How to get HTML Elmement (or DOM) in Vaadin ?In GWT I can use as DOM.getElementById("myId");
I can set id attribute on my Vaadin components by setId() method. For example:
Button button = new Button("Say Hello");
button.setId("myButton");
So, how can I retrieve this DOM Element in Vaadin ?
You can use this:
public static Component findComponentById(HasComponents root, String id) {
for (Component child : root) {
if (id.equals(child.getId())) {
return child; // found it!
} else if (child instanceof HasComponents) { // recursively go through all children that themselves have children
Component result = findComponentById((HasComponents) child, id);
if (result != null) {
return result;
}
}
}
return null; // none was found
}
Source: https://vaadin.com/forum/#!/thread/3199995/3199994
Vaadin 10 (Vaadin Flow)
The new Vaadin Flow generation replaces the internal use of GWT for Web Components.
This new architecture provides us with easy direct access to the DOM from the Java-based server-side, if you so desire. You can read the DOM, and you can manipulate elements in the DOM. Read about the new Element API in the manual.
Vaadin 6, 7, & 8 (Vaadin Framework)
This Answer expands on the comment by Vaadin expert, Henri Kerola.
Vaadin is a server-side app framework. It's purpose is to shield the app developer from the details of HTML, CSS, JavaScript, DOM, GWT, HTTP, WebSocket, and such web technologies. The app developer writes in pure Java (and maybe a tiny touch of CSS for tweaking). Vaadin transparently and auto-magically generates the HTML-CSS-JavaScript-GWT-DOM necessary to render a representation of the app’s user-interface within a web browser.
So there is no way to access the DOM from that Java server-side, nor any need to do so generally.
If you want to take control of the web technologies then Vaadin is probably not the best framework for you.
In Vaadin8 you may try this:
JavaScript.getCurrent().execute("document.getElementById('refreshButton').click()");
Related
Say I have a AngularDart component that adds a div and an iframe to that div as it's template.
I have the element passed for the outer component in the components constructor
#Component(
selector: "input-html",
templateUrl: "packages/myproject/components/inputs/html.html",
useShadowDom: false
)
class HtmlComponent implements ShadowRootAware {
HtmlComponent(NgModel ngModel, Element element):super(ngModel, element){
}
}
I have shadowdom turned off because I'm using Bootstrap for styling and want the elements easily accessible for the bootstrap css.
My template code is along the lines of
<div>
<iframe id="my-iframe"></iframe>
</div>
It's more complicated than that, there's a bunch of buttons etc, as I'm porting a javascript html editor to angulardart.
My problem is, I need to get the iframe element, but whenever I query element.querySelector("#my-iframe") or even window.document.querySelector("#my-iframe") the object is null. I believe this is because the template hasn't been added to the DOM yet.
I need the iframe object because I need to set the iframe content for the HTML editor to work. There's a few other areas of my project that I wanted to get the template dom objects but couldn't either.
I've tried onShadowRoot, which worked in AngularDart 0.14 but no longer works in 1.0. I've tried ScopeAware and querying for the iframe when the scope is set, but that didn't work (ScopeAware fires before shadowroot event).
I have a hack that's messy that works, by using ng-show="init()" and in that init method I have
bool _initDone = false;
bool init() {
if(_initDone == false) {
iframe = element.querySelector("#my-iframe")
_initDone = true;
}
return true;
}
Which works, but it's messy and I don't like that solution and obviously isn't the correct way to do it.
Anyone know how I can achieve this in AngularDart 1.0?
I think onShadowRoot is the right place for the code to query the element. If it really doesn't work wrap it in a Future to add it as a task at the end of the event queue to delay it a bit more.
onShadowRoot() {
new Future(() {
querySelector(...);
});
}
On many places one can find usage of a function PF with Primefaces. For example in this answer
From what I have seen so far it seems to be a magic "make it work a little better" function. But I don't believe in this kind of stuff so:
What does this function do?
And where can I find documentation about it?
PF is a Javascript function.
In Primefaces 4.0 the Javascript scope of widgets changed. Prior to version 4.0 you could open a dialog widget with widgetVar.show();.
In Primefaces 4.0 and above the widgets are stored in a Javascript widget array. When you call PF('widgetVar') it is looking for the widget in the array and returning it.
PF=function(d){
var c=b.widgets[d];
if(!c){
if(a.console&&console.log){
console.log("Widget for var '"+d+"' not available!")
}
b.error("Widget for var '"+d+"' not available!")
}
return c
};
I could not find much on this either this is what I was able to decipher using Chrome's developer tools.
The PF function is a part of PrimeFaces's JavaScript API. It looks up a Javascript object that is the backbone of the JSF component on the client-side. Here is its definition (source):
PF = function(widgetVar) {
var widgetInstance = PrimeFaces.widgets[widgetVar];
if (!widgetInstance) {
PrimeFaces.error("Widget for var '" + widgetVar + "' not available!");
}
return widgetInstance;
};
PF is a shortcut for PrimeFaces.widgets['someWidgetId'], which just looks-up a Javascript object in global scope, and so the Javascript object can also be retrieved using window['someWidgetId'].
The PrimeFaces's Javascript API has no official documentation online, so to understand what you can really "do" with the Javascript object, you'll need to take a deep dive into PrimeFaces.
See also
"Intro To PrimeFaces widgetVar" blog post
PrimeFaces source code
For other Primefaces users coming here when upgrading to version 4.0 and above, it's possible to bypass the need to use PF('yourWidgetVar').someFunction() and just use yourWidgetVar.someFunction() directly as you would have before version 4.0. You just need the following configuration in web.xml:
<context-param>
<param-name>primefaces.LEGACY_WIDGET_NAMESPACE</param-name>
<param-value>true</param-value>
</context-param>
From the Primefaces User Guide:
Enables window scope so that widgets can be accessed using
widgetVar.method() in addition to default PF namespace approach like
PF('widgetVar').method().
Obviously you'd be susceptible to the namespace clash/pollution this feature was created to avoid, but it's useful if you want to migrate to a new version in little steps and isolate what incompatibilities the new version has introduced.
Mobile display modes in ASP.NET MVC 4 stop serving the correct views after about an hour of uptime, despite browser overrides correctly detecting an overridden mobile device.
Recycling the application pool temporarily solves the problem.
The new browser override feature correctly allows mobile devices to view the desktop version of a site, and vice-versa. But after about an hour of uptime, the mobile views are no longer rendered for a mobile device; only the default desktop Razor templates are rendered. The only fix is to recycle the application pool.
Strangely, the browser override cookie continues to function. A master _Layout.cshtml template correctly shows "mobile" or "desktop" text depending on the value of ViewContext.HttpContext.GetOverriddenBrowser().IsMobileDevice, but the wrong views are still being rendered. This leads me to believe the problem lies with the DisplayModes.
The action in question is not being cached:
[OutputCache(NoStore = true, Duration = 0, VaryByParam = "*")]
I am using 51Degrees for mobile detection, but I don't think this should affect the overridden mobile detection. Is this a bug in DisplayModes feature for ASP.NET MVC 4 Beta & Developer Preview, or am I doing something else wrong?
Here is my DisplayModes setup in Application_Start:
DisplayModeProvider.Instance.Modes.Insert(0, new DefaultDisplayMode("iPhone")
{
ContextCondition = context =>
context.GetOverriddenBrowser().IsMobileDevice
&& (context.Request.UserAgent.IndexOf("iPhone", StringComparison.OrdinalIgnoreCase) >= 0
|| context.Request.UserAgent.IndexOf("Android", StringComparison.OrdinalIgnoreCase) >= 0
|| !context.Request.Browser.IsMobileDevice)
});
/* Looks complicated, but renders Home.iPhone.cshtml if the overriding browser is
mobile or if the "real" browser is on an iPhone or Android. This falls through
to the next instance Home.Mobile.cshtml for more basic phones like BlackBerry.
*/
DisplayModeProvider.Instance.Modes.Insert(1, new DefaultDisplayMode("Mobile")
{
ContextCondition = context =>
context.GetOverriddenBrowser().IsMobileDevice
});
This is a known issue in MVC 4 (Codeplex: #280: Multiple DisplayModes - Caching error, will show wrong View). This will be fixed in the next version of MVC.
In the meantime you can install a workaround package available here: http://nuget.org/packages/Microsoft.AspNet.Mvc.FixedDisplayModes.
For most applications simply installing this package should resolve the issue.
For some applications that customize the collection of registered view engines, you should make sure that you reference Microsoft.Web.Mvc.FixedRazorViewEngine or Microsoft.Web.Mvc.FixedWebFormViewEngine, instead of the default view engine implementations.
I had a similar issue and it turned out to be a bug when mixing webforms based desktop views with razor based mobile views.
See http://aspnetwebstack.codeplex.com/workitem/276 for more info
Possibly a bug in ASP.NET MVC 4 related to caching of views, see:
http://forums.asp.net/p/1824033/5066368.aspx/1?Re+MVC+4+RC+Mobile+View+Cache+bug+
I can't speak for this particular stack (I'm still in MVC2) but check your output caching setup (either in your controllers or views - and in your web.config in your app and at the machine level). I've seen it work initially for the first few users and then a desktop browser comes in right around the time ASP decides to cache, then everyone gets the same view. We've avoided output caching as a result, hoping this would get addressed later.
If you want all mobile devices to use the same mobile layout you can use
DisplayModeProvider.Instance.Modes.Insert(1, new DefaultDisplayMode("Mobile")
{
ContextCondition = context =>
context.GetOverriddenBrowser().IsMobileDevice
});
And of course you need to make a view in the shared layout folder named _Layout.Mobile.cshtml
If you want to have a separate layout for each type of device or browser you need to do this;
DisplayModeProvider.Instance.Modes.Insert(0, new DefaultDisplayMode("Android")
{
ContextCondition = (context => context.GetOverriddenUserAgent().IndexOf
("Android", StringComparison.OrdinalIgnoreCase) >= 0)
});
DisplayModeProvider.Instance.Modes.Insert(0, new DefaultDisplayMode("iPhone")
{
ContextCondition = (context => context.GetOverriddenUserAgent().IndexOf
("iPhone", StringComparison.OrdinalIgnoreCase) >= 0)
});
DisplayModeProvider.Instance.Modes.Insert(0, new DefaultDisplayMode("Mobile")
{
ContextCondition = (context => context.GetOverriddenUserAgent().IndexOf
("IEMobile", StringComparison.OrdinalIgnoreCase) >= 0)
});
And of course you need to make a view in the shared layout folder for each named
_Layout.Android.cshtml
_Layout.iPhone.cshtml
_Layout.Mobile.cshtml
Can you not just do this?
protected void Application_Start()
{
AreaRegistration.RegisterAllAreas();
// Code removed for clarity.
// Cache never expires. You must restart application pool
// when you add/delete a view. A non-expiring cache can lead to
// heavy server memory load.
ViewEngines.Engines.OfType<RazorViewEngine>().First().ViewLocationCache =
new DefaultViewLocationCache(Cache.NoSlidingExpiration);
// Add or Replace RazorViewEngine with WebFormViewEngine
// if you are using the Web Forms View Engine.
}
So guys here is the answer to all of your worries..... :)
To avoid the problem, you can instruct ASP.NET to vary the cache entry according to whether the visitor is using a mobile device. Add a VaryByCustom parameter to your page’s OutputCache declaration as follows:
<%# OutputCache VaryByParam="*" Duration="60" VaryByCustom="isMobileDevice" %>
Next, define isMobileDevice as a custom cache parameter by adding the following method override to your Global.asax.cs file:
public override string GetVaryByCustomString(HttpContext context, string custom)
{
if (string.Equals(custom, "isMobileDevice", StringComparison.OrdinalIgnoreCase))
return context.Request.Browser.IsMobileDevice.ToString();
return base.GetVaryByCustomString(context, custom);
}
This will ensure that mobile visitors to the page don’t receive output previously put into the cache by a desktop visitor.
please see this white paper published by microsoft. :)
http://www.asp.net/whitepapers/add-mobile-pages-to-your-aspnet-web-forms-mvc-application
Thanks and Keep coding.....
Would it be possible to use a project-specific stylesheet for JIRA projects?
For example, if I would like to include project X in an iframe, I'd like to hide the logo and possibly the JIRA toolbar - for specific user groups for example (it's only for viewing purpose, it is not a security feature)
Granted that I'd have to implement this myself (through the webservice api for example) - are there templates for the standard issue page?
Thanks in advance!
There is a (currently undocumented) plugin point in JIRA for inserting top navigation components, <top-navigation>.
You can use this plugin point to add your own navigation bar, and perhaps hide the normal bar using an inline CSS stylesheet. The following example triggers this behavior by using a ?hideit=true query parameter, which is the simplest way to approach the "embed in iframe" problem. You could make that "sticky" by storing it in a session or cookie.
Once you have created a plugin that plugins into the <top-navigation>, hiding the top bar is simple. Here is a velocity script that does it:
#if ($hideHeaderHack)
<style>
\#header {display:none;}
</style>
HIDDEN (remove this message eventually)
#else
NORMAL (remove this message eventually)
#end
To create such a plugin, use the Atlassian Plugin SDK (use atlas-create-jira-plugin). Your atlassian-plugin.xml should look like:
<atlassian-plugin key="${project.groupId}.${project.artifactId}" name="${project.name}" plugins-version="2">
<plugin-info>
<description>${project.description}</description>
<version>${project.version}</version>
<vendor name="${project.organization.name}" url="${project.organization.url}" />
</plugin-info>
<top-navigation key="standard-navigation-top"
name="Tigerblood"
class="com.madbean.topnavhack.TopNav" state='enabled'>
<resource type="velocity" name="view" location="topnav.vm"/>
<order>5</order>
</top-navigation>
</atlassian-plugin>
Your top-navigation implementation class (called com.madbean.topnavhack.TopNav above) should look like:
public class TopNav implements PluggableTopNavigation {
private TopNavigationModuleDescriptor descriptor;
public void init(TopNavigationModuleDescriptor descriptor)
{
this.descriptor = descriptor;
}
public String getHtml(HttpServletRequest request) {
Map<String,Object> params = new HashMap<String, Object>();
params.put("hideHeaderHack", "true".equals(request.getParameter("hideit")));
return descriptor.getTopNavigationHtml(request, params);
}
}
Your plugin will be laid out something like:
./pom.xml
./src/main/java/com/madbean/topnavhack/TopNav.java
./src/main/resources/atlassian-plugin.xml
./src/main/resources/topnav.vm
Disclaimer I work for Atlassian as a developer in the JIRA team.
I don't believe this functionality is exposed directly, and you don't state what JIRA version you are using, but in 4.x in \atlassian-jira\includes\decorators there is a file called bodytop.jsp the has the following fragment that renders the top level navigation and toolbar elements:
// Render all the top nav plugins
for (Iterator iterator = topNavPlugins.iterator(); iterator.hasNext();) {
TopNavigationModuleDescriptor topNavModuleDescriptor = (TopNavigationModuleDescriptor) iterator.next();
PluggableTopNavigation pluggableTopNavigation = (PluggableTopNavigation) topNavModuleDescriptor.getModule();
%>
<%= pluggableTopNavigation.getHtml(request) %>
<%
}
%>
If you wanted to you could create a version of the dashboard rendering jsp that calls a modified bodytop.jsp that renders none of the usual nav elements.
I would be tempted to write a basic plugin to do this.
Take a look at http://confluence.atlassian.com/display/JIRA/Web+Resource+Plugin+Module
If you have yet to write a jira plugin, now might be the time to try it out http://confluence.atlassian.com/display/DEVNET/Developing+your+Plugin+using+the+Atlassian+Plugin+SDK .
I'm currently running Jira 4.2.2 and wrote a plugin that implements PluggableTopNavigation for a custom navigation bar. Unfortunately, this functionality, as detailed in the awarded question, is now depreciated.
My plugin added a div to the top of the Jira header that created a nice menu for use with our development pages. The source of the menu was hard-coded into the plugin and located as a static menu.html file on our server for sharing across different pages.
Since I'd have to completely redesign the plugin for Jira 5.2, I started searching for different ways to re-implement the menu. Here's what I settled on. It's not pretty, but it makes it so you don't have to write a plugin.
Change your announcement banner (quickly get there by typing 'gg', then search for announcement banner) to the following:
<script type="text/javascript">
jQuery(document).ready(function() {
jQuery.get("http://path.to.server/menu.html", function(data){
jQuery("#header").prepend('<nav class="global" role="navigation">'+data+'</nav>');
jQuery("#top-level-id-of-navbar a").css("color", "white")
});
});
</script>
Replace the menu.html link with your own link. The color of the header was inherited by the links in my menu, so I had to change them back to white after inserting the html page.
The result looks identical to Jira 4.2.2, so I'm happy.
So, after much research on whether or not we should the CEWP or the HTML Field Control on an external facing SharePoint site, we settled on using the Field Control (much thanks to AC). Now, we are having an issue that all the blogs I read say should not be an issue.
When we put a relative URL into the HTML Editor and hit OK, it is automatically changed to an absolute URL. This is apparently a "feature" of Internet Explorer from some of the research I have been doing. TinyMCE has a work around for this. I was wondering if there was some work around for the SharePoint control that I am missing.
This is kind of a big issue for us because we have an authoring site and the www site. So, when the authoring is done on the authoring site and all the links get migrated to the www site, they are http:// authoring.domain.com/en-us/Pages/... instead of /en-us/Pages/...
I encountered this issue as well. We had custom site fields and content types deployed via feature. The RichText property of the HTML Field is properly as true in caml, but once deployed the SPField in the root web fields collection and every Pages list the RichText attribute becomes false.
I was able to successfully resolve the issue by using a feature receiver on the feature that deploys the site columns and content types. My code loops every web in the site and then iterates over the fields to update them.
code snippet:
private void processweb(SPWeb web)
{
SPList list = web.Lists["Pages"];
SPField field;
for (int i = 0; i < list.Fields.Count; i++)
{
field = list.Fields[i];
//to work around a sharepoint defect ... make html fields work in richtext mode
if (field != null && string.Compare(field.TypeAsString, "HTML", true) == 0 && (field as SPFieldMultiLineText).RichText == false)
{
(field as SPFieldMultiLineText).RichText = true;
field.Update(true);
}
}
foreach (SPWeb w in web.Webs)
{
processweb(w);
}
}