Include event ID in findSimilarRegion_() check so that events with same
presentationTime and duration but different ID are treated as separate events.
Closes#2077
Safari 13 is out, and needs polyfills just like Safari 12 did
(although one fewer). This assumes that Safari 13+ all need the same
polyfills (to avoid surprise failures when Safari 14 comes out).
This also raises the minimum version from Safari 9 to Safari 11, since
usage stats for Safari 9 and 10 are vanishingly small. Dropping
support for Safari 9 and 10 allows us to delete a complicated polyfill
used for those versions.
Change-Id: I8f833d915215fc19a89e3aea292b67c5d677c779
According to RFC 3986, URL schemes should be case-insensitive. Our
scheme plugin map is registered with exclusively-lowercase scheme
names, so we should convert the input scheme to lowercase before
looking up the plugin.
Closes#2173
Change-Id: I78218c18d3df154aabd101a9ef5fd3156d0c57a8
We avoided bind() since we could use lambdas instead; but with the
console methods, we loose the link to where the log happened in the
DevTools. So by using bind(), we can see where the log came from.
Change-Id: I04ad1f360e459c3676eabb6d0651a7963d186b43
When we are in the "same" stall, we shouldn't handle the stall multiple
times. If we do, the video will step forward slowly over time, which
isn't desirable; especially on slower platforms where a seek would
restart the time it takes to start playing again.
Change-Id: I11a1811f9ecf754484d3530f4dc047cf95b007a4
For VOD, the first segment of every Media Playlist in every Variant
Stream must start at the same media timestamp. Thus, we can get the
start time once and store the value, and all the streams can use the
value directly. For live, we assume for the same.
This change reduces calling getStartTime_() from 15 times to once for
Angel One, and also fixes the text segment timestamp issue.
Also removed associated methods that no longer needed.
Closes#1558Closes#1563
Change-Id: I0e95ab93ea2b13758128f11019b262bc53dbcd38
Changes in the Closure Compiler have made it so that goog.define
without using and assigning the return value will no longer be
allowed. There is no need for goog.define, though. Instead, we can
just use the @define annotation and assign directly to a variable.
This eliminates the use of goog.define.
A similar change was already made almost two years ago for the Player
version in lib/player.js
(in Change-Id: Iad9040d0f5ee397c7d46b1d64bc5bebf2def57bc)
Replaces PR #2158
Change-Id: Ia7c7de0407f9790284c418fb6edde948b922ab62
Previously, the Closure compiler was causing a strange problem in
non-debug compiled builds in the HLS parser: when we make an alias for
shaka.net.NetworkingEngine.RequestType, if we access RequestType.SEGMENT
the value is undefined. However, if we store
shaka.net.NetworkingEngine.RequestType.SEGMENT in a variable, it works.
Looking at the compiled code, when shaka.net.NetworkingEngine.RequestType
is defined, SEGMENT seems to be treated as some special case, which might
explain this odd behavior.
Fixes#2156
Change-Id: I272dbc8b8db08ef8f9067c535c25890adae32440
We would incorrectly initialize the embedded captions multiple times
during a Period transition, which caused duplicate cues to be given to
the displayer. We also wouldn't handle the case when switching between
embedded captions and external text during a Period transition. This
fixes both cases and adds tests for them.
This also avoids passing an empty cue list to the displayer.
Fixes#2076
Change-Id: I89add3eb86ad8d93644bba14eabd11f98d57bc5e
Because src= content lacks codec information, ClearKey setup fails.
However, because there was no DrmInfo prior to the ClearKey settings
being applied, the variable hadDrmInfo was false, which led to a
suppression of the resulting DRM error.
Instead, we should count ClearKey settings toward hadDrmInfo. So now,
ClearKey settings are applied before we calculate hadDrmInfo.
Closes#2139
Change-Id: I14d5bfe63fd2ce1a461ae7ef8c2fee7f42bb1b3e
We are seeing spurious errors from the load graph tests. These errors
don't appear to be thrown from any of our code and I can't find what is
causing the uncaught rejection errors. But adding a delay seems to
remove the errors.
Fixes: 140725109
Change-Id: Ia08e7a4c12a17cf5610db64b146b863c8e5ecbeb
When using DOMParser parseFromString with an empty string, it returns an errored XML document. The next part of the code is then executed and throws an unnecessary error.
This adds a special case for an empty string.
In compiled mode, UI calls player methods by their compiled, renamed names.
The proxy object doesn't know those, so an error used to be thrown while
casting in compiled mode. This change adds a map from compiled to extern
names that cast proxy can use.
It also cleans up proxy code a bit and points the demo to the nightly
build of cast receiver.
Closes#2130.
Change-Id: I39b83c6a1e050a68d70e7923d1a07ad8929daeeb
Before we would only update the current Period when we got a key status
change; this caused us to fail playback for the second Period if the
streams were restricted. Now we update all the Periods.
Closes#2135
Change-Id: I2238a90c1d367a1d898ec26e4318852cd26da4b7
This standardizes how we handle the spacing in argument comments. The
other way is much more common, so we standardized on that. This add a
new linter rule to catch and automatically fix those issues.
Change-Id: I8ea54c47ae4d34cf8e1646e56c6ed8142b42afbe
For audio only content that the audio variant tag and media tag has the
same uri, we should compare all media tag uris with the variant tag uri,
instead of comparing only the first one.
Change-Id: Ifa985eae441076418fc289e5601020c06c135c24
This adds support for a DASH MIME type that predates standardization
on "application/dash+xml". Some servers still use the older, now
non-standard MIME type of "video/vnd.mpeg.dash.mpd". This is trivial
to support, and it helps some folks who needs it, so there's no good
reason not to support it.
Change-Id: I17e85fdb74491e9e32628eddf30ef9e88acf5221
These are intermittently used and cannot be verified. Most of our
methods can throw a shaka.util.Error, so having it doesn't really add
anything. Plus, if we change a function to throw, we'd need to update
all their callers to ensure they have an accurate description of what
they throw; otherwise we can't trust the directive.
Change-Id: I520bd0fc4c33443e967bf5b103ca5aa9e3274884
Apparently, the 100ms fudge factor we had was not sufficient for
some assets, which could cause those assets to get stuck buffering at
the very end of the video, when on native controls.
This just increases the fudge factor to a whole 1000ms (1 second).
Fixes#2117
Change-Id: Idf3c09f397ebb6abf7f681efaa042608eed60152
Since embedded text tracks are not "real" tracks, we need to ignore them
when handling Period transitions. This ensures that the other streams
will handle the transition instead of waiting for the text stream to get
updated.
Fixes#2075Fixes#2094
Change-Id: I2c7c92cb04795edb3f7695677ee6745bad3f4471
A new conformance rule was added after the PR was started and it is
violating it. This fixes the violation and adds another assertion.
Change-Id: Ie18760d9e1db1959d9587682ff6c9c8ca4fe6a8f
We can't parse webm segments to extract segment start
time yet. Getting webm streams in hls stops playback
even if there are other formats we can play. This change
skips webm segments for now.
Fixes#2108
Change-Id: I4ff167efea45967cdf95a91afbe75efef3d9aa32
Using "new Uint8Array" with a TypedArray creates a copy of the buffer;
this is unnecessarily expensive for large buffers. This adds a rule
to disallow using it in favor of a new utility that correctly creates
a new "view" on the same buffer.
Note it is fine to pass an ArrayBuffer to "new Uint8Array" and it won't
copy; but there there are many cases where the type is BufferSource,
so it could be a TypedArray. Unfortunately, there are many other cases
where we explicitly pass an ArrayBuffer; but the compiler rules don't
allow us to whitelist this case (since ArrayBuffer is part of
BufferSource).
Change-Id: I58696a85a9cbcc188c0b16919c9eeb63e56edca1
Now the EbmlParser and DataViewReader types can accept BufferSource
objects to avoid the caller having to convert it at each callsite.
Change-Id: I527fbc638485214bfe976d7da89b79b6098033a8
This changes the network API to use BufferSource instead of ArrayBuffer,
which allows plugins to return a "view" on a buffer instead of the
whole buffer. This also adds some utilities for changing between
views and buffers.
Lastly this forbids the use of the "buffer" property of TypedArrays
since it doesn't work with partial "views". This audits and fixes the
usages of the "buffer" property to ensure correct usage.
It should be noted that both MSE and EME accept a BufferSource as input,
so we don't need to convert a "view" into an ArrayBuffer before passing
to it.
Change-Id: Iaa417773f8ce5304424e43c7372ce10ebf540d2a
Chrome has fixed their behavior involving endOfStream(), but both Tizen
and IE11 have the same broken behavior. Unfortunately these platforms
will likely never be fixed, so we need to keep this forever.
Fixes#1357
Change-Id: Id540170f8c8cb6e0ddf3a30bde5035016fa61df1
During startup, there was a race between the video element's ready
state and seeking to the requested start position. If the video
became ready before the seek to the start position, there could
briefly be 0 tracks. Reproducing this required a first period start
> 0 and a start position > 0, such that a lookup of the "current"
period during this brief window would return null.
The problem was more likely to be seen on Chromecast, where the time
between load() and the completion of a seek was much larger than on
desktops.
The bug was caused by Playhead, which both seeks to the start position
and manages the concept of "current time", returning the start
position during startup and the video element's position thereafter.
This fixes the race, making sure that startup is not considered
complete by Playhead until the initial seek is over. This adds an
appropriate regression test and simulated content that can reproduce
the conditions in the original bug report.
Fixes b/138941217
Fixes#2045
Change-Id: Ie8a57283468d88697619349a29e37a804586e5ef
According to the style guide, protected members should not end with
an underscore. This renames them to fix it. This also marks some
members as private that aren't used outside the class. Lastly, this
fixes a possible bug with the use of "this".
Issue #1553
Change-Id: I6c7ecdcd228ec4e4aded4af52e2262d0bf5beaf4
This adds a separate configuration value for this because (a) the HLS
parser can't access the streaming configuration and (b) this behavior
is different in that the streams are removed instead of just not playing
them.
Closes#2065
Change-Id: Ic9b1e0d40b161cd4ba0eb6d7c922882112034626
- Use <code> blocks.
- Use <p> to separate blocks (since newlines in comments are ignored).
- Use <ul> for lists.
- Give messages for deprecated tags.
Change-Id: I06d5e0e75eb3e498063cf0c291e6031af5d93b2d
Our previous approach:
For every variant tag, find the associated media tags(with the same
group id), create stream info for each media tag, and then create stream
info for variant tag based on that, and create variants.
This approach has duplicate calls for the media tags with the same group
id to create stream infos.
New approach:
1. Get codecs from variant tags
2. Create stream info from each media tag (with the codecs from variant
tags)
3. Create stream info for each variant tag, and then create variants.
This creates stream info for each media tag once, and avoids duplicate
calls to guess the codecs.
Issue #1558Closes#2066
Change-Id: Icdf3133b1b15d1baf60624ff449e0a61fdb78d8a
Now there is a generic callback to transform the init data before
passing it to the browser. This can be used by apps to use a custom
content ID in FairPlay content. This also adds some utilities to help
in writing these functions and moves the default behavior to DrmEngine.
Closes#1951
Change-Id: I78ce660c126b53a69d5f55b16775ffcdbbe4d748
michellezhuo
Satheesh Velmurugan
Joey Parrish
Jacob Trimble
Álvaro Velad Galván
Theodore Abshire
fadomire
Sandra Lokshina