otp.net
otp.net copied to clipboard
saleyn
Fork or?
We are planning on making some very significant additions to this library (type mapping support, possibly moving all comms to iocp, wrap support for rx etc, move away from java styled code, etc). Would we be better to issue pull requests here or fork and continue there?
Cheers,
Greg
Actually, I recently did a complete rewrite of this library that soon will be published in open source. Here's the relevant blog: http://blog.aumcode.com/2013/10/nfx-native-interoperability-of-net-with.html
In that implementation I haven't used iocp yet, though it was the next thing I planned to address in the TODO list for this library as I recently finished another unrelated component that uses completion ports in another project, and they do perform better in asynchronous communications. However, the new implementation is much more efficient, has many bug fixes, contains a proper support of io server protocol, RPC, and is much cleaner and simpler.
I believe it'll be open sourced within 1-2 months as it's part of a larger library NFX that has a ton of other useful functionality. So at this point it doesn't make sense to fork from otp.net, but rather wait and fork from NFX, as I will discontinue support for otp.net in favor of the later.
If you can get that code up sooner than later I'd be happy to start committing on it (I have quite a few big things I can quickly drop in)
On Saturday, October 19, 2013, Serge Aleynikov wrote:
Actually, I recently did a complete rewrite of this library that soon will be published in open source. Here's the relevant blog:
http://blog.aumcode.com/2013/10/nfx-native-interoperability-of-net-with.html
In that implementation I haven't yet used iocp yet, though it'll be the next thing to address. However, the new implementation is much more efficient, has many bug fixes, contains a proper support of io server protocol, RPC, and is much cleaner and simpler.
I believe it'll be open sourced within 1-2 months as it's part of a larger library NFX that has a ton of other useful functionality. So at this point it doesn't make sense to fork from otp.net, but rather wait and fork from NFX, as I will discontinue support for otp.net in favor of the later.
— Reply to this email directly or view it on GitHubhttps://github.com/saleyn/otp.net/issues/2#issuecomment-26649585 .
Le doute n'est pas une condition agréable, mais la certitude est absurde.
Also my direct email is [email protected]
To be fair both models are needed as the thread per is actually quite a bit better in small deployments (say up to 25-50 nodes)
On Saturday, October 19, 2013, Greg Young wrote:
If you can get that code up sooner than later I'd be happy to start committing on it (I have quite a few big things I can quickly drop in)
On Saturday, October 19, 2013, Serge Aleynikov wrote:
Actually, I recently did a complete rewrite of this library that soon will be published in open source. Here's the relevant blog:
http://blog.aumcode.com/2013/10/nfx-native-interoperability-of-net-with.html
In that implementation I haven't yet used iocp yet, though it'll be the next thing to address. However, the new implementation is much more efficient, has many bug fixes, contains a proper support of io server protocol, RPC, and is much cleaner and simpler.
I believe it'll be open sourced within 1-2 months as it's part of a larger library NFX that has a ton of other useful functionality. So at this point it doesn't make sense to fork from otp.net, but rather wait and fork from NFX, as I will discontinue support for otp.net in favor of the later.
— Reply to this email directly or view it on GitHubhttps://github.com/saleyn/otp.net/issues/2#issuecomment-26649585 .
Le doute n'est pas une condition agréable, mais la certitude est absurde.
Le doute n'est pas une condition agréable, mais la certitude est absurde.
Just a ping back this is going a bit long. We would really prefer to contribute than to reimplement but we are quickly hitting that timeframe.
On Saturday, 19 October 2013, Serge Aleynikov wrote:
I'll keep you posted when we open-source NFX.
— Reply to this email directly or view it on GitHubhttps://github.com/saleyn/otp.net/issues/2#issuecomment-26655388 .
Le doute n'est pas une condition agréable, mais la certitude est absurde.
Unfortunately the legal stuff is making very slow progress at open sourcing the library. I expect it should take at least another month. A sneak peek at the list of classes is documented here: http://itadapter.com/nfxhelp/html/N_NFX_Erlang.htm
I would be happy to NDA and go to a private repo.
Cheers,
Greg
On Wed, Jan 8, 2014 at 3:58 PM, Serge Aleynikov [email protected]:
Unfortunately the legal stuff is making very slow progress at open sourcing the library. I expect it should take at least another month. A sneak peek at the list of classes is documented here: http://itadapter.com/nfxhelp/html/N_NFX_Erlang.htm
— Reply to this email directly or view it on GitHubhttps://github.com/saleyn/otp.net/issues/2#issuecomment-31832703 .
Le doute n'est pas une condition agréable, mais la certitude est absurde.
Pingback on this.
Also if you are still in nyc and want to grab a pint let me know as we will be in nyc in a few weeks.
On Wed, Jan 8, 2014 at 6:56 PM, Greg Young [email protected] wrote:
I would be happy to NDA and go to a private repo.
Cheers,
Greg
On Wed, Jan 8, 2014 at 3:58 PM, Serge Aleynikov [email protected]:
Unfortunately the legal stuff is making very slow progress at open sourcing the library. I expect it should take at least another month. A sneak peek at the list of classes is documented here: http://itadapter.com/nfxhelp/html/N_NFX_Erlang.htm
Reply to this email directly or view it on GitHubhttps://github.com/saleyn/otp.net/issues/2#issuecomment-31832703 .
Le doute n'est pas une condition agréable, mais la certitude est absurde.
Le doute n'est pas une condition agréable, mais la certitude est absurde.
That library should be available within a couple of weeks.
I am in NJ area, not far from NYC, but but regretfully extremely busy now.
Has iocp been implemented yet? This is the key for us. As I mentioned I have no problem at all implementing it it just seems silly to have duplicate efforts
On Fri, Mar 7, 2014 at 8:17 PM, Serge Aleynikov [email protected]:
That library should be available within a couple of weeks.
I am in NJ area, not far from NYC, but but regretfully extremely busy now.
Reply to this email directly or view it on GitHubhttps://github.com/saleyn/otp.net/issues/2#issuecomment-37051672 .
Le doute n'est pas une condition agréable, mais la certitude est absurde.
Hi, haven't seen any update for a while, is the NFX project still alive and going to be available soon?
I am wondering as well if not we will start on it if you want to join :)
On Sat, Jul 26, 2014 at 8:11 PM, Yan Cui [email protected] wrote:
Hi, haven't seen any update for a while, is the NFX project still alive and going to be available soon?
— Reply to this email directly or view it on GitHub https://github.com/saleyn/otp.net/issues/2#issuecomment-50241349.
Studying for the Turing test
The successor of the OTP.NET is part of the NFX framework that will be posted here: https://github.com/itadapter/nfx you can create an issue there to expedite committing the library to that repository.
I'd feel a lot more comfortable with continuing to wait (been about 9 months now) if there were anything at all in the repository
On Wed, Jul 30, 2014 at 4:20 AM, Serge Aleynikov [email protected] wrote:
The successor of the OTP.NET is part of the NFX framework that will be posted here: https://github.com/itadapter/nfx you can create an issue there to expedite committing the library to that repository.
— Reply to this email directly or view it on GitHub https://github.com/saleyn/otp.net/issues/2#issuecomment-50562580.
Studying for the Turing test