Initial avc3 support #72
Conversation
WalkthroughAdds an Avc3 sample-entry variant and wires it throughout the codebase. Introduces Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Tip 📝 Customizable high-level summaries are now available in beta!You can now customize how CodeRabbit generates the high-level summary in your pull requests — including its content, structure, tone, and formatting.
Example instruction:
Note: This feature is currently in beta for Pro-tier users, and pricing will be announced later. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (4)
src/moov/trak/mdia/minf/stbl/stsd/mod.rs (1)
55-57:Codec::Avc3integration matches existing patternsDefining
Codec::Avc3(Avc3)and wiring it throughCodec::decode(Any::Avc3(atom) => atom.into()) andCodec::encode(Self::Avc3(atom) => atom.encode(buf)) is consistent with the other video codecs and leverages the existingderive(From)nicely. Only minor nit: the “SPS/PPS/VPS is inline” comment mentions VPS, which is HEVC terminology; if you want to be precise, you might drop “VPS” from the H.264 comment, but that’s purely cosmetic.Also applies to: 98-105, 122-129
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (3)
3-12: Avc3 Atom shape and decode logic look solidThe
Avc3struct layout (visual + avcc + optional btrt/colr/pasp/taic) andAtomimpl withKIND: "avc3"are consistent with the existing stsd/H.264 design.decode_bodycorrectly:
- decodes the leading
Visual,- iterates child atoms via
Any::decode_maybe(buf)?,- populates the optional fields, and
- enforces presence of
AvccwithError::MissingBox(Avcc::KIND).Logging unknown children with
tracing::warn!is a reasonable compromise for forward‑compatibility. If you ever want stricter behavior, a future enhancement could be to optionally preserve unknown children instead of dropping them, but the current behavior matches typical MP4 parsing patterns.As per coding guidelines.
Also applies to: 14-44
46-62: Encoding optional child boxes is correct; can simplify slightlyEncoding
visualandavccfirst, followed by the optionalbtrt,colr,pasp, andtaicboxes, mirrors the decode path and keeps the layout straightforward. The explicitis_some()checks aroundself.*.encode(buf)?are correct; ifEncodeis already implemented forOption<T>as “encode if Some, otherwise no-op”, you could drop theifguards and just callself.btrt.encode(buf)?;etc. directly to reduce repetition, but that’s purely stylistic.
65-157: Tests cover primary paths; consider an extra negative caseThe two tests provide good coverage:
test_avc3validates a minimal Avc3 with onlyvisualandavcc.test_avc3_with_extrascovers round‑tripping with all optional boxes populated.Together they exercise both decode and encode and rely on
PartialEqfor a strong equality check. As an optional improvement, you might add a test that decodes an Avc3 without anAvccchild and assertsError::MissingBox(Avcc::KIND)to lock in that invariant, but the current tests are already useful and aligned with the guidelines of keeping tests inline.As per coding guidelines.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
src/any.rs(1 hunks)src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs(1 hunks)src/moov/trak/mdia/minf/stbl/stsd/h264/mod.rs(2 hunks)src/moov/trak/mdia/minf/stbl/stsd/mod.rs(3 hunks)
🧰 Additional context used
📓 Path-based instructions (4)
**/*.rs
📄 CodeRabbit inference engine (CLAUDE.md)
**/*.rs: Code must be clippy-clean (cargo clippy --all-targets --all-features -- -D warnings)
Code must be formatted with rustfmt (cargo fmt) and pass formatting checks (cargo fmt -- --check)
Write unit tests inline within modules using #[cfg(test)]
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/mod.rssrc/any.rssrc/moov/trak/mdia/minf/stbl/stsd/mod.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
src/**/*.rs
📄 CodeRabbit inference engine (CLAUDE.md)
src/**/*.rs: Each MP4 atom type must implement the Atom trait with KIND, decode_body(), and encode_body()
Use Header::read_from() to handle large atoms without loading entire contents into memory
All operations should return Result using custom error types defined in error.rs
Use the custom Buf/BufMut traits for contiguous byte slices instead of ad-hoc buffer handling
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/mod.rssrc/any.rssrc/moov/trak/mdia/minf/stbl/stsd/mod.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
src/moov/**
📄 CodeRabbit inference engine (CLAUDE.md)
Place movie metadata atom implementations under src/moov/ (and its hierarchy: trak/mdia/minf/stbl/stsd)
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/mod.rssrc/moov/trak/mdia/minf/stbl/stsd/mod.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
src/any.rs
📄 CodeRabbit inference engine (CLAUDE.md)
Register new atom types in the Any enum (src/any.rs) to enable flexible decoding
Files:
src/any.rs
🧠 Learnings (4)
📓 Common learnings
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/**/*.rs : Each MP4 atom type must implement the Atom trait with KIND, decode_body(), and encode_body()
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/any.rs : Register new atom types in the Any enum (src/any.rs) to enable flexible decoding
Applied to files:
src/any.rssrc/moov/trak/mdia/minf/stbl/stsd/mod.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/**/*.rs : Each MP4 atom type must implement the Atom trait with KIND, decode_body(), and encode_body()
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/mod.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/moov/** : Place movie metadata atom implementations under src/moov/ (and its hierarchy: trak/mdia/minf/stbl/stsd)
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
🧬 Code graph analysis (1)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (1)
src/moov/trak/mdia/minf/stbl/stsd/mod.rs (1)
decode(99-119)
🪛 GitHub Actions: PR
src/moov/trak/mdia/minf/stbl/stsd/h264/mod.rs
[warning] 1-1: Rustfmt diff detected: 'mod avc1;' vs '-mod avcc;' indicates formatting changes needed in mod.rs.
[warning] 8-8: Rustfmt diff detected: public re-export lines differ; formatting changes needed in mod.rs.
[error] 16-16: Rustfmt check failed. Run 'cargo fmt' to fix formatting.
🔇 Additional comments (1)
src/any.rs (1)
271-277:Anyregistration ofAvc3is consistent with existing H.264 atomsAdding
Avc3alongsideAvc1in theany!macro’sbasiclist correctly plugs it intoAny,DecodeAtom,AnyAtom, and the conversion impls. This matches how other stsd entries are wired.Based on learnings.
I said they should start by copy-pasting AVC1 and sure enough, it decodes properly. But it's definitely not implementing the full spec and yeah the code reuse is frustrating. Maybe until there's interest in full Avc3 support, a type alias might just work? |
- Introduced a new AvcSampleEntry struct to encapsulate common fields for AVC sample entries. - Updated Avc1 and Avc3 to utilize the new AvcSampleEntry type, improving code reusability. - Added a from_u32 method to the FourCC struct for better FourCC initialization from u32 values.
Not an expert in Rust, so after a quick Cursor investigation, I got the following answer:
My latest commit looks to minimize code reuse and plan for future additions to other avc types by:
Happy to defer the decision to you guys! Or if we want to leave this open for now... |
bradh
left a comment
bradh
left a comment
There was a problem hiding this comment.
Minor suggestions only, I can do those as a follow-up if you'd prefer.
I'm going to guess you have some background in an object-oriented language. Trying to force everything to be an object is making it hard to see alternatives. (Maybe I'm just projecting here, because I sometimes still have this problem). I haven't actually tried it, but we can probably just turn the Visual parsing into a |
|
You guessed correctly 😂 I'll take a look at those suggestions during the weekend. Thanks for the suggestions! |
- Introduced a constant for BBC AVC3 sample data extracted from a test stream. - Implemented a function to create an expected Avc3 structure for comparison. - Added a test to validate the decoding of the BBC AVC3 sample, ensuring correct structure and child atom handling. - Refactored AVC1 code to use a more explicit byte array initialization.
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rs (1)
50-66: Fixencode_bodyto encode inner optional atoms instead of theOptionwrapper
self.btrt,self.colr,self.pasp, andself.taicareOption<...>fields. Calling.encodeon theOptionitself (even behind anis_some()check) is incorrect and will either fail to compile or rely on an unintended trait impl.Encode the inner atom only when present, e.g. using
if let Some(...).Suggested diff:
fn encode_body<B: BufMut>(&self, buf: &mut B) -> Result<()> { self.visual.encode(buf)?; self.avcc.encode(buf)?; - if self.btrt.is_some() { - self.btrt.encode(buf)?; - } - if self.colr.is_some() { - self.colr.encode(buf)?; - } - if self.pasp.is_some() { - self.pasp.encode(buf)?; - } - if self.taic.is_some() { - self.taic.encode(buf)? - } + if let Some(btrt) = &self.btrt { + btrt.encode(buf)?; + } + if let Some(colr) = &self.colr { + colr.encode(buf)?; + } + if let Some(pasp) = &self.pasp { + pasp.encode(buf)?; + } + if let Some(taic) = &self.taic { + taic.encode(buf)?; + } Ok(()) }
🧹 Nitpick comments (1)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (1)
12-39: Avc3 round‑trip tests are solid; shared fixture abstraction is optionalThe
base()constructor plusroundtrip()helper and the two tests (test_avc3,test_avc3_with_extras) give good coverage of both minimal and fully-populated Avc3 entries and mirror the Avc1 coverage.If more AVC-style entries appear later, consider extracting a shared test fixture helper (e.g., in a small
testshelper module) to avoid repeating the same synthetic Visual/Avcc setup in multiple files, but it's fine as-is for now.Also applies to: 83-117
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
⛔ Files ignored due to path filters (1)
tests/data/bbc_avc3.binis excluded by!**/*.bin
📒 Files selected for processing (2)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rs(3 hunks)src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.rs
📄 CodeRabbit inference engine (CLAUDE.md)
**/*.rs: Code must be clippy-clean (cargo clippy --all-targets --all-features -- -D warnings)
Code must be formatted with rustfmt (cargo fmt) and pass formatting checks (cargo fmt -- --check)
Write unit tests inline within modules using #[cfg(test)]
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
src/**/*.rs
📄 CodeRabbit inference engine (CLAUDE.md)
src/**/*.rs: Each MP4 atom type must implement the Atom trait with KIND, decode_body(), and encode_body()
Use Header::read_from() to handle large atoms without loading entire contents into memory
All operations should return Result using custom error types defined in error.rs
Use the custom Buf/BufMut traits for contiguous byte slices instead of ad-hoc buffer handling
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
src/moov/**
📄 CodeRabbit inference engine (CLAUDE.md)
Place movie metadata atom implementations under src/moov/ (and its hierarchy: trak/mdia/minf/stbl/stsd)
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
🧠 Learnings (4)
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/**/*.rs : Each MP4 atom type must implement the Atom trait with KIND, decode_body(), and encode_body()
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/moov/** : Place movie metadata atom implementations under src/moov/ (and its hierarchy: trak/mdia/minf/stbl/stsd)
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/any.rs : Register new atom types in the Any enum (src/any.rs) to enable flexible decoding
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rssrc/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/test/** : Put integration tests under src/test/ with sample MP4 files for different codecs
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
🧬 Code graph analysis (2)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rs (1)
src/types.rs (1)
from_u32(22-24)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (2)
src/types.rs (9)
new(18-20)new(193-195)new(304-306)decode(78-80)decode(101-103)decode(144-146)decode(207-212)decode(286-294)decode(325-329)src/moov/trak/mdia/minf/stbl/stsd/mod.rs (1)
decode(99-119)
🪛 GitHub Actions: PR
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
[error] 40-40: Rustfmt check failed. Formatting differences detected in the file. Run 'cargo fmt' to fix formatting issues.
[warning] 1-1: Diff in code formatting detected by cargo fmt --check (reported as a warning in logs) but caused the step to fail with exit code 1.
🪛 GitHub Check: build
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
[warning] 40-40:
Diff in /home/runner/work/mp4-atom/mp4-atom/src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
🔇 Additional comments (3)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (2)
1-6: Avc3 alias overAvcSampleEntrycleanly reuses the generic Atom implementationThe
AVC3_CODEconstant pluspub type Avc3 = AvcSampleEntry<{ AVC3_CODE }>;ties Avc3 into the genericAvcSampleEntry<const KIND_CODE: u32>and itsAtomimpl nicely, so Avc3 gets the correct KIND and behavior without duplicate logic.
41-143: Real BBC avc3 sample test is a very useful spec/interop checkUsing
include_bytes!for the BBC-derivedBBC_AVC3_SAMPLEand then:
- Parsing the header and asserting
header.kind == Avc3::KIND,- Decoding
Visualand walking child atoms viaAny::decode_maybe,- Finally doing a full
Avc3::decodeand comparing tobbc_expected(),gives strong confidence that the generic
AvcSampleEntry+ Atom wiring matches real-world avc3 boxes and that child atoms (avcC,pasp,colr) are handled correctly.src/moov/trak/mdia/minf/stbl/stsd/h264/avc1.rs (1)
3-3: GenericAvcSampleEntry<const KIND_CODE>design cleanly unifies Avc1/Avc3 Atom handlingDefining
AVC1_CODEonce and using:
pub struct AvcSampleEntry<const KIND_CODE: u32> { ... }pub type Avc1 = AvcSampleEntry<{ AVC1_CODE }>;impl<const KIND_CODE: u32> Atom for AvcSampleEntry<KIND_CODE> { const KIND: FourCC = FourCC::from_u32(KIND_CODE); ... }is a nice way to satisfy the Atom trait requirement (KIND/decode_body/encode_body) for all AVC-style sample entries while keeping Avc1 and Avc3 as simple aliases. This will also make future fourcc variants trivial to add.
Also applies to: 7-7, 16-16, 18-20, 21-48
|
I've fixed the formatting nit. The other parts are messier than I'd hoped, and won't get looked at until the weekend (UTC+11 time). |
There was a problem hiding this comment.
Actionable comments posted: 0
♻️ Duplicate comments (1)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (1)
40-40: Verify rustfmt compliance.A past review comment flagged a rustfmt issue around this line. Please ensure
cargo fmthas been run and the file passescargo fmt -- --checkto comply with project formatting requirements.As per coding guidelines, code must be formatted with rustfmt and pass formatting checks.
Run the following to verify:
#!/bin/bash # Verify rustfmt compliance for this file cargo fmt -- --check src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
🧹 Nitpick comments (1)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (1)
5-5: Consider adding documentation for the publicAvc3type.Since
Avc3is a public type alias that will be part of the crate's API, adding a doc comment would help users understand its purpose and relationship to the AVC3 codec specification.Example documentation:
+/// AVC3 sample entry. +/// +/// Similar to AVC1, but parameter sets are stored in-band rather than in the sample entry. +/// See ISO/IEC 14496-15 for the AVC file format specification. pub type Avc3 = AvcSampleEntry<{ AVC3_CODE }>;
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.rs
📄 CodeRabbit inference engine (CLAUDE.md)
**/*.rs: Code must be clippy-clean (cargo clippy --all-targets --all-features -- -D warnings)
Code must be formatted with rustfmt (cargo fmt) and pass formatting checks (cargo fmt -- --check)
Write unit tests inline within modules using #[cfg(test)]
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
src/**/*.rs
📄 CodeRabbit inference engine (CLAUDE.md)
src/**/*.rs: Each MP4 atom type must implement the Atom trait with KIND, decode_body(), and encode_body()
Use Header::read_from() to handle large atoms without loading entire contents into memory
All operations should return Result using custom error types defined in error.rs
Use the custom Buf/BufMut traits for contiguous byte slices instead of ad-hoc buffer handling
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
src/moov/**
📄 CodeRabbit inference engine (CLAUDE.md)
Place movie metadata atom implementations under src/moov/ (and its hierarchy: trak/mdia/minf/stbl/stsd)
Files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
🧠 Learnings (5)
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/**/*.rs : Each MP4 atom type must implement the Atom trait with KIND, decode_body(), and encode_body()
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/any.rs : Register new atom types in the Any enum (src/any.rs) to enable flexible decoding
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to src/test/** : Put integration tests under src/test/ with sample MP4 files for different codecs
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to **/*.rs : Code must be formatted with rustfmt (cargo fmt) and pass formatting checks (cargo fmt -- --check)
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
📚 Learning: 2025-08-31T04:06:57.364Z
Learnt from: CR
Repo: kixelated/mp4-atom PR: 0
File: CLAUDE.md:0-0
Timestamp: 2025-08-31T04:06:57.364Z
Learning: Applies to **/*.rs : Code must be clippy-clean (cargo clippy --all-targets --all-features -- -D warnings)
Applied to files:
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs
🧬 Code graph analysis (1)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (1)
src/moov/trak/mdia/minf/stbl/stsd/mod.rs (1)
decode(99-119)
🔇 Additional comments (2)
src/moov/trak/mdia/minf/stbl/stsd/h264/avc3.rs (2)
48-92: LGTM!The
bbc_expected()function provides comprehensive expected data, and theroundtrip()helper follows standard encode/decode test patterns.
94-145: Excellent test coverage!The tests are comprehensive and well-structured:
- Basic roundtrip validation
- Extended testing with all optional fields (btrt, colr, pasp, taic)
- Real BBC stream test with thorough structural validation
The BBC test (lines 121-145) directly addresses the past review comment requesting validation with real avc3 boxes, and it carefully verifies both the structure and decoded content.
3 participants
This PR aims to add avc3 support.
It's an initial implementation, reusing all the avc1 codebase. Tested it with a bbc test stream.
Tested with this live stream: https://vs-dash-ww-rd-live.akamaized.net/pl/testcard2020/avc-full.m3u8