I've browsed the Modernizr docs for a property that tells whether Ecmascript 6' support for module scripts is available in the client or not, as Firefox currently doesn't. I found none. Am I missing something obvious or could I deduce this from a property that regards something else?
Yes,
but that's not available as built-in test in Modernizr as far as I know (Modernizr 3.5.0). So, you could use:
Modernizr.addTest('es6module', function(){
return 'noModule' in document.createElement('script')
});
In order for Modernizr.addTest to work, of course, you have to add addTest option to your build.
Testing script nomodule feature should be fine. That means, false positives and negatives are imaginable to happen.
Related
Is it possible to take screenshots inside Chutzpah?
I can't seem to get to the underlying phantom object.
I understand now looking through the chutzpahRunner.js that the test itself is called via page.open() so appears to be in a different scope.
In the end I'm hoping to diff svg elements similar to how PhantomCSS does it with casperjs.
I'm new to phantomjs / chutzpah so maybe I'm missing something obvious.
From the co-ordinator of Chutzpah:
This is not something Chutzpah supports. Chutzpah is focused on
running js tests. I think CasperJS is a better fit to what you are
looking for.
Though it would be a cool feature, it's being discussed in the Chutzpah forum: Screenshots via phantomjs
I have used pseudo-classes in the project, that i am currently working on, and since it is already javascript heavy I am trying to avoid using selectivizr. Is there are way making psedoclass compatible with IE8? Is it possible to use Modernizr to resolve this issue?
Any advise would be much appreciated.
You could check for css-lastchild with Modernizr (you'll find it in the Non-core detects section) and customize your styles if the browser doesn't support css-lastchild - which would mean it doesn't support any other of the CSS3 selectors either, which are:
:root
:nth-child()
:nth-last-child()
nth-of-type
nth-last-of-type()
:last-child
:first-of-type
:last-of-type
:only-child
:only-of-type
:empty
:target
:enabled
:disabled
:checked
:not().
I have custom frameworks in my app bundle for WebKit, WebCore, and JavaScriptCore. I would like all other frameworks that depend on the system versions of WebKit, WebCore, or JavaScriptCore to use my custom versions also. For instance my custom version of Webkit loads private system frameworks that in turn depend on the system version of WebKit. This means both my WebKit and the system WebKit get loaded, and usually a crash happens shortly after.
The way I understand this should be done, is to set the DYLD_FRAMEWORK_PATH environmental variable before your app bundle begins execution (Search order for loading frameworks can be found here: http://code.google.com/p/macdependency/wiki/SearchPaths). You can set environmental variables in code with setenv, but it won't take effect for the currently running process. I would have to re-launch the process again for it to take effect. I would like to avoid this too.
So my question is what is the best way to set the DYLD_FRAMEWORK_PATH before the execution of the my app bundle? This has to work in a release app bundle too. Is there a way to run a script whenever someone clicks on my app bundle before the executable starts running. Or is there any other suggestions out there?
Thanks in advance.
Jeff Wolski has the right idea by referring directly to Apple's documentation on the subject. This thread also provides excellent advice on how to get that going in Xcode, including corner cases associated with alternative methods of specification (for example, by using ~/.bash_profile directly for your user).
My reason for chiming in is you also asked for a script that might be able to assist you (and a bounty provider appears to have the same issue). It turns out Webkit references such a script in its documentation, which you might want to pick apart from the applicable Webkit source code. This should give you additional guidance on how, at least according to the developers, you should do this properly.
Best of luck with your project(s).
I would recommend environment.plist or the LSEnvironment key in the info.plist. Check out the link below.
https://developer.apple.com/library/mac/#documentation/MacOSX/Conceptual/BPRuntimeConfig/Articles/EnvironmentVars.html
WebKit currently has a script to do this for you, called run-webkit-app. See http://trac.webkit.org/browser/trunk/Tools/Scripts/run-webkit-app
Newer versions of ld understand the -dyld_env flag, which inserts a LC_DYLD_ENVIRONMENT load command into the binary (and as such applies essentially as early as possible in app execution). Perhaps adding -Wl,-dyld_env,DYLD_FRAMEWORK_PATH=/folder/encosing/WebKit/and/related/frameworks to your compiler flags might do what you want?
I have tried out the NPRuntime sample provided with the Gecko SDK 1.9.1 with the help of this link.
It works perfectly fine with Firefox 3.6.
Following the above steps I created a new scripting plugin (npmyplugin.dll). I am currently placing npmyplugin.dll in a folder other than Firefox/plugins. I am registering the associated mime-type and plugin path in the registry. The plugin gets loaded properly. But now the browser is not calling NPP_GetValue.
I was suspecting that it must be because I am placing the plugin in a directory other than Firefox/plugins. Hence I did the same with npruntime.dll (the sample which comes with Gecko sdk). Suprisingly npruntime.dll scripting was working. Please guide me whether I missing anything important while creating the new plugin.
NPP_GetValue doesn't get called to get the NPObject until the first time you access it with javascript in many cases. When you do a getElementById, it will make the call.
Likely other calls to GetValue are being made, but you aren't seeing them because you aren't handling them. However, that's still not guaranteed.
Sorry for replying so late. I somehow solved the problem. I was embedding the plugin using tag
then i wrote a small javascript code
var MyWorkingPlugin = document.getElementById('MyPlugin');
It seems that after executing the above javascript NPP_GetValue was called. Currently I have no convincing reason why it happened but it worked and I am happy for the moment. If anyone can explain the above behaviour please let us know.
I wrote a firefox plugin using C++ and used the <EMBED> to load it to html.
In javascript I got the embedded plugin by using document.getElementByID,
but when I tried to call a plugin function, the function was undefined. Moreover,
plugins constructors did not run while loading the page.
The same html file and plugin seems to work on some computers, while it doesnt work on others.
Does anyone have an idea what might by wrong?
How can I debug such an issue?
Can you reproduce it on your computer at first? If not, then try to figure out what systems, browsers, architecture, versions, they are using.
I recall, there are many GECKOSDK's, and each one has a specific version it works for. https://developer.mozilla.org/en/gecko_sdk For example, different SDK's for Firefox 1.5-2.0, FireFox 3.0, Firefox 3.5, and Firefox 3.5
For different browers, make sure you are exposing the right DLL Exports, browsers can vary: http://git.webvm.net/?p=npsimple is a good starting point. You can use firebreath.googlecode.com too as an alternative.
My suggestion would be to use an object tag instead of <embed>. In my experience, doesn't seem to work as reliably.
If I had more information on how your plugin is structured (are you using XPCOM or npruntime?), I might be able to help more.