incubator-callback-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Brian LeRoux...@brian.io>
Subject Re: [DISCUSS] Ripple Emulator CLI
Date Tue, 16 Oct 2012 03:47:11 GMT
Cool, checking that I didn't mess something up!

What if, as a first step, we just made the Ripple cordova.js avail via
the static server? (Not sure how feasible this is given the entirety
of the Ripple UI but the mocked cordova env is the first big win given
dev tools are then avail.)

On Mon, Oct 15, 2012 at 6:21 PM, Dan Silivestru
<dan.silivestru@gmail.com> wrote:
> Hey Brian,
>
> Yup, static server. But if you go to localhost:4400/ripple/enable that
> is essentially emulate.phonegap.com. Also have ripple proxy for cross
> origin xhr. Really this is just the foundation for much more :)
>
> This is just the first step :)
>
> Dan Silivestru
> +1 (519)-589-3624
>
> On 2012-10-15, at 17:19, Brent Lintner <brent.lintner@gmail.com> wrote:
>
>> Hey Brian (and all),
>>
>> Dan is out of the country (vacation) for the next week, so it is just me at
>> the moment. :-)
>>
>> First off, I squashed all the commits in that branch into a manageable
>> amount, and tried to make a very detailed message on the why/what/where,
>> that will hopefully explain the gist of the functionality. There is also
>> this Pull Request (as a reference) ->
>> https://github.com/blackberry/Ripple-UI/pull/595/files
>>
>> As for `emulate`, yeah, currently it is (pretty much) a static web server
>> (first step), except (if I am not repeating what is already known) there is
>> also an (http) route bound to it that can be used to do various things like
>> proxying cross origin XHR calls locally, vs shelling out to a remote web
>> service like we do now), same with the serving up of 'ripple/enable/'.
>>
>> Definitely like the idea of pumping device events/data to Ripple, btw! I
>> know one of the guys who used to be on our team did something similar with
>> an installed app that pumped Accelerometer data via websockets to Ripple to
>> test a game with (vs using the virtual phone to manipulate accel data) that
>> never made it in- it was epic.. :-D
>>
>> And.. as another idea to add onto what Dan was saying earlier about
>> injecting the UI & code via webserver middleware (or at least exploring
>> it), I also think (as we have discussed) something like a "remote web
>> inspector" style approach would be really cool (if not useful). I.e. where
>> the UI is (instead) separated from the core emulation bits (that get
>> injected when you app is loaded), and your app can run normally (i.e.
>> running as an installed app, in a mobile browser, in another desktop
>> browser (tab), whatever).. so when any data that is changed via Ripple's UI
>> (or from wherever), it is propagated to the emulated API code in the app
>> (wherever it is). Sort of like the above idea, but reverse. A thought.. If
>> that makes sense.. :-)
>>
>> Brent - \m/
>>
>> On Mon, Oct 15, 2012 at 5:14 PM, Brian LeRoux <b@brian.io> wrote:
>>
>>> +1 Duck Wings
>>>
>>> * * *
>>> Dan, I installed now and when I run `ripple emulate` I'm just getting
>>> a static file server?
>>>
>>> On Sun, Oct 14, 2012 at 6:08 AM, Michal Mocny <mmocny@chromium.org> wrote:
>>>> On Sat, Oct 13, 2012 at 7:27 PM, Dan Silivestru <
>>> dan.silivestru@gmail.com>wrote:
>>>>
>>>>> well, that's embarrassing :( you're right I got the path wrong.
>>>>>
>>>>> Really really sorry about that error, I staged the commit on my end
>>>>> changing our package name back to ripple from ripple-cli and forgot to
>>>>> push... Got really excited about going to have duck wings for dinner
:-)
>>>>>
>>>>
>>>> Duck wings?  Who _wouldn't_ be excited?!  How were they?
>>>>
>>>>
>>>>>
>>>>> Want to give it another shot now?
>>>>>
>>>>> As for the rest. we are already thinking of a ripple middleware that
>>> would
>>>>> allow us to "inject" the ripple ui and code before the content is
>>> served up
>>>>> to the browser. That will only work for locally hosted projects but the
>>>>> upside would be that we would no longer depend on the chrome extension
>>> in
>>>>> those cases... and as a result... we can become browser independent :-)
>>>>>
>>>>> This is just the foundation for our CLI, the sky is the limit of where
>>> we
>>>>> go from here.
>>>>>
>>>>> On Sat, Oct 13, 2012 at 6:07 PM, Brian LeRoux <b@brian.io> wrote:
>>>>>
>>>>>> Dan: awesome! Echoing Mike, it has always been idea (mostly discussed
>>>>>> over beers) that the default implicit behavior for `emulate` be Ripple
>>>>>> as soon as we could get it hosted/browser based. (And make so mobile
>>>>>> operating system emulator invocation is explicit.) Zero dependency
dev
>>>>>> workflow. (Excluding Node and a text editor of course. =)
>>>>>>
>>>>>> Not to get too far but having a little phonegap app that proxies
>>>>>> device events back to a socket server that propagates to Ripple would
>>>>>> sure be cool.
>>>>>>
>>>>>> ****
>>>>>> Tried following the instructions you provided but failed at step
4
>>>>>> ---there is no pkg/rim-cli.js ...I think you meant: sudo npm install
>>>>>> -g pkg/rim.npm/ ??
>>>>>>
>>>>>> Once I do install I get a node error--->
>>>>>>
>>>>>> Error: Cannot find module 'ripple/cli/help'
>>>>>>
>>>>>> (Uh, do you guys want me to file bugs on the github for this stuff?)
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Sat, Oct 13, 2012 at 12:30 PM, Dan Silivestru
>>>>>> <dan.silivestru@gmail.com> wrote:
>>>>>>> Hey Mike,
>>>>>>>
>>>>>>> Thanks, great idea. I saw that command, but didn't want to assume
>>> that
>>>>>>> would be the right place for Ripple :)
>>>>>>>
>>>>>>> I really like the idea of adding Ripple as another platform or
>>> emulator
>>>>>>> type.
>>>>>>>
>>>>>>> Cheers,
>>>>>>>
>>>>>>> Dan.
>>>>>>>
>>>>>>> On Sat, Oct 13, 2012 at 3:04 PM, Mike Reinstein <
>>>>>> reinstein.mike@gmail.com>wrote:
>>>>>>>
>>>>>>>> Hey Dan,
>>>>>>>>
>>>>>>>> Pretty cool! The cordova client has a command called emulate,
I
>>> could
>>>>>> see
>>>>>>>> ripple being a great option here. Maybe we could have a a
way to
>>>>> specify
>>>>>>>> ripple as the emulator type. Maybe it makes sense to have
that as
>>> the
>>>>>>>> default fallback if the platform emulators arent present.
>>>>>>>>
>>>>>>>> -Mike
>>>>>>>>
>>>>>>>> On Sat, Oct 13, 2012 at 3:01 PM, Dan Silivestru <
>>>>>> dan.silivestru@gmail.com
>>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> Hi All,
>>>>>>>>>
>>>>>>>>> Thought I would share some of what we've been up to on
Ripple
>>> with
>>>>> the
>>>>>>>>> group here, since I think it could mesh really well with
the CLI
>>>>> work
>>>>>>>>> that's being done for Cordova.
>>>>>>>>>
>>>>>>>>> We've been working on a CLI for Ripple. All work can
be found in
>>>>> this
>>>>>>>>> branch:  https://github.com/tinyhippos/Ripple-UI/tree/server.cli
>>>>>>>>>
>>>>>>>>> It's not on npm yet as we have to work through some internal
>>>>>> processes to
>>>>>>>>> make that happen and there already is a ripple npm package
out
>>> there
>>>>>>>>> (hopefully we can sort thing out, or we might go with
>>>>>>>>> npmjs.org/packages/ripple-cli)
>>>>>>>>>
>>>>>>>>> But you can easily install the CLI by doing the following:
>>>>>>>>>
>>>>>>>>> 1. clone our repo and checkout the server.cli branch
>>>>>>>>> 2. run ./configure
>>>>>>>>> 3. run jake
>>>>>>>>> 4. npm install -g pkg/rim.cli (in the future that will
be npm
>>>>> install
>>>>>> -g
>>>>>>>>> ripple OR npm install -g ripple-cli)
>>>>>>>>> 5. accept the EULA (url points to the wrong spot right
now, this
>>> is
>>>>> a
>>>>>>>> work
>>>>>>>>> in progress :) )
>>>>>>>>> 6. type in: ripple help
>>>>>>>>> 7. play around :-)
>>>>>>>>>
>>>>>>>>> Would really love to get the community's feedback on
this and
>>> see if
>>>>>>>> there
>>>>>>>>> might also be some interest in adding a command to the
cordova
>>>>> CLI...
>>>>>>>>> something along these lines:
>>>>>>>>>
>>>>>>>>> cordova test [--port xxxx] [--path pathToTest]
>>>>>>>>>
>>>>>>>>> The idea is that this will enable an http server over
the current
>>>>> dir
>>>>>> or
>>>>>>>>> the specified path and Ripple will auto enable once the
provided
>>> URL
>>>>>> is
>>>>>>>>> pasted into Chrome.
>>>>>>>>>
>>>>>>>>> Looking forward to hearing everyone's thoughts.
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Dan Silivestru
>>>>>>>>> +1 (519) 589-3624
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Dan Silivestru
>>>>>>> +1 (519) 589-3624
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Dan Silivestru
>>>>> +1 (519) 589-3624
>>>>>
>>>
>>
>>
>>
>> --
>> Brent

Mime
View raw message