I have deployed the play framework 2.0's sample websock-chat app onto cloud foundry, but the part after you join the chatroom doesn't display anything. I wondered if this was a websocket support issue, because the app works fine locally.
Yes, CloudFloundry supports Websockets, thanks to its gorouter component.
Support is also available on the PWS instance, on wss: port 4443, see official documentation.
Answering my own question, it appears according to https://cloudfoundry.atlassian.net/browse/CF-45, that the answer is no.
Now with latest updates in CFv2 it has a support for websockets (see this line in the code of gorouter, which is responsible for redirecting requests to your apps). I've written a proof-of-concept app in Node.js (for complete example see my answer here) and it is working fine for my local CF deployment, so I expect it would work for other frameworks and languages also.
However it is not yet working on https://console.run.pivotal.io/, as I suppose because they are not updating deployment of CF with each commit. Also there are some issues with IE 10 and Firefox (which might be CF issues), but in Chrome it works as expected.
if its websockets , then I'll suggest you must go for Jelastic , its easy and reliable.
http://jelastic.com/docs/websockets
Related
So I have multiple clients using an app built in electron. The entire application is actually a number of electron windows that talk to each other. When I have an update for the client side code (html/js/css) I have to have them shutdown, and run a utility that downloads from our internal server to update their app. I would like to know if there is a way I can either push new code to the clients (maybe through a socket) to overwrite the old code, or even maybe poll our 'code server' for updates, and then have it automatically update/overwrite existing code.
Is this possible? Is there functionality built in to electron that allows this?
And, if possible, how can it be accomplished? i.e. is there a library I can look at that will help me?? (i found a filesaver.js library, but its not exactly what I need). Thanks in advance.
You can have your Electron app load all code from a server every startup and cache this locally. You can do this by simply hosting your electron code on a web server and pointing Electron at the URL. You could make the app work offline by using a Service Worker.
This isn't a great idea though as code loaded from the internet will have access to all the node APIs. You will have essentially made a DIY botnet and securing it from abuse can be tricky.
You should read Security, Native Capabilities, and Your Responsibility in the Electron docs and be sure you understand the implications.
You can use the built-in autoUpdater of Electron: https://electronjs.org/docs/api/auto-updater
You need a server the autoUpdater can talk to, to download the updates from. The updates are installed after download.
You can host a server by yourself or use a service like https://www.update.rocks/
What you need is electron updater and you can use electron-builder for that.
I loved with loopback. But I am new with fireloop. fireloop working fine with angular 4 webapp client but when I try to use fireloop with nativeapp client (nativescript) its SDK only work for REST API. its Real-time feature not working.
debug logs only show 'Creating a new connection with: '
for example:
Creating a new connection with: http://192.123.1.13:3000
only this just logging in console.
I need some guide to settings up fireloop real-time feature with nativeapp.
sorry for my language.
I think it has to do with the websocket api. I presume for the real time you might actually be using that. Its is related to a question I asked earlier regarding angularfire2 and nativescript.
Here is one response: angular-fire2 is expected to run in a browser/server. Mobile devices have different specifications, more limitations, they work differently. Mobile SDKs are also optimized and made to work with the device's operating system and application lifecycle events. The same does not apply to node.js backends.
angular fire2 and nativescript
Since CodenameOne doesn't support "the cloud storage API" any more and the parse.com is going to retire soon as well. Does CodenameOne has any plan to release a new Cloud Storage API or provide suggestions/guidelines to help developers to deal with the parse4cn1 library code, cloud code, database structure and data in parse.com?
That is something you will have to figure out yourself as parse4cn1 was initially contributed by a community member and wasn't developed by Codenameone team.
You can use a simple webservices created in php, python or java, hosted along your content with any ISP.
You may also have a look at amazon aws which is promising, they provide a cloud solution but their SDKs is not yet integrated to Codenameone.
I made the parse4cn1 lib and I'm also wondering what's smartest to do. With the announcement of Parse.com's imminent shutdown, there's been a lot of discussion around alternatives. My feeling is that "the dust is yet to settle" as per what options are best and reliable for the longer term (it would be a pity to migrate to another service only for it to be shut down soon). So I personally plan to wait till sometime in Q2 to do a proper evaluation of the alternatives. Hopefully, there'll be more clarity then.
The option to host one's own Parse server (e.g. on AWS or Heroku) is getting interesting. They recently announced support for push notifications on iOS and Android. If (when?) they open source the Parse.com dashboard code, I think that option would be much more interesting.
At some point in the coming months, I plan to make a parse4cn1 release that exposes an option to set the server path. With that, anyone migrating to the Parse server option should, in principle, be able to continue to use the cn1lib. Of course, for features that are supported by the open source Parse server.
PS: Here are pointers to some of such discussions on Parse alternatives:
https://github.com/relatedcode/ParseAlternatives
http://www.slant.co/topics/5219/compare/~firebase_vs_kumulos_vs_kinvey
I'm evaluating using SignalR in an app hosted at AppHarbor running on 2+ instances (web workers) but reading around it looks like thsi won't work:
SignalR wiki says that scaling in a web farm is still in development (and 2+ web workers sounds like web farm to me). Another question here on StackOverflow says it won't work on more than one iss server.
On the other side, on AppHarbor support site they say it works great without giving much info thought (didn't answer to all questions like # of simultaneous connection, limits of load balancers etc).
Can someone confirm if SignalR is the right path to take on AppHarbor?
Thanks!
David Fowler is working on a Redis message store for SignalR. The code is on Github and I believe it is what will let SignalR apps scale to multiple AppHarbor instances.
Updated links for 2014:
SignalR's Redis page on GitHub, which is currently just a URL to a Redis doc on ASP.Net website.
Is there any library that allows to handle Skype messages using Ruby?
Take a look at this: http://rubyforge.org/projects/skyperapper
If you are using JRuby http://skype.sourceforge.jp/ might be a better option
For the public API doc: http://developer.skype.com/accessories
You can use the RestClient gem and the tropo.com service to send and receive Skype IMs. See the documentation for their REST api here:
https://www.tropo.com/docs/scripting/message.htm?search=skype#anchor
They're currently not charging for the service that sends/receives instant messages.
There are currently two ways of accessing the Skype network in a sanctioned manner. The first is the Skype Public API, which requires talking to the official Skype client running on a local machine. The other is SkypeKit which I am led to believe has a very restrictive agreement attached to it.
SkypeKit for ruby is available as a gem here.
The following is historical only; Skype no longer supports the Public API for their client. I have also been working on a Ruby wrapper around the Skype Public API. It's far from finished at the moment but supports Windows and Linux currently (OSX should be easy to implement as well, however I currently don't have a Mac) and allows you to connect and send commands to Skype manually. Long term the entire API will be wrapped.
In the interests of full disclosure, there seem to be a few other projects hanging around too, however they seem to only target a single platform and the ones I found seem to be abandonware at the moment.
libpurple have a plugin for Skype, but this is not easy to Setup.
Unfortunately, Skype protocol is closed & encripted, so there are no good libs out there.