Spoiler alert: this is NOW a question, so apologies to anyone that read it purely as a discursive topic :)
Anyway, I was doing a little research today re adding routes via javascript when i thought that a bit of google research wouldn't hurt. Basically, my aim was to do away with the following type of construct within my views:
and replace it with something akin to:
well, i lucked out a little today after finding this fantastic article (which isn't mine nor do i have any affiliation other than respect for the piece of work):
http://weblogs.asp.net/zowens/archive/2010/12/20/asp-net-mvc-javascript-routing.aspx
this really has been a missing link (or so i thought) for me when dealing with routes via javascript. However, the 2nd code example is misleading and actually won't produce what the example leads on. Can anyone suggest a fix for this and/or an alternative solution to allow this fluent convention of js routes within mvc views??
cheers...
[edit] - question edited 22:16 GMT to explore deeper options on this topic, plus changed title (removed OT portion).
So the question is why the second code example won't work as expected. Here's the answer, post currently doesn't return anything. This is an example of a certain developer not looking at the details of the code. When you use homePageUrl, the value will be undefined.
To actually get the home page URL, you'd do the following:
$.routeManager.action({controller:'Home', action:'Index'}).toUrl()
So, the moral of the story is that the code is a bit broken. The post action SHOULD return an object where you can put "toUrl()" right after the post is performed, like this:
$.routeManager.action({controller:'Home', action:'Index'})
.post(function(data){ alert(data); })
.toUrl();
I'll be fixing this bug in a bit!
Related
Trying to find help for this problem has taken me to a whole new one: complete lack of ressources, books and samples vor MVC 4 in VB.NET. I am having to choose between learn by experience (and the associated feeling of banging your head against a wall) or give it up and move to C# alltogether.
No company should ship a product if they are not willing to give it the same support as its sibling pruduct. They should drop VB for MVC completely or give us the means to learn it.
With that out of the way, here's my question. This line:
#Html.EditorFor(Function(x) x.UsersData(temp).Roles(Role))
Is a nice line of code. Works wonders. But How can I add a class to it, so I can change the style on my css files?
Well, it seems that this should do the trick:
#Html.EditorFor(Function(x) x.UsersData(temp).Roles(Role), New With {.class = "users-manage-check-box"})
But guess what, it doesn't. Ever. The result is the same.
What is wrong and how can I fix it?
And to be completely honest, I did come up with a solution. One that makes me feel dirty.
Looking at the output from that code, i see that the boxes classes are "check-box".
So what I've been doing is this:
#html.Raw(Html.EditorFor(Function(x) x.UsersData(temp).Roles(Role), New With {.class = "users-manage-textbox"}).ToHtmlString.Replace("check-box", "user-manage-checkbox"))
This feels wrong. So wrong. And not only is it a sad piece of code, it introduces security risks, which I'll have to fix before my solution is out of the development phase.
Any clues on why the additional view data is not working as it should? Am I getting something wrong? Am I asking too much?
Thanks a lot!
I don't think EditorFor allows that. So you need to create a custom editor template yourself.
You can read more about creating custom templates in this blog post
Update:
Take a look at the answer for this issue from http://aspnetwebstack.codeplex.com/
This behavior is by design.
The overload that you are calling accepts an object parameter called
additionalViewData (http://msdn.microsoft.com/en-us/library/ff406462).
The default implementation of EditorFor ignores this value. You would
have to write a custom editor template to be able to access that
information.
I am working on a project and was wondering if anyone else has messed with the view model much. Im looking for some examples on how to inject other view models i.e header,footer content.
I already have the template and layout path switching Im just trying to figure out the best way to handle if I select layout1 and it has 3 footer content fields, and a slider so how would put a class in front of the render to gather and inject required data
EDIT
Ok I guess I should clarify LOL .... slight ADHD where my mind wanders.
What should I listen for before injecting the viewmodels or can I do this in my onBootStrap() in my main module. I currently
$sites = $e->getApplication()->getServiceManager()->get('Application\Model\Sites');
$sr = $sites->getSiteByDomain();
Since many domains and/or subdomains can point to this and puts info in a session. Maybe im over thinking it and should just extend actionController like I did in ZF1
LOL PHP so many ways to do something ......
Thx for any pointers
There's two helpful links i can give you. One is the playground of Rob Allen alias Akrabat, you can find his playground right over here at github. The other one would be the official documentation which is nicely documented on this part.
If those don't help you, you should specify your question and show us what you've tried so far.
This is my most confused aspect that I continually have to ask people about. They have given me answers like 'if it works with data it goes in the model'... but to me that is pretty much ALL of it.
Does anyone have a better way to explain this?
Perfect example from my current code:
I have a listing of posts that can be favorited or not favorited. On the front end, i differentiate the different ones by dynamically adding class="favorite" to the HTML depending on if its a favorite or not.
So basically ... <li class="item<%= is_favorite?(current_user.id) %>">
Part of me thinks this should go in the model because its going to be running a statement to find a record that matches :resource_id and :user_id...
but another part of me thinks its going to be in the controller because its directly outputting the word " favorite" which is used in html
My second mini question is, is it the-rails-way to put methods that the controller uses in the same controller? as long as they remain un-routed, etc. Or is that not the right spot?
You're thinking properly.
The right answer is that you should use presenters or decorators.
See this railscast for inspiration.
I've got a need where each user can customize their own page on a replicated site. In grails it seems the most straightforward way to do this is:
somedomain.com/someController/JohnDoe
spelling out a controller, except this forces folks to type in a longer domain name, versus something like
somedomain.com/JohnDoe
Using sub-domains may be another approach, however they would need to be created automatically, i.e. when someone joins.
Can you please clarify the main ways Grails supports this kind of requirement/need (replicated site), and some of the pros/cons of each?
Thanks, Ray
Edit: Per Tomasz's edit below, the simplest course of action isn't clear. If you have insights on this please do share.
It is called UrlMappings in grails. You need to declare:
"/$username?" {
controller = 'someController'
action = 'user'
}
It redirects to someController, action user and optional variable called username.
This solution has one catch. Every one level path you visit passes this rule and takes you to someController. You cannot go to somedomain.com/books because it passes rule above and it follows you to someController#user with params['username']='books'. Then you can't use default actions. But if you decide that all your other paths have at least one slash, e.g. /books/list then you can follow this solution
Edit: I was wrong. It doesn't work as I've expected. I thought that UrlMappings are applied in order they are defined. That's not true, as explained here. Even worse - it's not documented (GRAILS-6246). Most specific explanation comes from Peter Ledbrook :
It uses a specificity algorithm, so the most specific match should apply
You must experiment then. I suggest you use safest solution and stick with /user/username solution.
Why is this:
http://MySite.com/Project/24/Search/32/Edit/49
preferred over this?
http://MySite.com/Project/24?Search=32&Edit=49
I'm not sure where your premise is coming from? It looks like an artificial example, which makes it hard to comment on.
A better comparison would be something like:
http://MySite.com/Project/24/Members/Edit
As opposed to:
http://MySite.com/Projects.aspx?id=24§ion=Members&action=Edit
Where, among other things, the hierarchy of entities is immediately obvious from the first example (ie, a Project contains Members). It also suggests that you can use other URLs that contain similar structures to the first (ie, /Projects/24 and /Projects/24/Members), so in that sense it's more concise.
If it comes down to actions that have a variable number of parameters, such as searching, then it's totally fine to use URL parameters as this will give you more flexibility, eg:
http://MySite.com/Projects/Search?name=KillerApp&type=NET
You could construct a URL using the first style, but you don't really gain anything, and managing the route could add unnecessary overhead:
http://MySite.com/Projects/Search/name/KillerApp/type/NET
I would argue that this (or any similar construction, eg if you removed the param names) suffers from an artificial hierarchy - the action in this case is really Search, and everything else is just a parameter of the Search, so it's in the same hierarchy, not some "sub" hierarchy.
Not really a fair comparison. The style allows you to drop the GET parameter names, so the routed one should read something like
http://MySite.com/Project/24/32/49
It's really an aesthetic improvement, though -- it's both neater-looking, and easier to type or read out to someone.
Its mostly a human readability issue, although (since most search engine ranking algorithms are not publically disclosed), it is believed to have SEO value as well.
In the example case, it may not be any better. But it's a Search Engine Optimization in general. Here are some SEO best practices -- from that article ...
Ideally, the URL structures should be
static, and reveal what the page is
about. A simple and clear URL
structure is much easier for both
search engine spiders and human
beings.
Easier to remember as well. It's easier for a user to remember /Employee/1 to get the information for employee #1 rather than understand a querystring. Not a reason to use it but I think its a small improvement.