brainfood / videojs-contrib-hls

Gary Katsevman committed 2013-12-12 16:31:30 -0500

Gary Katsevman committed 2013-12-12 15:45:43 -0500

Gary Katsevman committed 2013-12-12 15:39:19 -0500

Gary Katsevman committed 2013-12-12 15:36:30 -0500

Gary Katsevman committed 2013-12-12 15:32:30 -0500

Rename mediaFile to mediaURL in segment and rendition
Rename byteRange into byterange and default it to -1 in segment
Rename duration to targetDuration in segment

Gary Katsevman committed 2013-12-12 13:59:04 -0500

Comments, EXTINF tags, and EXT-STREAM-INF tags get turned into an array
in the outputted format.

Gary Katsevman committed 2013-12-12 13:55:49 -0500

Gary Katsevman committed 2013-12-11 19:45:25 -0500

Gary Katsevman committed 2013-12-11 19:23:28 -0500

Gary Katsevman committed 2013-12-11 19:20:42 -0500

Gary Katsevman committed 2013-12-11 18:44:12 -0500

Now, it'll use peg's line and column tracking. So, each extinf would
output an object with property #EXTINF + line number.

Gary Katsevman committed 2013-12-11 18:25:28 -0500

Gary Katsevman committed 2013-12-11 13:46:55 -0500

Gary Katsevman committed 2013-12-11 13:40:50 -0500

Gary Katsevman committed 2013-12-11 13:40:03 -0500

Generate the parser file via 'npm run peg'.
There is a simple test script in test/pegtest.js.
The test can be run via 'npm run testpeg' which will generate a new copy
of the parser and then run the test file.

Gary Katsevman committed 2013-12-10 15:48:37 -0500

Lee Whitaker committed 2013-12-01 21:46:16 -0500

Lee Whitaker committed 2013-11-19 10:48:28 -0500

Lee Whitaker committed 2013-11-19 10:38:36 -0500

Lee Whitaker committed 2013-11-18 18:38:43 -0500

Lee Whitaker committed 2013-11-18 18:31:39 -0500

…quence, playlist_type, and target_duration.

Lee Whitaker committed 2013-11-18 18:12:33 -0500

Lee Whitaker committed 2013-11-18 10:11:13 -0500

Not sure how this got in here but it doesn't belong.

David LaPalomento committed 2013-10-31 18:10:51 -0400

Joining videojs-contrib so change the name to reflect that.

David LaPalomento committed 2013-10-31 18:09:36 -0400

vjs-media-sources is now a "contrib" project, so make sure the dependency name reflects that.

David LaPalomento committed 2013-10-31 17:57:42 -0400

Update videojs-media-source to videojs org repo version

David LaPalomento committed 2013-10-31 15:10:49 +0000

sheffernan committed 2013-10-30 15:50:35 -0700

Fix the tests. Only manifests with the X-STREAM_INF tag should be consid...

David LaPalomento committed 2013-10-30 21:43:50 +0000

Lee Whitaker committed 2013-10-30 17:35:58 -0400

Grow FLV tags as necessary and make some performance tweaks

David LaPalomento committed 2013-10-30 20:11:14 +0000

The example page should always work without additional services running. Use a checked-in HLS stream to guarantee this.

David LaPalomento committed 2013-10-26 18:47:12 -0400

The difference isn't particularly significant but we can get away with a couple fewer shifts when parsing TS packets. Also, avoid an additional branch by default the dts value to the calculated pts.

David LaPalomento committed 2013-10-26 18:45:24 -0400

Drop the extra null byte when closing an "ecmascript array" in the metadata tag. The spec specifies a 3 byte tag and I'm not sure how the 4-byte version managed to work in practice. This makes the output differ significantly from the original actionscript version of the code but the converted video still plays back without any issues.

David LaPalomento committed 2013-10-26 15:36:04 -0400

Forgot that time was being measured in ms so the throughput values were off by 1000. Changed reporting unit to MB/s and fixed the precision to 3 digits so it's not overwhelming.

David LaPalomento committed 2013-10-26 15:35:03 -0400

Replace more one-letter variables with more descriptive names.

David LaPalomento committed 2013-10-26 15:28:36 -0400

Exclude the performance page from the qunit grunt task.

David LaPalomento committed 2013-10-25 19:36:56 -0700

Calculate parser throughput and add a column to the result table to show it for each run.

David LaPalomento committed 2013-10-25 19:15:37 -0700

Check whether the byte buffer is large enough before performing a write and allocate a larger buffer if necessary. Do less byte-by-byte ops when writing metadata double tags since they were a significant portion of parse computation when testing on my laptop.

David LaPalomento committed 2013-10-25 19:13:19 -0700

Created a page that runs the segment parser repeatedly on the bunnies test segment. Initial testing indicates garbage parseTSPacket and garbage collection take up about a quarter of the parse run.

David LaPalomento committed 2013-10-25 14:18:35 -0700