-server
network-http-v2-client
network-logging
network-message
network-message-directives
network-message-wrappers
network-uri
network_concurrency
network_error

What is the next step?

ruslo · 2014-01-30T07:51:22Z

At least it's working here:
https://travis-ci.org/glynos/cpp-netlib
I'll see what I can do set it up for the cpp-netlib repo.

@deanberris @glynos Yep, my repo working too: https://travis-ci.org/ruslo/cpp-netlib
And uri repo of cpp-netlib organization: https://travis-ci.org/cpp-netlib/uri
but not cpp-netlib repo of cpp-netlib organization: https://travis-ci.org/cpp-netlib/cpp-netlib

deanberris · 2014-01-30T08:19:01Z

Are network_concurrency and network_error typo's? It'd be great if we can make this consistent as network-concurrency and network-error.

Does anybody know how to make Travis work with cpp-netlib?

glynos · 2014-01-30T08:22:08Z

On 30 January 2014 09:19, Dean Michael Berris notifications@github.comwrote:

Are network_concurrency and network_error typo's? It'd be great if we can
make this consistent as network-concurrency and network-error.

Does anybody know how to make Travis work with cpp-netlib?

Yeah, it's not obvious, but I just triggered a build for this repo.

Reply to this email directly or view it on
GitHubhttps://github.com//pull/372#issuecomment-33667838

.

deanberris · 2014-01-30T08:24:13Z

@glynos Huh. I think we may be clobbering settings on each other. 😀 I'll let you figure this out going forward. :)

ruslo · 2014-01-30T08:32:27Z

Are network_concurrency and network_error typo's?

No

It'd be great if we can make this consistent as network-concurrency and network-error

@deanberris sure, I just think that if you want to remove/merge this libraries then there is no difference
what is the library name (:

I'll open new PR if you don't mind. This PR becomes a little bit messy.

glynos · 2014-01-30T08:33:18Z

On 30 January 2014 08:41, Ruslan Baratov notifications@github.com wrote:

Thanks, Dean!

Libraries list's current state:

network-constants
network-http-message
network-http-message-wrappers
network-http-server
network-http-v2-client
network-logging
network-message
network-message-directives
network-message-wrappers
network-uri
network_concurrency
network_error

What is the next step?

The next step is to finish the HTTP client and HTTP server, before removing
the old libraries.

Glyn

deanberris · 2014-01-30T08:34:43Z

@ruslo Yes, please send in a new PR. network-concurrency should be the correct name, and AFAIK it's not going away. If anything I'd like to implement the standards executor proposal.

network-error on the other hand is something worth keeping unless @glynos thinks otherwise.

glynos · 2014-01-30T08:36:35Z

I don't think it's worth keeping network-error over the long run. I intended it to be a specialization of std::system_error for this library but I no longer feel this is necessary, and it's not being used by the HTTP v2 client.

ruslo · 2014-01-30T09:02:57Z

@glynos @deanberris Thank! It's working: https://travis-ci.org/cpp-netlib/cpp-netlib/builds/17892236

ruslo · 2014-01-30T09:07:40Z

The next step is to finish the HTTP client and HTTP server, before removing the old libraries.

What part is unfinished C++ or CMake?

glynos · 2014-01-30T09:10:19Z

None, for the time-being.

deanberris reviewed Jan 24, 2014
View reviewed changes

ruslo reviewed Jan 26, 2014
View reviewed changes

ruslo mentioned this pull request Jan 26, 2014

Remove unnecessary 'add_dependencies' commands #370

Merged

s,cppnetlib,network,

19cc4d0

deanberris added a commit that referenced this pull request Jan 30, 2014

Merge pull request #372 from ruslo/libnames-refactor

d85820e

s,cppnetlib,network,

deanberris merged commit d85820e into cpp-netlib:master Jan 30, 2014

ruslo deleted the libnames-refactor branch January 30, 2014 08:36

s,cppnetlib,network, #372

s,cppnetlib,network, #372

Uh oh!

Conversation

ruslo commented Jan 24, 2014

Uh oh!

ruslo commented Jan 24, 2014

Uh oh!

deanberris commented Jan 24, 2014

Uh oh!

deanberris Jan 24, 2014

Choose a reason for hiding this comment

Uh oh!

ruslo Jan 24, 2014

Choose a reason for hiding this comment

Uh oh!

ruslo Jan 24, 2014

Choose a reason for hiding this comment

Uh oh!

deanberris Jan 25, 2014

Choose a reason for hiding this comment

Uh oh!

glynos commented Jan 24, 2014

Uh oh!

ruslo commented Jan 24, 2014

Uh oh!

deanberris commented Jan 25, 2014

Uh oh!

ruslo Jan 26, 2014

Choose a reason for hiding this comment

Uh oh!

deanberris Jan 26, 2014

Choose a reason for hiding this comment

Uh oh!

ruslo commented Jan 26, 2014

Uh oh!

ruslo commented Jan 26, 2014

Uh oh!

glynos commented Jan 26, 2014

Uh oh!

deanberris commented Jan 28, 2014

Uh oh!

ruslo commented Jan 28, 2014

Uh oh!

deanberris commented Jan 30, 2014

Uh oh!

ruslo commented Jan 30, 2014

Uh oh!

ruslo commented Jan 30, 2014

Uh oh!

deanberris commented Jan 30, 2014

Uh oh!

glynos commented Jan 30, 2014

Uh oh!

deanberris commented Jan 30, 2014

Uh oh!

ruslo commented Jan 30, 2014

Uh oh!

glynos commented Jan 30, 2014

Uh oh!

deanberris commented Jan 30, 2014

Uh oh!

glynos commented Jan 30, 2014

Uh oh!

ruslo commented Jan 30, 2014

Uh oh!

ruslo commented Jan 30, 2014

Uh oh!

glynos commented Jan 30, 2014

Uh oh!

Uh oh!