When a use accesses my application, I want to return a web page that I construct using brew. Right now I have an init() - this is the function that the user specifies. The function executes brew and creates a web page. It then returns the file name of the web page.
When I attempt the following ../library/xbrowser/R/init/file I get the following error reported to the browser:
cannot coerce type 'closure' to vector of type 'character'
In call:
as.character(x)
The documentation indicates that I could use the /file output as a means to create and return a web page. Am I using this feature incorrectly? Is there a better way to create and return a web page?
OK, I think my problem is an operator error.
I should have been testing the function with POST instead of GET which is, of course, the method used when you use a browser.
I don't get this message if I POST with curl - for instance.
My apologies for wasting anyone's time.
Peter
One more observation: don't use init as a function name; it seems to use some other namespace and complains about lazy-loading a database named 'P'
Related
I want to add "Weather: 24C" to the rental-listing component of the super-rentals tutorial app.
Where would be the "best-practices" place to put this ajax request?
Ember.$.getJSON(`http://api.openweathermap.org/data/2.5/weather?q=${location}&APPID=${apiKey}`)
.then(function(json) {
return JSON.parse(json).main.temp;
});
Do I need to add a component, add a model, add a service, add a second adapter, modify the existing adapter? Something else? All of these? Is the problem that the tutorial uses Mirage? I ask this because when I think I'm getting close, I get an error like this:
Mirage: Your Ember app tried to GET
'http://api.openweathermap.org/data/2.5/weather?q=london&APPID=5432',
but there was no route defined to handle this request.
Define a route that matches this path in your
mirage/config.js file. Did you forget to add your namespace?
You need to configure mirage to allow you making calls to outside in case mirage is active; what I mean is using this.passthrough function within mirage/config.js, that is explained in api documentation quite well.
Regarding your question about where to make the remote call is; it depends:
If you need the data from the server to arrive in case a route is about to open; you should prefer putting it within model hook of the corresponding route.
If you intend to develop a component that is to be reused from within different routes or even from within different applications with the same remote call over and over again; you can consider putting the ajax remote call to a component. Even if that is not a very common case usually; it might be the case that a component itself should be wrapped up to fetch the data and display it by itself for reusing in different places; there is nothing that prevents you to do so. However by usually applying data-down action-up principle; generally the remote calls fall into routes or controllers.
Whether using an ember-data model is another thing to consider. If you intend to use ember-data; you should not directly use Ember.$.ajax but rather be using store provided by ember-data and perhaps providing your custom adapter/serializer to convert data to the format ember-data accepts in case the server do not match to the formats that ember-data accepts. In summary; you do not need to use models if you use pure ajax as you do in this question.
I'm trying to use express-flash in a standard web express js app. I don't want to use session, because I want to do the app as stateless as possible, but when I try to use without session, the app show me this error:
req.flash() requires sessions
Can I use express-flash without session? Can I use other alternatives for this kind of messages?
Thanks.
Note: A flash message is a variable stored within a session that is only available once, for the next request. That is if we put a flash variable and renders a page, the flash variable is available but if we render the same (or other) page again the flash variable is not present (it is destroyed).
-- acanimal
Based on this premise, you need to have sessions to use message flashing.
One way I think you can accomplish what you want is to add an item to the request (req) object in your middleware, and then in your controller, check if the key exists. You can then pass a specific message to your template, assuming you're using a template engine or pass it as part of your response.
Hope this helps.
I'm putting together a desktop application in TideSDK and am having some trouble finding the parameters passed to the application via a custom launch URL.
The application launches when the appropriate protocol is invoked (call it aaa://), but I haven't been able to figure out how to grab the URL string. I read a couple of threads that suggested I could get the string with the Ti.API.application.getArguments() call, but it returns something odd (see below).
// launch application with aaa://some_args_here
var args = Ti.API.application.getArguments();
// returns (StaticBoundList) [ /path/to/app, "-psn_0_721072", ]
I'm not completely surprised that this doesn't seem to work, as the API documentation says the getArguments method returns a list of command line arguments.
Any insight as to how to access the launch URL would be appreciated!
have a look at window.location.search
https://developer.mozilla.org/en-US/docs/DOM/window.location
It doesn't appear that there's currently an "official" way to do this, so I came up with my own solution that seems to work pretty well (though currently its only implemented for OSX).
Searching through the TideSDK source, I found the place where the native application delegate is created. I added a URL launch handler in the app delegate, which stores the launch URL in a new app delegate member, and connected it to the API with a binding in Ti.UI.
If anybody else is interested in this functionality, or have insights as to how to accomplish this task for Windows, please get in touch!
Although I don't have your problem, just want to say the following function works fine for me.
var args = Ti.API.application.getArguments();
is it possible this is the problem of the urlprotocl registry?
my url protocol is set by this : MSDN
thus the value in "Command" is
"C:\YOUR_APP_FOLDER\YOURAPP.exe" "%1"
tested on win7 and winxp,
both successfully get the arguments.
I'm creating an external service of my rails app. This is always listening a rabbitmq queue and all their messages should be redirected to some methods of the rails controllers.
I've tried these approaches:
AMQP subscriber inside Rails app
What's the correct way to run one controller action from another controller action without an HTTP redirect?
http://www.misuse.org/science/2007/07/24/firing-rails-controller-actions-from-command-line-console/
The first one allowed me only to access the model (Anyway I suppose must be the base for what I want). The second one... never worked for me. And the last one doesn't works on rails 3 ( ActionController::Integration doesn't exists)
I think that the last approach still could be used if I figure out how the sessions are handled in rails 3. In any case, somebody had tried something similar before? Any suggestion will be appreciated.
Why not just send the request via an HTTP request? Your controller basically makes actions visible via URLs - simply making a request to the URL is essentially the same as calling the controller code, and it keeps your code working like a software service/API. You could even just use curl to do this if you want.
Otherwise, if the two files are in the same machine/folder, you could try to explicitly include the one controller code in the other, but that seems like maybe not the way to go. Depends on how you want the end result to work.
Short description
I'm getting used to CakePHP right now and am wondering about how to get more debug-information about what is happening inside the framework.
Let me please explain my situation a little more detailed
As you know CakePHP does a lot for you without putting you into the need to write additional code. One example is the handling of models.
I just created a model User and added validation-rules (no other methods). As described in the API methods like save will just work.
After that I created the needed controller and view to add a new user. When I try to add a user from the view I just receive the flash-message The user could not be created. Please, try again. No validation-violations are flashed.
I also did set the debug-level to 2: Configure::write('debug', 2); but do not receive any errors. The error.log inside \tmp\logs is also empty.
I really do want to learn how to solve those issues in the future.
So what else can I do to debug / display inner processes of cake?
Thank you very much for your help!
DebugKit is an official plugin that gives you lots of information on the request, queries and variables produced by Cake:
https://github.com/cakephp/debug_kit
You can also use trace() and other methods in the Debugger to show what is being executed in the background:
http://book.cakephp.org/2.0/en/development/debugging.html
Use a PHP IDE with an integrated debugger. That will allow you to follow execution line by line as it is executed and even inspect variable values as you go. Netbeans is a free one.