Playing with the new(ish) url rewriting functionality for web forms, but I'm running into trouble trying to declare parameters as optional.
Here's the scenario. I've got a search function which accepts two parameters, sku and name. Ideally I'd like the URL for this search function to be /products/search/skuSearchString/nameSearchString. I also have various management pages that need to map to things like /products/management/ or /products/summary/. In other words, the last two parameters in the URL need to be optional - there might be one search string, or two, or none.
This is how I've declared my virtual URL:
Friend Const _VIRTUALURL As String = "products/{action}/{sku}/{*product}"
And added the following defaults:
Me.Defaults = New Web.Routing.RouteValueDictionary(New With {.sku = "/"})
Me.Defaults = New Web.Routing.RouteValueDictionary(New With {.product = "/"})
I have two problems with this setup. The most pressing is that the url seems to expect an sku parameter. So, /products/summary/ cannot be found but /products/summary/anyTextAtAll/ maps to the correct page. You get the same result whether the defaults are set to "/" or "". How do I ensure both sku and product parameters are optional?
The second is more a matter of interest. Ideally, I'd like the url to be able to tell whether or not it's got a product search string or a url search string. The obvious way to do this is to make one or the other default to a value I can just pick up and ignore, but is there a neater way of handling it?
I'm not sure I entirely understood the question, but I have some comments about what you've shown so far:
The manner in which you're setting defaults seems incorrect. You're first setting a default value dictionary with a value for "sku". You're then replacing the default value dictionary with a value for "product".
A default value of "/" is unlikely to be what you want. In this case it sounds like you want a default value of just "" (empty string).
Try something like:
Me.Defaults = New Web.Routing.RouteValueDictionary(New With {
.sku = "",
.product = "" })
My VB skills are rather weak, so the syntax I showed might not be exactly right.
I think that if you change both of these then you should be good to go.
Related
i have create module using module builder , now i am having a field called as book Name
now if i give same book name 2 time t is accepting .
i don't want to use and plug in for checking duplicate value because i want to learn the customization through code .
so i can call ajax and check in data base weather the same book name is exist in db or not but i don't know how controller works in sugar crm . and how to call ajax in sugar crm .
can any one guide me , your help is much appreciated .
If you really want to accomplish this using ajax then I'd recommend an entryPoint as the way to go. This customization will require a couple of simple things. First you'll write a little bit of javascript to perform the actual ajax call. That ajax call will post to the entryPoint you write. The entryPoint will run the query for you and return a response to you in the edit view. So lets get started by writing the entryPoint first.
First, open the file custom/include/MVC/Controller/entry_point_registry.php. If the folder structure and file do not exist yet, go ahead and create them.
Add the following code to the entry_point_registry.php file:
$entry_point_registry['test'] = array('file' => 'custom/test.php', 'auth' => true);
Some quick explanation about that line:
The index value of test can be changed to whatever you like. Perhaps 'unique_book_value' makes more sense in your case. You'll see how this value is used in a minute.
The file value in the array points to where you're gonna put your actual code. You should also give this a more meaningful name. It does NOT need to match the array key mentioned above.
The 'auth' => true part determines whether or not the browser needs to have an active logged in session with SugarCRM or not. In this case (and almost all) I'd suggest keeping this to true.
Now lets look at the code that will go in custom/test.php (or in your case unique_book_name.php):
/* disclaimer: we are not gonna get all crazy with using PDO and parameterized queries at this point,
but be aware that there is potential for sql injection here. The auth => true will help
mitigate that somewhat, but you're never supposed to trust any input, blah blah blah. */
global $db; // load the global sugarcrm database object for your query
$book_name = urldecode($_REQUEST['book_name']); // we are gonna start with $_REQUEST to make this easier to test, but consider changing to $_POST when confirmed working as expected
$book_id = urldecode($_REQUEST['book_id']); // need to make sure this still works as expected when editing an existing record
// the $db->quote is an alias for mysql_real_escape_string() It still does not protect you completely from sql injection, but is better than not using it...
$sql = "SELECT id FROM book_module_table_name WHERE deleted = 0 AND name = '".$db->quote($book_name)."' AND id <> '".$db->quote($book_id)."'";
$res = $db->query($sql);
if ($db->getRowCount($res) > 0) {
echo 'exists';
}
else {
echo 'unique';
}
A note about using direct database queries: There are api methods you can use to accomplish this. (hint: $bean->retrieve_by_string_fields() - check out this article if you wanna go that route: http://developer.sugarcrm.com/2012/03/23/howto-using-the-bean-instead-of-sql-all-the-time/) However, I find the api to be rather slow and ajax should be as fast as possible. If a client asked me to provide this functionality there's a 99% chance I'd use a direct db query. Might use PDO and parameterized query if I'm feeling fancy that day, but it's your call.
Using the above code you should be able to navigate to https://crm.yourdomain.com/index.php?entryPoint=test and run the code we just wrote.
However at this point all you're gonna get is a white screen. If you modify the url to include the entryPoint part and it loads your home page or does NOT go to a white screen there are 3 potential causes:
You put something different for $entry_point_registry['test']. If so change the url to read index.php?entryPoint=whatever_you_put_as_the_array_key
You have sugar in a folder or something on your domain so instead of crm.yourdomain.com it is located somewhere ugly and stupid like yourdomain.com/sugarcrm/ if this is the case just make sure that your are modifying the url such that the actual domain portion is preserved. Okay I'll spell it out for you... https://yourdomain.com/sugarcrm/index.php?entryPoint=test
This is more rare, but for some reason that I cannot figure out apache sometimes needs to be reloaded when adding a new entrypoint. If you have shell access a quick /etc/init.d/apache2 reload should do the trick. If you don't have shell access you may need to open a ticket with your hosting provider (or get a fricking vps where you have some control!!!, c'mon man!)
Still not working? Did you notice the "s" in https? Try http instead and buy a fricking $9 ssl cert, geez man!
Okay moving on. Let's test out the entryPoint a bit. Add a record to the book module. Let's add the book "War of Art" (no, not Art of War, although you should give that a read too).
Now in the url add this: index.php?entryPoint=test&book_name=Art%20of%20War
Oh gawd that url encoding is hideous right! Don't worry about it.
You should hopefully get an ugly white screen with the text "exists". If you do let's make sure it also works the other way. Add a 2 to the book name in the url and hopefully it will now say "unique".
Quick note: if you're using Sugar you're probably also using mysql which is case insensitive when searching on strings. If you really need case sensitivity check out this SO article:
How can I make SQL case sensitive string comparison on MySQL?
Okay so now we have our entryPoint working and we can move on to the fun part of making everything all ajaxical. There are a couple ways to go about this, but rather than going the most basic route I'm gonna show you what I've found to be the most reliable route.
You probably will need to create the following file: custom/modules/CUSTOM_BOOK_MODULE/views/view.edit.php (I hope by now I don't need to point out changing that path to use your module name...
Assuming this file did not exist and we are starting from scratch here is what it will need to look like:
if(!defined('sugarEntry') || !sugarEntry) die('Not A Valid Entry Point');
class CUSTOM_BOOK_MODULEViewEdit extends ViewEdit
{
public function display()
{
// make sure it works in the subpanel too
$this->useForSubpanel = true;
// make the name value available in the tpl file
$this->ss->assign('name_value', $this->bean->name);
// load the parsed contents of the tpl into this var
$name_input_code = $this->ss->fetch('custom/modules/CUSTOM_BOOK_MODULE/tpls/unique_book_checker.tpl.js');
// pass the parsed contents down into the editviewdefs
$this->ss->assign('custom_name_code', $name_input_code);
// definitely need to call the parent method
parent::display();
}
}
Things are looking good. Now we gotta write the code in this file: custom/modules/CUSTOM_BOOK_MODULE/tpls/unique_book_checker.tpl.js
First a couple of assumptions:
We're going to expect that this is Sugar 6.5+ and jquery is already available. If you're on an earlier version you'll need to manually include jquery.
We're going to put the event listener on the name field. If the book name value that you want to check is actually a different field name then simply adjust that in the javascript below.
Here is the code for custom/modules/CUSTOM_BOOK_MODULE/unique_book_checker.tpl.js:
<input type="text" name="name" id="name" maxlength="255" value="{$name_value}" />
<span id="book_unique_result"></span>
{literal}
<script type="text/javascript">
$(document).ready(function() {
$('#name').blur(function(){
$('#book_unique_result').html('<strong> checking name...</strong>');
$.post('index.php?entryPoint=test', {book_name: $('#name').val(), book_id: $('[name="record"]').val()}, function(data){
if (data == 'exists') {
removeFromValidate('EditView', 'name');
addToValidate('EditView', 'name', 'float', true, 'Book Name Must be Unique.');
$('#book_unique_result').html('<strong style="color:red;"> ✗</strong>');
}
else if (data == 'unique') {
removeFromValidate('EditView', 'name');
addToValidate('EditView', 'name', '', true, 'Name Required');
$('#book_unique_result').html('<strong style="color:green;"> ✓</strong>');
}
else {
// uh oh! maybe you have php display errors on?
}
});
});
});
</script>
{/literal}
Another Note: When the code detects that the name already exists we get a little hacky and use Sugar's built in validation stuff to prevent the record from saving. Basically, we are saying that if the name already exists then the name value MUST be a float. I figured this is pretty unlikely and will do the trick. However if you have a book named 3.14 or something like that and you try to create a duplicate this code will NOT prevent the save. It will tell you that a duplicate was found, but it will not prevent the save.
Phew! Okay last two steps and they are easy.
First, open the file: custom/modules/CUSTOM_BOOK_MODULE/metadata/editviewdefs.php.
Next, find the section that provides the metadata for the name field and add this customCode attribute so that it looks like this:
array (
'name' => 'name',
'customCode' => '{$custom_name_code}',
),
Finally, you'll need to do a quick repair and rebuild for the metadata changes to take effect. Go to Admin > Repair > Quick Repair & Rebuild.
Boom! You should be good to go!
I'd like for changes in the URL to drive my application, and for changes in the application to change the URL, but not actually change state.
I have a route like this. The country/city example is a bit contrived, hopefully that doesn't confuse things. The relationship in the real application is somewhat hierarchical. Child views don't seem a fit though because there's no need for nested views.
$stateProvider.state( 'viewMap', {
url: '/viewMap/:country/:city',
templateUrl: 'pages/viewMap/viewMap.html',
controller: 'ViewMapController'
};
In ViewMapController, I can construct the page based on $stateParams.country and .city. As these values change, my application reacts and I want the url to stay in sync. I don't want to reload the whole page, however. I just want to update the url and push a history state on to the stack.
I understand I could manually construct a string:
updateUrl = function() {
window.location.hash = '#/viewMap/'+ $stateParams.country +'/'+ $stateParams.city
}
This feels fragile, as the way I build the string is separate from the way the framework parses it. I would prefer for the framework to build me a string based on the current params, but $state.href('.') describes the current route, which doesn't include $stateParams that haven't yet been activated/navigated to.
I've also looked at reloadOnSearch, but I think it only applies to query params.
Is there a better way to model this? It feels like I'm fighting the framework over something simple.
You can pass state params to $state.href function to get the complete URL
$state.href('.', $stateParams)
To generate arbitrary urls you can pass non-current params and/or configuration:
$state.href('.', {country:'usa',city:'sf'}, {absolute:true})
I have a set of cascading dropdowns where the value in one defines the options for the next. For various reasons, it makes more sense for my application to represent this in the url path than url args. So I want urls like
mysite/#/Supermarkets/Produce/Apples
Where Supermarkets is not a dropdown but Produce and Apples are. The trick is that I don't know ahead of time that the path hash has three parts; the first part always exists, but the fact that there are two (or three or one or zero) dropdowns is only known after the first /Supermarkets/ part of the path resolves.
I'm not sure how to set this up with sammy. Does it even support this scenario?
You can make a route with parameters. The parameters then are accessed by the params object. I tried the code below which worked for me:
var app = $.sammy('header', function () {
//this allows us to use the sammy templates
this.use('Template');
//load header template
this.get('/Supermarkets/:type/:item', function (context) {
alert(this.params['type']);
alert(this.params['item']);
});
});
$(document).ready(function (){
app.run();
});
So now if the url is /Supermarkets/Produce/Apples I will get two alerts one for the type parameter which is Produce and one for the item parameter which returns Apples.
My problem is i want to add '.html' extension to my every blog post. For example right now i have a blog post url 'http://www.indu.com/blog/-/blogs/creating-a-custom-portlet-in-liferay' and what i want is 'http://www.indu.com/blog/-/blogs/creating-a-custom-portlet-in-liferay.html'
I would be grateful if any one can help?
I wonder why would you want this.
Anyways, I think you can change the urlTitle field of BlogsEntry and append .html to the string. May be use a service-wrapper-hook to append .html to the urlTitle whenever a blog is created or updated.
Or you can even use a ModelListener if it exists for BlogsEntry hook to update the blog's urlTitle by using onCreate & onUpdate methods.
Edit
The urlTitle field is present in the BlogsEntry table.
You can access this in java with the following methods:
BlogsEntry blog = BlogsEntryLocalServiceUtil.getBlogsEntry(90989); // retrieves the blog-entry
String blogUrlTitle = blog.getUrlTitle();
blog.setUrlTitle(blogUrlTitle + ".html"); // this would set the string
You can have a check for the blogUrlTitle so that you don't have repeated .html appended to the string:
if (!blogUrlTitle.contains(".html")) { // append only if the title does not contain `.html`
blog.setUrlTitle(blogUrlTitle + ".html");
}
You can refine your code as you like, the above is just a guide-line.
As a side-note, I would always try to reason with the client why they want something, this helps not only to fend-off bad-change-requests but also helps in giving an alternative to the client (which is less taxing on the developers like us ;-) ). In most cases this helps to understand the client's business better & provide better returns.
I am trying to load a rewrite rule based on a product's URL path.
I am using the loadByRequestPath() method in Mage_Core_Model_Url_Rewrite to accomplish this. However, no matter what I supply this method I get the following result (Check comment in code):
public function loadByRequestPath($path)
{
Zend_Debug::dump($path); // returns the path to my module
$this->setId(null);
$this->_getResource()->loadByRequestPath($this, $path);
$this->_afterLoad();
$this->setOrigData();
$this->_hasDataChanges = false;
return $this;
}
Here is my module code:
$productRewrite = Mage::getModel('core/url_rewrite') ->loadByRequestPath($product->getUrlPath());
Oddly, I get this back:
Array ( [0] => rewrites/getProductRewrites
[1] => rewrites/getProductRewrites/ )
Array ( [0] => 01003-product-name )
So loadByRequestPath() is getting called twice for whatever reason. $productRewrite still returns an empty object.
I have verified that $product->getUrlPath() returns the correct path. (As seen in the second array)
I am on Magento 1.6.1.
Your question is still a little unclear, so this answer might not address the specific problem you're seeing.
Magento's core team hasn't done a great job of communicating these sorts of things over the years, but loadByRequestPath is one of those methods that's best thought of as a "private api". Not in the OOP sense, but in the "this is a method used to implement core system functionality, and probably won't work like you think it should work, so use at your own risk".
The PHP code you're trying to use
$productRewrite = Mage::getModel('core/url_rewrite') ->loadByRequestPath($product->getUrlPath());
won't work with a default installation of Magento because the rewrite object doesn't have a store ID set. Trying something like this should work. (assuming the sample data, with an installed store object that has an ID of "1" and that the product in question exists in that store)
$productRewrite = Mage::getModel('core/url_rewrite');
$productRewrite->setStoreId(1);
$productRewrite->loadByRequestPath($product->getUrlPath());
The loadByRequestPath method assumes that a rewrite already has a store ID set, as it's part of Magento's larger dispatching process. (self-link to article describing the role of rewrites in Magento's routing system)
All that said, the problem you're describing is somewhat confusing. You say that
Zend_Debug::dump($path);
returns
an array that contains the path to my module
While I'm sure you know what the phrase "path to my module" means, it's a meaningless term in the larger magento universe. Being more specific about the literal value will help people understand what you mean.
Additionally, you also say
I have verified that $product->getUrlPath() returns the correct path.
but you're not clear on the value of "the correct path".
My guess would be the path you're seeing in Zend_Debug::dump is the call that's coming through as a part of the standard dispatch and not your later call using $product->getUrlPath(). However, the lack of clarity in your question makes that hard to tell.
If setting the store ID doesn't get you what you want, update your question with a full explanation of how you're running your code, and what you see displayed. With that information more people will be able to help you.