As part of Period-flattening, I'm trying to remove our dependence on
the "position" field of SegmentReference. With that eliminated, we
can more easily concatenate Arrays of SegmentReferences without
modifying them.
- Make SegmentIndex iterable
- Add specialized seek() and current() methods to SegmentIterator
- Remove position from SegmentReference
- Make positions in SegmentIndex API stable without field in
reference
- Remove brittle hard-coded positions in tests (except SegmentIndex
tests, where they would be hard to avoid in testing methods
separately)
- Use SegmentIterator in StreamingEngine to track the next segment
between switches
Issue #892 (refactor StreamingEngine)
Issue #1339 (period flattening)
Change-Id: I666cc21249c34ee6cbc138a59109d9f1159fa127
This reverts commit 235e4e11ad.
The effort to remove SegmentReference's position field will be handled
in a different way.
Issue #892 (refactor StreamingEngine)
Issue #1339 (period flattening)
Change-Id: I62b115137abc89f498b30467e574b0401dcad05d
This removes the tests added by #2412 since they didn't test what
was wanted. This also fixes our existing tests to ensure they work
with autoCorrectDrift. Lastly this adds a new test to ensure the
UTCTiming is ignored when autoCorrectDrift is true.
Change-Id: If511fca5d3f360cdf373229961b2a01fdbda31bb
As part of Period-flattening, I'm trying to reduce our dependence on
the "position" field of SegmentReference. If it can be eliminated, we
can more easily concatenate Arrays of SegmentReferences without
modifying them.
SegmentIndex can now track the last reference you asked for and
iterate through the list of references. This means we don't need the
"position" field of SegmentReference, which means we don't need to
know positions in advance or globally. StreamingEngine will no longer
use position to request segments.
The old methods find(time):position and get(position):SegmentReference
have been replaced with seek(time), current(), and next(), all of
which return a SegmentReference and maintain an internal pointer to
the "current" reference. Care has been taken to maintain that pointer
during the evict() and fit() operations. Recent changes to merge()
made sure that the pointer does not need to change during that
operation.
All test updates are related to the SegmentIndex API change, not
changing expectations or behavior.
Issue #892 (refactor StreamingEngine)
Issue #1339 (period flattening)
Change-Id: I1682dcc2dd625c6e390711538e46d31e6eb6cea8
The recent DASH refactor broke the details of multi-period segment
references for SegmentTemplate+duration.
This change introduces a corrected universal model for generating
segment references for fixed-duration SegmentTemplate content. It has
been tested with a variety of streams from third-parties, most notably
the DASH-IF's live simulator.
Issue #1339 (flatten periods)
Issue #892 (refactor StreamingEngine)
Closes#2111
Bug: 28087927
Change-Id: Ife4e9ae2bb726e24e97dfe153882dabba4c03f2d
In an in-progress DASH stream (in which the content is static, but not
fully available yet), we will need to produce additional segment
references as time progresses. This was broken recently during
refactors for period flattening, when SegmentTemplate with a fixed
duration was changed to use a SegmentIndex instead of generating
references on-demand.
Issue #1339 (flatten periods)
Issue #892 (refactor StreamingEngine)
Change-Id: I0239335fa7c569063f06f4416982fc5ca7d5873d
This changes the eslint rule to enforce a strict pattern for the
argument comments. The comment must appear before the argument and
must be /* foo= */. This still ignores line comments.
Change-Id: I3afb01c65e1088eda13facb3aeeaa7595a2f5aee
This is a step along the way to flattening periods, and also
simplifies a few small things in StreamingEngine and tests.
Issue #1339 (flatten periods)
Issue #892 (refactor StreamingEngine)
Change-Id: Ie17cd5e15ed6ec9290a918c3a69c05c74581e0fc
This reflects changes in Google's policy on JavaScript license
headers, which should be smaller to avoid increasing the size of the
binary unnecessarily.
This also updates the company name from "Google, Inc" to "Google LLC".
Change-Id: I3f8b9ed3700b6351f43173d50c94d35c333e82b4
Include event ID in findSimilarRegion_() check so that events with same
presentationTime and duration but different ID are treated as separate events.
Closes#2077
During a refactor ("Replace find/get callbacks with SegmentIndex"),
this test changed from using a hard-coded segment position to looking
one up based on the live edge. But using the exact live edge leads to
some amount of flakiness.
Back off the live edge enough to make sure you're not on the trailing
edge of the segment.
Fixes b/137286528
Change-Id: Ic4e869e8a3b1ceaf45dc14b04ea0e987bd690453
This replaces find/get callbacks in Stream with a SegmentIndex. With
the exception of DASH's SegmentTemplate+duration, all manifests were
already backed by SegmentIndex. Now, all manifests are backed by
SegmentIndex. This will simplify Period-flattening, in which all
tracks will be represented by a list of segments, some of which come
from different Periods.
The SegmentIndex in Stream will not be created until
createSegmentIndex is called. Prior to this change, the find/get
callbacks could be invoked without createSegmentIndex() in some cases
(excluding DASH's SegmentBase), which some lazy tests took advantage
of. Now that find/get are methods on SegmentIndex, createSegmentIndex
must be called in all cases. The tests have been updated accordingly.
Making SegmentIndex generation async in all cases exposed some issues
with the parser context being modified in-place between one
Representation and the next. So the parser now makes a shallow copy
of the context before it is bound into an async callback.
To facilitate updating the SegmentIndex for SegmentTemplate+duration
content, SegmentIndex now has a method to update its list on a timer.
Once per segment duration, the index will be updated to add and remove
SegmentReferences.
The initial expansion of SegmentTemplate+duration will be limited to a
relatively small number of segments, to avoid excessive CPU or memory
consumption. This defaults to 1000 segments, but is configurable.
Issue #1339
Change-Id: I99c007b1096c3b396d04a729750cd7b743cb899d
This ensures that we get the expected types and that type coercion
doesn't convert between types. This also ensures we are consistent
in how we check for equality of primitives in tests.
Change-Id: I9f3aacdf25ab1afe5e8d6e4b895b5299ee687d54
This converts them to be normal async tests instead of mocking the
clock and Promises. This uses some internals on the DASH parser for
updates to allow the tests to complete quickly.
Issue #1379
Issue #1953
Change-Id: I6d21d124a290920e568b8a275221140e1b27941b
- Reduce times for "short delays".
- Remove backoff delay from networking tests.
- Avoid hard-coding delays in the library.
- Move Storage tests to integration tests since they use indexedDB (and
take over 200ms each to run).
This reduces the time it takes to run unit tests (with --quick) from
50 seconds to about 6 seconds. Now all but one unit test finish <100ms.
Change-Id: I88461472a87c4cf750a36d07d07422818e069a4d
This is a fully automated change. The linter will fail because the
extra indentation caused line-length errors. These won't be fixed
automatically. They are fixed in a follow-up to make this one fully
automated.
Change-Id: I4d8cf9c998985add2bcd24a81c8d65495668c4f3
With the new style rule, we cannot have two statements on the same line.
So we can no longer have an "if" on a single line and we cannot have
an arrow function with a body on the same line as when it is used.
This is mostly a manual change.
Change-Id: I2285202dd5ecbad764308bc725e6d317ff2ee7f0
This is an automated change to convert use of "function" functions
to arrow functions. This doesn't change all uses of bind() that
could be converted. This also doesn't remove all "function" functions.
Change-Id: I40ac7d086bcef947a1be083359c8fd1d4499a9c3
A coming update to the Google eslint config will require using "const"
over "let". This makes that one change to isolate the big changes.
Change-Id: I7d0974c3ae15c53cc45a6b07bf9f6586e2d34aca
The configured clock sync URI was only being used for true live
streams, but there are cases where it is needed for in-progress
recordings, as well. This came up while working with a partner on
issues with CEA caption support.
Related to b/70902665
Change-Id: Ie7199dec3177a7ed014d9d0d9aaeb9f1bb1dbeac
Instead of duplicating the default configuration values in each test,
this now uses the default configuration methods.
Change-Id: Ifd2ab349db7903a2acb0d06fed4bd0ccd5050b35
Add the config field manifest.dash.ignoreMinBufferTime, which will default to false. If true, the DASH parser will ignore minBufferTime in the manifest, such that streaming.rebufferingGoal is the only factor in play.
Closes#1547
Before, segment timestamps were used in PresentationTimeline without
regard for the period start. This means they were not truly relative
to the presentation, but to the period.
The "isFirstPeriod" argument was also broken. It was meant to be true
for segments from the first period *ever*, but was passed true for the
first period *in the latest manifest update*. So data calculated from
that was bogus for live streams.
Now, notifySegments() is supplied with a period start time, so that
segment references can be combined with the period start to give
presentation timestamps. This fixes a major issue with the original
fix for #999.
Closes#999
Change-Id: Id0fe450f3ce4f90a2387d7103c75eb88f0c69c72
Went through all the functions in FakeNetworkingEngine used to
register responses and made them all follow the same language.
Change-Id: I0f197254fa40e1d97985a3070b3cda3bbff70df2
This calculates the DASH live edge from explicit segment descriptions
when we have them. The net effect is that availabilityStartTime will
only be used with SegmentTemplate+duration, in which we have no
explicit segment times.
With this, a DASH live stream experiencing encoder drift can still be
played so long as we know the segment times. This makes playback of
DASH live streams work more like HLS live streams.
Please note that DASH live streams using SegmentTemplate+duration may
still suffer from encoder drift.
This change also:
- Avoids DASH clock sync when availabilityStartTime is not used,
which should reduce startup latency
- Removes the simulation of a presentation start time for HLS live
- Renames some variables and improves comments for clarity
- Fixes some brittle tests that made bad assumptions or instrumented
the wrong methods
- Adds new tests that show the new behavior in PresentationTimeline
Closes#999
Change-Id: I21d7f3ccc81c9d9e218857a9b41882a7609ca36a
This reverts the implementation to an earlier draft state. During
review of the first version, I recommended moving the implementation
to Player so that it would be independent of the manifest parsers.
That was bad advice on my part, because this overlooked updates made
by the parsers when live manifests are updated later.
Closes#1177Closes#1307
Change-Id: I4611e00824dead83c6467da04a2a11afa892ace7
This config lets you override the availability window of a live stream.
It is in config.manifest, and passed along on start.
This will let users configure the parser so that they can seek with HLS
live streams, for example.
Closes#1177Closes#1307
Change-Id: Icd3c1d81c6b52ebdbb72137df42fc91cd73a0207
If a single manifest update takes a long time, it should not hinder
future ones; but if there are repeated slow updates, we should delay
updates. This uses an EWMA to keep a rolling average of update times.
Closes#1399
Issue #1408
Change-Id: I6a662ce51ba9c3e1e4a87b2418e35fc3948043c1
For slow embedded devices, such as the v1 Chromecast, updating the
DASH manifest takes far too long. The result is that we can sometimes
repeatedly fall out of the seek range when playing near the left edge
of a live stream.
Until we can dig into and optimize the DASH parser, we will detect
and adapt to slow parsing by updating the manifest less often.
This is done based on how long updating actually takes, and does not
hard-code either the platforms that need this workaround, nor the
frequency of updates they can handle.
Bug: 75276747
Change-Id: I6565192b4ce6935ee8416ea4843fee851f4bd990
This is an automated fix to be inline with Google style guides, this
was created with eslint's --fix option.
Change-Id: I860eecbc8152603e730aa17a1393f16d26b3b6fc
This is part of a change to convert all usages of 'var' with either
'let' or 'const'. This takes a conservative approach for 'const' where
it will only be used for aliases and storing the "original" values in
tests.
Change-Id: Iebba756b5d0e68c41292ecabda89503682d8d434
A bug in our Promise polyfill caused issues with the recently-added
AbortableOperation class on IE11. Since external polyfills for this
are smaller, it is easier to remove ours in favor of a third-party
polyfill. Applications that wish to support IE11 must now load this
additional polyfill.
We are using the "es6-promise-polyfill" module from npm, but any
compliant polyfill should suffice.
One feature our own polyfill offered was the ability to flush all
Promises, which allowed us to write synchronous unit tests that
simulated async processes. To get this ability back, we are now
using the "promise-mock" module in our tests.
Getting "promise-mock" to load correctly involved switching from
"requirejs" to "cajon", which builds on requirejs and supports
AMD modules more directly.
Closes#1260
Change-Id: I5de48e88a910736ae5c1897a7a509bc5641acb70
This uses AbortableOperation in all networking, from the scheme
plugins all the way to the request interface.
This also updates all default scheme plugins, docs, and sample code.
Backward compatibility is provided for scheme plugins and the
request API in NetworkingEngine. This compatibility will be
removed in v2.5.
Two cancelation-related tests have been disabled in
player_integration until the new abort interface has been adopted
in the manifest parsers.
Issue #829
Change-Id: I91c8e6efe97798d111e8ddca5655cddc1f6bcbf3
It is useful to have the default presentation delay configurable when the stream provider isn't able to add `suggestedPresentationDelay` in the manifest
Fixes#1234
When calculating segment references, the end time should be capped at
the period's end, except when the period has no end.
Closes#1204
Change-Id: Ia608452706bbc3fd847a645197e8a0cb57d7aaec
If the parser is failing to load a manifest, perhaps due to being offline,
and the player is destroyed mid-load, it will wait for all of the retries to
run out before canceling. This causes a significant wait time on destroy.
This change adds an optional isCanceled callback to the request method in
NetworkingEngine, allowing requests to be made that can be canceled via a
provided callback. This is then used to quickly skip past all of the
retries, when destroying the parser.
This doesn't stop any backoff in progress, so there still might be a delay of
a second or two (depending on networking settings, of course) if you destroy
during a backoff period. That will be changed in a followup CL.
Issue #1084
Change-Id: I0b99616ed4d5be078161122532ea271e84613170
We misinterpretted the EXT-X-START tag before. Now, we ignore this
tag and parse segment times from the segments themselves. This is
more robust for both VOD and live HLS content.
We avoid segment parsing when updating normal, well-behaved live
streams. For poorly-behaved live streams, we will fall back to
segment parsing and recover.
This also addresses the issue of VOD content which does not start at
zero. Instead of using configuration to make this playable, we will
now offset VOD content back to 0 automatically.
Issue #740Closes#1011
Change-Id: Ib3a59f87e3a050244cd39854409d8e3542c50b0a
HLS Live comes in two flavors: LIVE and EVENT.
EVENTs append segments to the end of the playlist, but do not
ever remove segments. LIVE can both append segments to the end
and remove segments from the beginning of the playlist.
This CL lays the ground for supporting both types and introduces
support for EVENTs.
Issue #740.
Change-Id: I2042660daf9d1e7209cb1b40f50ab8b24571faa4
This fixes a bug in which we did not track DASH period IDs during the
initial manifest parse, which led to duplication of all periods during
the first update.
Closes#963
Change-Id: Ife9be7e5f5e158a565b0095d7cbb81b50c681422
Joey Parrish
Jacob Trimble
Tomohiro Matsuzawa (Tomo)
Satheesh Velmurugan
Álvaro Velad Galván
Michelle Zhuo
fadomire
Aaron Vaage
Theodore Abshire
Patrick Cruikshank
Sandra Lokshina
Benjamin Wallberg