|IRC Nick||Time (GMT-7)||Message|
|TheSHEEEP||02:02||Hey, guys. I've got a "small" problem with building FB and using the examples. I've posted it all on the forums: http://forum.firebreath.org/topic/247/ . Would be cool if someone could help me :)|
|linearray||03:02||I wonder why it tries to include a Mac specific file on windows
that doesn't look right
|TheSHEEEP||03:02||what exactly do you mean?|
|TheSHEEEP||03:02||Ah, now I see
Indeed, doesn't look right.
|linearray||03:02||Does building your own plugins work?|
|TheSHEEEP||03:02||I tried creating an "empty" project using the python script. It had the same problems, yes.
I also noticed that all cases where the wrong tag was, it is either a X11 or Mac ClInclude
Yep, just deleted the examples and ran "prep2010 examples" again, and in all files where such an error occurs it is when including a Mac or X11 file that has sub-nodes.
|linearray||03:02||oh, there's a new ticket about this
|FireBreathBot||03:02||FIREBREATH-166: Summary: Firebreath 1.6 will not build in Visual Studio 2010
FIREBREATH-166: Assigned To: richard
FIREBREATH-166: Priority: Major, Status: Open, http://jira.firebreath.org/browse/FIREBREATH-166
|linearray||03:02||are you running HEAD or 1.6?|
|TheSHEEEP||03:02||Yeah, I'll add that to the thread. Anyway, you can fix that pretty easily and it still doesn't work for me. It does build the *.dlls, though.
That one: https://github.com/firebreath/FireBreath/zipball/master
|Nalu||03:02||Hi, Can anyone please tell how to render an opengl image on browser.
Do i need to install glut?
Thanks in advance
Hi, Can anyone please tell how to render an opengl image on browser. Do i need to install glut? Thanks in advance
|EvilTengil||06:02||I've implemented OpenGL rendering from a separate thread. However the plugin crashes in Chrome whenever the Plugin is being occluded by the scrollbox in the browser (that is, the visible region of the plugin is changed!)
I think Chrome is destroying the window handle and creates a new one without telling me about it...
Anyone with experience of this issue?
|dougma||06:02||are you monitoring onWindowAttached/onWindowDetached messages?|
|pingvishal||07:02||Hi, am new to firebreath and was looking around for ways to develop a plugin to access the usb port through the browsers. Is this possible with Fb ?|
|dougma||07:02||umm... you'll probably need a device driver first.|
|pingvishal||07:02||am planning to interact with mobile phones when they are connected in mass storage mode. would i still need any device driver apart from what windows already provides ?|
|dougma||07:02||no, you access it as a storage device as per normal.|
|Nalu||07:02||Hi, Can anyone please tell how to render an opengl image on browser. Thanks in advance|
|dougma||07:02||Nalu: maybe EvilTengil can help you.
Nalu: why not use webgl? :)
|pingvishal||07:02||Dougma: yeah i thought so. which APIs should i look at within FB ?|
|dougma||07:02||well, you'll just use regular old c++ to access the filesystem won't you?
there is some degree of sandboxing going on with some browsers... so i think you'll just have to try it and see.
|pingvishal||07:02||i was also looking at accessing the USB descriptors, events|
|dougma||07:02||well i can tell you ie9 runs the plugin in a separate process at low integrity level if that helps.
you may find you can't get the events you need
i'd advise a regular old desktop app. :)
|pingvishal||07:02||:) i had thought about that but it'd defeat the purpose.|
|dougma||07:02||what is the purpose?|
|pingvishal||07:02||want to try displaying the contents on the phone and a way to add/delete files. more like a webpage where people could visit and do it using the browser.|
|dougma||07:02||they still have to download and install your plugin|
|pingvishal||07:02||the idea being some of the content will be available from the web portal|
|dougma||07:02||you could always bundle an .exe with your plugin
install a service or something to assist your plugin
|pingvishal||07:02||thats true .. but usually the trusting a browser plugin is easier than a desktop app|
i don't trust em
|pingvishal||07:02||how can i understand the service-plugin interaction ?
|dougma||07:02||how can you understand?|
|pingvishal||07:02||am new to this and hence if there are any place where i can learn about it, would be grateful.|
|dougma||07:02||communicating from service to plugin is some form of inter-process-comms. firebreath has nothing to help you there.
but we made be over-thinking.
just try it from the plugin if that's what you want to do.
anyway, i gotta go. good luck.
|pingvishal||07:02||ok. let me explore.
|EvilTengil||08:02||Nalu: It's not completely trivial to render with OpenGL in the browser. Be prepared for a journey full of crashes if you want to implement it :)
There is some horrible example code for how to set up opengl on windows linked from the docs. However, at least for me it doesn't work in Chrome when the plugin container get's occluded in some way.
|taxilian||09:02||dougma: surprisingly most USB devices you can access just fine w/ low integrity mode; mass storage devices may be different, however|
|FireBreathBot||10:02||JIRA issue http://jira.firebreath.org/browse/FIREBREATH-167 issue created by georg|
|FireBreathBot||10:02||Could not find any tags matching JSAPI.h|
|FireBreathBot||10:02||Found 2 matching file(s) in the master branch. First 2 are:
|FireBreathBot||11:02||JIRA issue http://jira.firebreath.org/browse/FIREBREATH-167 issue commented by georg "Repro:|
|FireBreathBot||11:02||Found 2 matching file(s) in the master branch. First 2 are:
|linearray||11:02||not quite true|
|linearray||11:02||that the API is copyable
at least JSAPI derives from noncopyable
|taxilian||11:02||yeah, we're discussing it
he's getting an error that looks like it's not
but probably it generates the constructor for the subclass and generates an error before it even gets to the base class
|linearray||11:02||some people reported problems with prep2010.cmd
I'm going to look at it
|linearray||11:02||I couldn't find a change that might have caused this|
|taxilian||11:02||trying to fix jira right now, though =]|
|taxilian||11:02||actually; they probably need to upgrade cmake
I have seen the error before
|taxilian||11:02||could be something new, but that's my guess|
|FireBreathBot||11:02||JIRA issue http://jira.firebreath.org/browse/FIREBREATH-166 issue commented by richard "What version of cmake are you on? Can you try upgrading? I've seen this error before with an old..."|
|FireBreathBot||11:02||JIRA issue http://jira.firebreath.org/browse/FIREBREATH-167 issue assigned by georg
JIRA issue http://jira.firebreath.org/browse/FIREBREATH-167 issue updated by georg
JIRA issue http://jira.firebreath.org/browse/FIREBREATH-167 issue updated by georg
|taxilian||12:02||dougma: your variant patch looks good to me; I'm going to run a refresh test on it for a couple of hours, then probably go ahead and pull it in|
I want to know if I can build a plugin for chrome that add an extra text box in browser top bar
|reichi||13:02||that would be an extension
and not npapi plugin
ron__: you want to read this: http://colonelpanic.net/2010/08/browser-plugins-vs-extensions-the-difference/
|aakash||14:02||For a downloader should I use - https://github.com/taxilian/firebreath-curl/ or http://www.firebreath.org/display/documentation/Handling+downloads+using+FireBreath|
|taxilian||14:02||aakash: it kinda depends on what you need in a downloader|
|aakash||14:02||mp3 downloader with progress bar|
|taxilian||14:02||hmm. You could probably do it either way; the main advantage to using the browser APIs (BrowserStreams) is that proxies should be taken care of for you
libcurl is harder to use but much more powerful
also will make your binary larger
|aakash||14:02||Trying to get BrowserStreams to work since an hour, onStreamCreated fires but onStreamDataArrived never fires|
|taxilian||14:02||I'd start with SimpleStreamsHelper and modify that to meet your needs|
|aakash||14:02||I am extending DefaultBrowserStreamHandler|
|taxilian||14:02||it's a working example|
|aakash||14:02||ok, will look into. What I could be missing though?|
|taxilian||14:02||I honestly don't know
I don't use BrowserStreams much; I wrote SimpleStreamsHelper and it's always worked fine =]
|taxilian||14:02||dougma: interestingly enough, there seems to be a fairly major memory leak in FBTestPlugin; I'm not yet sure if it's related to your variant replacement, going to test wtihout it|
|shenberg||19:02||Hey. I'm getting weird behavior with IE on my plugin|
|taxilian||19:02||shenberg: what kind of weird behavior?|
|shenberg||19:02||I think understand the problem now|
|shenberg||19:02||I'm doing something like browserPtr->getDOMDocument()->callMethod<FB::JSObjectPtr>("createElement", FB::variant_list_of("div"))
This works just fine in chrome/firefox
but fails on IE9
specifically, invoking createElement fails
|taxilian||19:02||hmm. yeah, things like that sometimes don't work well in IE and you have to expand the activex side of the DOM abstraction to add support|
|shenberg||19:02||My guess would have been that since I'm calling it from my plugin JSAPI c'tor then it's failing
but if it isn't supposed to work
that's good to know
|taxilian||19:02||it wouldn't surprise me if it doesn't work
you could add a createelement to the AXDOM::Document object and the DOM::Document abstraction
|shenberg||19:02||Hmm, seems easy enough (this is instead of invoking it using the JSObject and hoping for the best, I assume)
then, I use the element I get back
and do some more invocations on it
am I to expect I'll need to abstract out those too? (specifically, I'm setting the div's style.cssText property
|taxilian||19:02||using the activex stuff to call CreateElement it'll return a IHTMLElement|
Yeah, I read the docs for IHTMLDocument2
|taxilian||19:02||and that can be converted to an IDispatchEx with QueryInterface and then wrapped back into a IDispatchAPI
and then you keep using it like normal; however, depending on what you're doing you may want to use the AXDOM::Element type
and add to the abstraction there
anyway, I gotta run. I accept patches, remember ;-)
|shenberg||19:02||If I get working code, I'll send a pull request on github|
|shenberg||20:02||taxilian: the "Element inherits Node" abstraction is breaking down in IE since calling createElement returns an IHTMLElement interface to an object which does not have an IHTMLDOMNode interface until it's added to the document|
|shenberg||21:02||d'oh, my previous statement just plain wrong. COM is a world of hurt that I'd sort of forgotten
taxilian: Never mind my previous statement - I got the code to work. I might be leaking some COM references, but it all sort of works. I'll clean things up tomorrow and send a pull request
|taxilian||22:02||shenberg: excellent. Try to use CComPtr objects to avoid leaking objects
oh, he's gone
FireBreathBot: tell shenberg uuse CComPtr and CComQIPtr to avoid leaking