I have a video element:
var video = window.content.document.createElement(\"video\");
video.width = width;
video.height = height;
video.style.backgroundColor =
Actually this seems to be a bug in FF: https://bugzilla.mozilla.org/show_bug.cgi?id=926753
You should add a loadeddata
event listener to the video, and try to read the size then, which is when enough information about the stream has been decoded and the dimensions can be accurately determined.
Still, it sometimes takes a bit in some cases to get the dimensions ready, so you might want to try several times (with delay) until it works.
That might be why it is not working for you but it's working for Sam.
Here's how I check the video size, with several attempts if required, in my gumhelper library: https://github.com/sole/gumhelper/blob/master/gumhelper.js#L38
Notice how we try several times and if it doesn't work we "give up" and default to 640x480.
"loadeddata" should only fire once. It's better to use "timeupdate" to repeatedly check if the video width and height have been set, in particular with getUserMedia where you don't really pause the video, but go straight into playback.
There are two issues here:
video.videoWidth
and video.videoHeight
properties weren't getting set as soon as the loadedmetadata
event fired. This was a bug in FireFox, which is now fixed (thanks to @Martin Ekblom for pointing out the bug).I don't think there's a direct way to get the dimensions of the active area, but after struggling with this myself, I wrote up a solution to calculate it from the values we do know:
function videoDimensions(video) {
// Ratio of the video's intrisic dimensions
var videoRatio = video.videoWidth / video.videoHeight;
// The width and height of the video element
var width = video.offsetWidth, height = video.offsetHeight;
// The ratio of the element's width to its height
var elementRatio = width/height;
// If the video element is short and wide
if(elementRatio > videoRatio) width = height * videoRatio;
// It must be tall and thin, or exactly equal to the original ratio
else height = width / videoRatio;
return {
width: width,
height: height
};
}
Essentially, we take the aspect ratio of the video element, the aspect ratio of the video, and the dimensions of video element, and use those to determine the area the video is occupying.
This assumes the video's fitting method hasn't been modified via CSS (not sure if that's even possible at this time, but the spec allows for it). For more details on that and a few other things, see the blog post I wrote ("Finding the true dimensions of an HTML5 video’s active area"), inspired by this question and its lack of complete answers.
It's interesting to note that while the spec specifically mentions the possible edges around a video (letterboxing and pillarboxing), I wasn't able to find any other mentions of it, apart from your question.