wrtc 0.0.2__tar.gz → 0.2.0__tar.gz

{wrtc-0.0.2 → wrtc-0.2.0}/CHANGES.md

@@ -1,5 +1,89 @@
 # Change Log
 
+## Version 0.2.0
+
+**02.11.2026**
+
+**🤖 Pipecat bots run on the engine: install `wrtc[pipecat]` and change the imports, and clients stay as they are. See [Use with Pipecat](https://wrtc.marshal.dev/guides/pipecat/)**
+
+**🧩 More libraries built on aiortc run on the compat layer unchanged, with `webrtc.aiortc.install_alias()`**
+
+**🔌 Clients that can't use UDP reach a server over TCP, without TURN, with `RTCConfiguration.tcp_listen`. See [Clients without UDP](https://wrtc.marshal.dev/guides/connections/#clients-without-udp)**
+
+* Add `webrtc.pipecat`, Pipecat's SmallWebRTC transport on the native API, installed with `wrtc[pipecat]`, with `install_alias()` for Pipecat's development runner by [@MarshalX](https://github.com/MarshalX) in [#277](https://github.com/MarshalX/python-webrtc/pull/277)
+* Add a Pipecat voice bot example, with an echo that needs no keys by [@MarshalX](https://github.com/MarshalX) in [#277](https://github.com/MarshalX/python-webrtc/pull/277)
+* Add `webrtc.aiortc.install_alias()`, which makes `import aiortc` give the compat layer, for libraries that import aiortc themselves by [@MarshalX](https://github.com/MarshalX) in [#276](https://github.com/MarshalX/python-webrtc/pull/276)
+* Add aiortc's `rtp`, `jitterbuffer`, `utils` and `clock` modules and the ICE server helpers of `rtcicetransport` to the compat layer by [@MarshalX](https://github.com/MarshalX) in [#276](https://github.com/MarshalX/python-webrtc/pull/276)
+* Drop a compat layer track's buffered frames when its receiver's `_enabled` is set to `False`, and add the track's `_queue`, for libraries that skip stale frames by [@MarshalX](https://github.com/MarshalX) in [#276](https://github.com/MarshalX/python-webrtc/pull/276)
+* Add `RTCConfiguration.tcp_listen` for clients without UDP, also on `webrtc.pipecat`'s connection and request handler by [@MarshalX](https://github.com/MarshalX) in [#278](https://github.com/MarshalX/python-webrtc/pull/278)
+* Add a library comparison guide, a README demo and copy-page actions to the docs by [@MarshalX](https://github.com/MarshalX) in [#271](https://github.com/MarshalX/python-webrtc/pull/271)
+* Fix a video frame leak in the compat layer when a sender's track is replaced or stopped by [@MarshalX](https://github.com/MarshalX) in [#278](https://github.com/MarshalX/python-webrtc/pull/278)
+* Fix compat layer audio `pts` falling behind after dropped audio by [@MarshalX](https://github.com/MarshalX) in [#278](https://github.com/MarshalX/python-webrtc/pull/278)
+* Fix a cancelled stream read losing the next chunk by [@MarshalX](https://github.com/MarshalX) in [#278](https://github.com/MarshalX/python-webrtc/pull/278)
+* Fix ICE and DTLS transport states changing before their events by [@MarshalX](https://github.com/MarshalX) in [#278](https://github.com/MarshalX/python-webrtc/pull/278)
+* Fix the docs' contrast, mobile overflow and dead API links by [@MarshalX](https://github.com/MarshalX) in [#275](https://github.com/MarshalX/python-webrtc/pull/275)
+
+## Version 0.1.0
+
+**24.10.2026**
+
+**🧪 The first beta: the API follows the specifications and is ready to build on, while compatibility between versions isn't guaranteed until 1.0.0**
+
+**⚡ Data channels move 1.2–2.1 Gbps, 7.7–14.6x aiortc's throughput at 10–15% of its CPU per MB. See the [benchmarks](https://wrtc.marshal.dev/benchmarks/)**
+
+**🚀 A data channel opens 3.5 ms after `createOffer`, 7–8x faster than aiortc, and `getStats` takes 1.7 instead of 4.2 ms**
+
+**🎥 Simulcast for every encoder, and H.264 on macOS that recovers from encoder faults and falls back to OpenH264**
+
+**🧭 Transceiver, ICE, SCTP and track state change with the events that report them, as in browsers**
+
+**🧩 [Examples](https://wrtc.marshal.dev/examples/) of a server for a browser page, a Gemini Live voice agent, and WHIP/WHEP streams**
+
+* Add simulcast for every encoder, on Windows too: each layer is sent in the negotiated codec at its scaled size, whether or not the encoder does simulcast itself. VP9 and AV1 need a `scalabilityMode` such as `L1T1` for each layer by [@MarshalX](https://github.com/MarshalX) in [#254](https://github.com/MarshalX/python-webrtc/pull/254)
+* Fall back to OpenH264, once `await webrtc.openh264.install()` has run, when VideoToolbox can't encode or decode by [@MarshalX](https://github.com/MarshalX) in [#254](https://github.com/MarshalX/python-webrtc/pull/254)
+* Show transceiver, ICE, SCTP and track state with the events that change them: `mid`, `currentDirection` and the selected candidate pair change when their event fires, not before by [@MarshalX](https://github.com/MarshalX) in [#259](https://github.com/MarshalX/python-webrtc/pull/259)
+* Add a Python server for a browser page example: your camera comes back in grayscale, and Python answers a chat over a data channel by [@MarshalX](https://github.com/MarshalX) in [#268](https://github.com/MarshalX/python-webrtc/pull/268)
+* Add a Gemini Live voice agent example, with the API key kept on the server by [@MarshalX](https://github.com/MarshalX) in [#268](https://github.com/MarshalX/python-webrtc/pull/268)
+* Add a WHIP/WHEP example that plays a live stream in the terminal or publishes one by [@MarshalX](https://github.com/MarshalX) in [#268](https://github.com/MarshalX/python-webrtc/pull/268)
+* Add llms.txt, Markdown versions of the docs pages and Context7 and DeepWiki configs by [@MarshalX](https://github.com/MarshalX) in [#265](https://github.com/MarshalX/python-webrtc/pull/265)
+* Credit node-webrtc in the third-party licences by [@MarshalX](https://github.com/MarshalX) in [#262](https://github.com/MarshalX/python-webrtc/pull/262)
+* Speed up connection setup: keep libsrtp initialized and the network list updated, start the network manager and create port allocators without waiting on the network thread by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Speed up the first connection by running libsrtp's self-tests on their own thread by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Gather a candidate pool without waits between phases, and add ICE candidates in one signaling task by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Cut thread hops from negotiation and `getStats`, and share concurrent `getStats` calls: 20 of them take 5.7 instead of 80.9 ms, renegotiation 22% less time and warm setup with trickle 18% less by [@MarshalX](https://github.com/MarshalX) in [#255](https://github.com/MarshalX/python-webrtc/pull/255)
+* Cut signaling-thread waits from getters: `transceiver.mid` takes 1.0 instead of 9.8 µs, `sender.track` 38% less time, `getSelectedCandidatePair()` 46% less and renegotiation 15% less by [@MarshalX](https://github.com/MarshalX) in [#259](https://github.com/MarshalX/python-webrtc/pull/259)
+* Speed up data channels on macOS and Linux with a second, connected socket for a busy peer. Set the `PythonWebRTC-ConnectedSockets` field trial to `Disabled` to turn it off by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Speed up data channel sends with one network-thread task per send, and skip the local address lookup on every UDP packet by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Convert received data channel messages once and deliver them with less Python per message by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Copy data channel messages once: 10–11% more throughput and 7–11% less CPU per MB for 1 KB messages by [@MarshalX](https://github.com/MarshalX) in [#252](https://github.com/MarshalX/python-webrtc/pull/252)
+* Speed up event delivery, skip wake-ups while the event loop drains, and cache session descriptions between the events that change them by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Drop the unused audio processing and video render smoothing, which cost CPU on every received stream by [@MarshalX](https://github.com/MarshalX) in [#249](https://github.com/MarshalX/python-webrtc/pull/249)
+* Send VP8 and H.264 at 720p with 6–9% less CPU per frame by [@MarshalX](https://github.com/MarshalX) in [#256](https://github.com/MarshalX/python-webrtc/pull/256)
+* Wrap NV12 frames without copies, pass NV12 to VideoToolbox as it is, and pool frame buffers by [@MarshalX](https://github.com/MarshalX) in [#257](https://github.com/MarshalX/python-webrtc/pull/257)
+* Copy video frame data once into pooled buffers, and convert VideoToolbox output to I420 in one pass by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Copy encoded frames less and cache RTP sources: 14% less CPU per transformed frame, `getSynchronizationSources()` in 1.3 instead of 14.8 µs. A transform now queues up to 60 unread frames instead of 120 by [@MarshalX](https://github.com/MarshalX) in [#258](https://github.com/MarshalX/python-webrtc/pull/258)
+* Fix sockets raising SIGPIPE on macOS, which kills the process where its default handler is restored by [@MarshalX](https://github.com/MarshalX) in [#267](https://github.com/MarshalX/python-webrtc/pull/267)
+* Fix data channel stalls under load: UDP sockets get a receive buffer of at least 8 MB, which `WebRTC-SetSocketReceiveBuffer` overrides. On Linux, raise `net.core.rmem_max` to allow it. See [Field trials](https://wrtc.marshal.dev/guides/field-trials/) by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Fix a VideoToolbox frame that fails after the encoder took it: the session restarts with a key frame by [@MarshalX](https://github.com/MarshalX) in [#254](https://github.com/MarshalX/python-webrtc/pull/254)
+* Fix a connection kept alive after `close()` by the event loop it was bound to while closing by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Fix the operations of a closed event loop keeping their connections alive by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Fix the next operation starting before the code awaiting the previous one resumes, as in browsers by [@MarshalX](https://github.com/MarshalX) in [#259](https://github.com/MarshalX/python-webrtc/pull/259)
+* Fix `addIceCandidate` without a remote description rejecting late; it now rejects at once by [@MarshalX](https://github.com/MarshalX) in [#261](https://github.com/MarshalX/python-webrtc/pull/261)
+* Fix sends that fail as the remote end closes: they fire `error` with `data-channel-failure`, and their bytes stay in `bufferedAmount` by [@MarshalX](https://github.com/MarshalX) in [#252](https://github.com/MarshalX/python-webrtc/pull/252)
+* Fix remote data channels collected before their `open` event by [@MarshalX](https://github.com/MarshalX) in [#259](https://github.com/MarshalX/python-webrtc/pull/259)
+* Fix DTMF tones and gaps raised to longer minimums than the specification's by [@MarshalX](https://github.com/MarshalX) in [#253](https://github.com/MarshalX/python-webrtc/pull/253)
+* Fix standalone ICE gathering waiting 1 s between phases and gathering passive TCP candidates, and gather IPv6 where the host has it by [@MarshalX](https://github.com/MarshalX) in [#253](https://github.com/MarshalX/python-webrtc/pull/253)
+* Fix the end of candidates of an earlier remote description shown after an ICE restart by [@MarshalX](https://github.com/MarshalX) in [#253](https://github.com/MarshalX/python-webrtc/pull/253)
+* Fix the audio capture time, so the jitter doesn't jump after a send pause by [@MarshalX](https://github.com/MarshalX) in [#253](https://github.com/MarshalX/python-webrtc/pull/253)
+* Fix `VideoTrackGenerator` frames written back to back dropped as duplicates by [@MarshalX](https://github.com/MarshalX) in [#256](https://github.com/MarshalX/python-webrtc/pull/256)
+* Fix relayed remote video not scaled down for its encoder at low bitrates, and a sent track's own sinks getting the encoder's smaller frames by [@MarshalX](https://github.com/MarshalX) in [#256](https://github.com/MarshalX/python-webrtc/pull/256)
+* Fix encoded frames a transform drops staying in memory while the event loop is busy or closed by [@MarshalX](https://github.com/MarshalX) in [#260](https://github.com/MarshalX/python-webrtc/pull/260)
+* Fix writes to a writable stream slowing down as its queue grows by [@MarshalX](https://github.com/MarshalX) in [#264](https://github.com/MarshalX/python-webrtc/pull/264)
+* Fix the empty reason in the OpenAI example's connection error by [@MarshalX](https://github.com/MarshalX) in [#264](https://github.com/MarshalX/python-webrtc/pull/264)
+* Fix the OpenAI example printing the API key in `--help` by [@MarshalX](https://github.com/MarshalX) in [#268](https://github.com/MarshalX/python-webrtc/pull/268)
+* Fix the memory benchmark table when a run stops early at the memory limit, and benchmark charts crashing on a zero value or baseline by [@MarshalX](https://github.com/MarshalX) in [#264](https://github.com/MarshalX/python-webrtc/pull/264)
+* Fix the connections guide's advice to wait for complete ICE gathering, which can take about 40 seconds with a STUN server by [@MarshalX](https://github.com/MarshalX) in [#268](https://github.com/MarshalX/python-webrtc/pull/268)
+
 ## Version 0.0.2
 
 **15.10.2026**

{wrtc-0.0.2 → wrtc-0.2.0}/CMakeLists.txt

@@ -33,5 +33,8 @@ find_package(Python 3.9 REQUIRED COMPONENTS Interpreter Development.Module)
 find_package(pybind11 CONFIG REQUIRED)
 
 include(cmake/openh264.cmake)
+if(WIN32)
+  include(cmake/libwebrtc_windows.cmake)
+endif()
 
 add_subdirectory(python-webrtc/cpp)

{wrtc-0.0.2 → wrtc-0.2.0}/Makefile

@@ -2,6 +2,9 @@
 
 # pinned to the clang-tidy of .github/scripts/tidy.sh
 CLANG_FORMAT := uvx clang-format==22.1.8
+# pinned so a new release can't break lint; bump together with .github/workflows/ci.yml
+RUFF := uvx ruff==0.17.1
+PYREFLY := uvx pyrefly==1.3.2
 CPP_SRC = $(shell find python-webrtc/cpp/src -name '*.cpp' -o -name '*.h' -o -name '*.mm')
 
 # editable install; the extension is rebuilt automatically on import after C++ changes
@@ -50,15 +53,15 @@ stacks:
 	uv run --no-sync python -m scripts.debug.stacks $(PID) $(O)
 
 lint: format-check
-	uvx ruff check
-	uvx ruff format --check
+	$(RUFF) check
+	$(RUFF) format --check
 
 typecheck:
-	uvx pyrefly check
+	$(PYREFLY) check
 
 format:
-	uvx ruff check --fix
-	uvx ruff format
+	$(RUFF) check --fix
+	$(RUFF) format
 	$(CLANG_FORMAT) -i $(CPP_SRC)
 
 format-check:

{wrtc-0.0.2 → wrtc-0.2.0}/PKG-INFO

@@ -1,13 +1,13 @@
 Metadata-Version: 2.4
 Name: wrtc
-Version: 0.0.2
+Version: 0.2.0
 Summary: Chrome's WebRTC engine (libwebrtc M152), with the browser API, in Python
 Keywords: webrtc,libwebrtc,rtc,peer-to-peer,data-channel,video,audio,voip,asyncio,w3c
 Author-Email: "Ilya (Marshal)" <ilya@marshal.dev>
 License-Expression: BSD-3-Clause
 License-File: LICENSE.md
 License-File: THIRD_PARTY_LICENSES.md
-Classifier: Development Status :: 3 - Alpha
+Classifier: Development Status :: 4 - Beta
 Classifier: Natural Language :: English
 Classifier: Intended Audience :: Developers
 Classifier: Framework :: AsyncIO
@@ -51,6 +51,8 @@ Requires-Dist: typing_extensions>=4.10
 Provides-Extra: aiortc
 Requires-Dist: pyee<15,>=13; extra == "aiortc"
 Requires-Dist: av<18,>=14; extra == "aiortc"
+Provides-Extra: pipecat
+Requires-Dist: pipecat-ai<2,>=1.15; python_version >= "3.11" and extra == "pipecat"
 Description-Content-Type: text/markdown
 
 <p align="center">
@@ -73,6 +75,15 @@ Description-Content-Type: text/markdown
     <a href="https://pypi.org/project/wrtc/">PyPI</a>
 </p>
 
+<p align="center">
+    <picture>
+        <source media="(prefers-color-scheme: dark)" type="image/avif" srcset="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-dark.avif">
+        <source type="image/avif" srcset="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-light.avif">
+        <source media="(prefers-color-scheme: dark)" srcset="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-dark.gif">
+        <img src="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-light.gif" alt="A browser page sends its camera to a Python server, which sends it back in grayscale and answers a chat message over a data channel">
+    </picture>
+</p>
+
 ---
 
 Python WebRTC binds **libwebrtc M152**, the engine inside Chrome, and exposes it through the **W3C APIs you
@@ -153,11 +164,13 @@ equal to the strings of the specification, dictionaries are typed dataclasses, a
 | Browser API (`RTCPeerConnection`, `createOffer`, …)     | ✅                    | similar           | ❌ rooms                    | ❌ GObject signals     |
 | Media path                                              | native               | Python            | native                     | native                |
 | VP9 and AV1                                             | ✅                    | ❌                 | ✅                          | ✅                     |
-| `pip install` on Linux, macOS and Windows               | ✅                    | ✅                 | ✅                          | ❌ system packages     |
+| `pip install` on Linux, macOS and Windows               | ✅                    | ✅                 | ✅                          | ❌ no Linux wheels     |
 
-Against aiortc, it opens connections 5–7x faster, moves 4.5–9x more data through a data channel and spends less CPU on
-audio. See the [benchmarks](https://wrtc.marshal.dev/benchmarks/), which also measure Pion, libdatachannel and
-webrtc-rs.
+Against aiortc, it opens connections 7.3–8.3x faster, moves 7.7–14.6x more data through a data channel and spends
+less CPU on audio. See the [benchmarks](https://wrtc.marshal.dev/benchmarks/), which also measure Pion, libdatachannel
+and webrtc-rs.
+
+The full comparison, and when another library is the better choice: [Choosing a Python WebRTC library](https://wrtc.marshal.dev/guides/comparison/).
 
 ## Codecs
 
@@ -194,6 +207,13 @@ Complete programs, ready to copy. The ones with dependencies declare them inline
 uv run https://raw.githubusercontent.com/MarshalX/python-webrtc/main/examples/janus_streaming.py
 ```
 
+## Pipecat
+
+Install `wrtc[pipecat]` and import the SmallWebRTC transport from `webrtc.pipecat`, and your
+[Pipecat](https://www.pipecat.ai) bot runs on Chrome's engine, with its clients unchanged. For Pipecat's development
+runner, call `webrtc.pipecat.install_alias()`. See [Use with Pipecat](https://wrtc.marshal.dev/guides/pipecat/) and the
+[voice bot example](https://github.com/MarshalX/python-webrtc/blob/main/examples/pipecat_bot.py).
+
 ## Status
 
 Pre-1.0: the API follows the specifications, but compatibility between versions isn't guaranteed until 1.0.0. See

{wrtc-0.0.2 → wrtc-0.2.0}/README.md

@@ -18,6 +18,15 @@
     <a href="https://pypi.org/project/wrtc/">PyPI</a>
 </p>
 
+<p align="center">
+    <picture>
+        <source media="(prefers-color-scheme: dark)" type="image/avif" srcset="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-dark.avif">
+        <source type="image/avif" srcset="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-light.avif">
+        <source media="(prefers-color-scheme: dark)" srcset="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-dark.gif">
+        <img src="https://github.com/MarshalX/python-webrtc/raw/main/docs/source/_static/img/demo-light.gif" alt="A browser page sends its camera to a Python server, which sends it back in grayscale and answers a chat message over a data channel">
+    </picture>
+</p>
+
 ---
 
 Python WebRTC binds **libwebrtc M152**, the engine inside Chrome, and exposes it through the **W3C APIs you
@@ -98,11 +107,13 @@ equal to the strings of the specification, dictionaries are typed dataclasses, a
 | Browser API (`RTCPeerConnection`, `createOffer`, …)     | ✅                    | similar           | ❌ rooms                    | ❌ GObject signals     |
 | Media path                                              | native               | Python            | native                     | native                |
 | VP9 and AV1                                             | ✅                    | ❌                 | ✅                          | ✅                     |
-| `pip install` on Linux, macOS and Windows               | ✅                    | ✅                 | ✅                          | ❌ system packages     |
+| `pip install` on Linux, macOS and Windows               | ✅                    | ✅                 | ✅                          | ❌ no Linux wheels     |
 
-Against aiortc, it opens connections 5–7x faster, moves 4.5–9x more data through a data channel and spends less CPU on
-audio. See the [benchmarks](https://wrtc.marshal.dev/benchmarks/), which also measure Pion, libdatachannel and
-webrtc-rs.
+Against aiortc, it opens connections 7.3–8.3x faster, moves 7.7–14.6x more data through a data channel and spends
+less CPU on audio. See the [benchmarks](https://wrtc.marshal.dev/benchmarks/), which also measure Pion, libdatachannel
+and webrtc-rs.
+
+The full comparison, and when another library is the better choice: [Choosing a Python WebRTC library](https://wrtc.marshal.dev/guides/comparison/).
 
 ## Codecs
 
@@ -139,6 +150,13 @@ Complete programs, ready to copy. The ones with dependencies declare them inline
 uv run https://raw.githubusercontent.com/MarshalX/python-webrtc/main/examples/janus_streaming.py
 ```
 
+## Pipecat
+
+Install `wrtc[pipecat]` and import the SmallWebRTC transport from `webrtc.pipecat`, and your
+[Pipecat](https://www.pipecat.ai) bot runs on Chrome's engine, with its clients unchanged. For Pipecat's development
+runner, call `webrtc.pipecat.install_alias()`. See [Use with Pipecat](https://wrtc.marshal.dev/guides/pipecat/) and the
+[voice bot example](https://github.com/MarshalX/python-webrtc/blob/main/examples/pipecat_bot.py).
+
 ## Status
 
 Pre-1.0: the API follows the specifications, but compatibility between versions isn't guaranteed until 1.0.0. See

{wrtc-0.0.2 → wrtc-0.2.0}/THIRD_PARTY_LICENSES.md

@@ -242,7 +242,8 @@ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 # aiortc
 The code of webrtc/aiortc/sdp.py, mediastreams.py, rtcconfiguration.py, rtcrtpparameters.py,
-rtcsessiondescription.py, stats.py, contrib/media.py and contrib/signaling.py, copied from it:
+rtcsessiondescription.py, stats.py, rtp.py, jitterbuffer.py, utils.py, clock.py, _ice_servers.py, contrib/media.py and
+contrib/signaling.py, copied from it:
 ```
 Copyright (c) Jeremy Lainé.
 All rights reserved.
@@ -2185,6 +2186,32 @@ SPDX-License-Identifier: BSD-2-Clause
 
 ```
 
+# node-webrtc
+The first versions of the C++ bindings (2022) were derived from it:
+```
+Copyright (c) 2019 The node-webrtc project authors. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without modification,
+are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+   list of conditions and the following disclaimer.
+2. Redistributions in binary form must reproduce the above copyright notice,
+   this list of conditions and the following disclaimer in the documentation
+   and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+```
+
 # ooura
 ```
 /*

{wrtc-0.0.2 → wrtc-0.2.0}/cmake/libcxx/update.py

@@ -37,19 +37,27 @@ import subprocess
 import tempfile
 from concurrent.futures import ThreadPoolExecutor
 from pathlib import Path
-from typing import TypedDict
+from typing import NamedTuple, TypedDict
+
+
+class LlvmDir(NamedTuple):
+    """How WebRTC pins a directory of llvm-project."""
+
+    deps_path: str
+    cmake_variable: str
+
 
 REPO = 'repos/llvm/llvm-project'
 CHROMIUM_CONFIG = ('__config_site', '__assertion_handler')
-# llvm-project dir -> (its DEPS path in WebRTC, CMake variable of its commit)
 LLVM_DIRS = {
-    'libcxx': ('src/third_party/libc++/src', 'LIBCXX_LLVM_COMMIT'),
-    'libcxxabi': ('src/third_party/libc++abi/src', 'LIBCXXABI_LLVM_COMMIT'),
-    'libc': ('src/third_party/llvm-libc/src', 'LLVM_LIBC_COMMIT'),
+    'libcxx': LlvmDir('src/third_party/libc++/src', 'LIBCXX_LLVM_COMMIT'),
+    'libcxxabi': LlvmDir('src/third_party/libc++abi/src', 'LIBCXXABI_LLVM_COMMIT'),
+    'libc': LlvmDir('src/third_party/llvm-libc/src', 'LLVM_LIBC_COMMIT'),
 }
 # llvm-project dirs the runtime sources include from: their own trees, and llvm-libc (-I libc)
 RUNTIME_TREES = ('libcxx/src', 'libcxxabi/src', 'libc')
 INCLUDE = re.compile(rb'^\s*#\s*include\s*[<"]([^>"]+)[>"]', re.MULTILINE)
+GITHUB_TIME = '%Y-%m-%dT%H:%M:%SZ'
 
 
 class TreeEntry(TypedDict):
@@ -90,7 +98,8 @@ def gh(path: str) -> bytes:
 
 def git_tree(sha: str, *, recursive: bool = False) -> Tree:
     """A git tree of llvm-project."""
-    tree: Tree = json.loads(gh(f'{REPO}/git/trees/{sha}' + ('?recursive=1' if recursive else '')))
+    query = '?recursive=1' if recursive else ''
+    tree: Tree = json.loads(gh(f'{REPO}/git/trees/{sha}{query}'))
     return tree
 
 
@@ -141,22 +150,32 @@ def git_fetch(url: str, ref: str, repo: str) -> str:
     return subprocess.check_output([git, '-C', repo, 'rev-parse', 'FETCH_HEAD'], text=True).strip()
 
 
+def commits_near(name: str, committed: str) -> list[Commit]:
+    """The llvm-project commits that changed a directory within a day of an ISO 8601 commit time."""
+    date = dt.datetime.fromisoformat(committed).astimezone(dt.timezone.utc)
+    since = (date - dt.timedelta(days=1)).strftime(GITHUB_TIME)
+    until = (date + dt.timedelta(days=1)).strftime(GITHUB_TIME)
+    commits: list[Commit] = json.loads(gh(f'{REPO}/commits?path={name}&since={since}&until={until}&per_page=100'))
+    return commits
+
+
 def resolve_llvm(deps: str, name: str) -> str:
     """Chromium mirrors each llvm-project dir as its own repo: finds the llvm-project commit with the same tree."""
-    match = re.search(rf"'{re.escape(LLVM_DIRS[name][0])}':\s*'([^'@]+)@([0-9a-f]+)'", deps)
+    deps_path = LLVM_DIRS[name].deps_path
+    match = re.search(rf"'{re.escape(deps_path)}':\s*'([^'@]+)@([0-9a-f]+)'", deps)
     if match is None:
-        msg = f'{LLVM_DIRS[name][0]} is not in the WebRTC DEPS'
+        msg = f'{deps_path} is not in the WebRTC DEPS'
         raise SystemExit(msg)
+
     url, revision = match.groups()
     with tempfile.TemporaryDirectory() as repo:
         git_fetch(url, revision, repo)
-        tree, committed = subprocess.check_output(
+        log = subprocess.check_output(
             [tool('git'), '-C', repo, 'log', '-1', '--format=%T %cI', 'FETCH_HEAD'], text=True
-        ).split()
-    date = dt.datetime.fromisoformat(committed).astimezone(dt.timezone.utc)
-    since, until = ((date + dt.timedelta(days=d)).strftime('%Y-%m-%dT%H:%M:%SZ') for d in (-1, 1))
-    commits: list[Commit] = json.loads(gh(f'{REPO}/commits?path={name}&since={since}&until={until}&per_page=100'))
-    for commit in commits:
+        )
+    tree, committed = log.split()
+
+    for commit in commits_near(name, committed):
         if tree_sha(commit['sha'], name) == tree:
             return commit['sha']
     msg = f'no llvm-project commit has the {name} tree of {url}@{revision}'
@@ -167,10 +186,14 @@ def runtime_sources(tag: str) -> list[str]:
     """The libc++ and libc++abi sources of Chromium's Linux build, as llvm-project paths."""
     sources: list[str] = []
     for gn, llvm in (('libc%2B%2B', 'libcxx'), ('libc%2B%2Babi', 'libcxxabi')):
-        for line in chromium(f'buildtools/third_party/{gn}/BUILD.gn', tag).decode().splitlines():
+        build_gn = chromium(f'buildtools/third_party/{gn}/BUILD.gn', tag).decode()
+        for line in build_gn.splitlines():
             match = re.search(r'"//third_party/libc\+\+(?:abi)?/src/src/([^"]+\.cpp)"', line)
-            if match is not None and not line.lstrip().startswith('#') and 'win32' not in match[1]:
-                sources.append(f'{llvm}/src/{match[1]}')
+            if match is None or line.lstrip().startswith('#'):
+                continue
+            source = match[1]
+            if 'win32' not in source:
+                sources.append(f'{llvm}/src/{source}')
     return sorted(set(sources))
 
 
@@ -189,20 +212,27 @@ def pin_runtime(commits: dict[str, str], tag: str) -> None:
     """Pins the runtime sources and, transitively, every file they include from RUNTIME_TREES."""
     tree: dict[str, str] = {}
     for root in RUNTIME_TREES:
-        tree.update({f'{root}/{e["path"]}': e['sha'] for e in subtree(commits[root.split('/')[0]], root)})
+        llvm_dir = root.split('/')[0]
+        tree.update({f'{root}/{e["path"]}': e['sha'] for e in subtree(commits[llvm_dir], root)})
+
     sources = runtime_sources(tag)
     pinned: dict[str, bytes] = {}
     queue = list(sources)
     with ThreadPoolExecutor(8) as pool:
         while len(queue) > 0:
-            pinned.update(zip(queue, pool.map(blob, [tree[p] for p in queue])))
-            found = set[str]().union(*(includes(path, pinned[path]) for path in queue))
+            contents = pool.map(blob, [tree[path] for path in queue])
+            pinned.update(zip(queue, contents))
+            found: set[str] = set()
+            for path in queue:
+                found |= includes(path, pinned[path])
             queue = sorted(c for c in found if c in tree and c not in pinned)
+
     # CMake compiles every pinned libcxx/libcxxabi .cpp, so only the sources may be among them
     stray = [p for p in pinned if p.endswith('.cpp') and p not in sources and not p.startswith('libc/')]
     if len(stray) > 0:
         msg = f'sources include other .cpp files: {stray}'
         raise SystemExit(msg)
+
     lines = [f'{hashlib.sha256(data).hexdigest()}  {path}\n' for path, data in sorted(pinned.items())]
     (Path(__file__).parent / 'runtime.sha256').write_text(''.join(lines))
     print(f'{len(sources)} runtime sources, {len(lines)} files pinned')
@@ -227,7 +257,7 @@ def pin_headers(commit: str, headers: Path) -> None:
         msg = f'the prebuilt headers differ from llvm-project@{commit}: {mismatches[:10]}'
         raise SystemExit(msg)
 
-    missing = [e for e in tree if e['path'] not in local and '.' not in e['path'].rsplit('/', 1)[-1]]
+    missing = [e for e in tree if e['path'] not in local and '.' not in posixpath.basename(e['path'])]
 
     def sha256(entry: TreeEntry) -> str:
         data = blob(entry['sha'])
@@ -262,8 +292,8 @@ def main() -> None:
     pin_config(args.chromium_tag)
     pin_runtime(commits, args.chromium_tag)
 
-    for name, (_, variable) in LLVM_DIRS.items():
-        print(f'set({variable} {commits[name]})')
+    for name, llvm_dir in LLVM_DIRS.items():
+        print(f'set({llvm_dir.cmake_variable} {commits[name]})')
     print(f'set(LIBCXX_CHROMIUM_TAG {args.chromium_tag})')
 
 

{wrtc-0.0.2 → wrtc-0.2.0}/cmake/libwebrtc.cmake

@@ -69,6 +69,37 @@ function(_fetch_pinned manifest base_url dest)
   endforeach()
 endfunction()
 
+# Downloads a file of libwebrtc at LIBWEBRTC_COMMIT; googlesource serves files only base64-encoded
+function(_fetch_webrtc path sha256 dest)
+  set(_encoded "${dest}.b64")
+  foreach(_attempt RANGE 1 4)
+    file(DOWNLOAD "https://webrtc.googlesource.com/src/+/${LIBWEBRTC_COMMIT}/${path}?format=TEXT"
+        "${_encoded}" STATUS _status TLS_VERIFY ON)
+    list(GET _status 0 _code)
+    if(_code EQUAL 0)
+      execute_process(
+          COMMAND "${Python_EXECUTABLE}" -c
+              "import base64, pathlib, sys; p = pathlib.Path(sys.argv[2]); p.parent.mkdir(parents=True, exist_ok=True); p.write_bytes(base64.b64decode(pathlib.Path(sys.argv[1]).read_bytes()))"
+              "${_encoded}" "${dest}"
+          RESULT_VARIABLE _decoded)
+      if(_decoded EQUAL 0)
+        file(SHA256 "${dest}" _actual)
+        if(_actual STREQUAL sha256)
+          file(REMOVE "${_encoded}")
+          return()
+        endif()
+        set(_status "SHA256 mismatch: ${_actual}")
+      else()
+        set(_status "can't decode it")
+      endif()
+    endif()
+    if(_attempt EQUAL 4)
+      message(FATAL_ERROR "Failed to fetch ${path} at ${LIBWEBRTC_COMMIT}: ${_status}")
+    endif()
+    execute_process(COMMAND "${CMAKE_COMMAND}" -E sleep ${_attempt})
+  endforeach()
+endfunction()
+
 # Header directories (relative to include/) needed to compile against the public API.
 set(LIBWEBRTC_HEADER_DIRS
     api audio call common_audio common_video logging media modules net p2p pc

wrtc-0.2.0/cmake/libwebrtc_windows.cmake

@@ -0,0 +1,46 @@
+# Fetches the libwebrtc sources the Windows prebuilt compiles out: the simulcast encoder adapter, and the encoder
+# fallback wrapper it creates.
+#
+# Sets LIBWEBRTC_WINDOWS_SRC.
+
+set(_sources
+    "media/engine/simulcast_encoder_adapter.cc"
+    "0fac2e2e51e3bf0a595d10b3ce5e197b1c40e4d11cbd8852d84e1fcf104a6e45"
+    "api/video_codecs/video_encoder_software_fallback_wrapper.cc"
+    "52251651cb273455adcc8db4dfe1dc654a243e709beb5a528bdbb0279f614908")
+
+if(LIBWEBRTC_CACHE_DIR)
+  set(_dir "${LIBWEBRTC_CACHE_DIR}/libwebrtc-windows-src-${LIBWEBRTC_VERSION}")
+else()
+  set(_dir "${CMAKE_BINARY_DIR}/libwebrtc-windows-src")
+endif()
+
+set(LIBWEBRTC_WINDOWS_SRC "")
+set(_stamp "${_dir}/.complete")
+set(_stamp_content "${LIBWEBRTC_COMMIT};${_sources}")
+set(_stamp_current "")
+if(EXISTS "${_stamp}")
+  file(READ "${_stamp}" _stamp_current)
+endif()
+set(_fetch FALSE)
+if(NOT _stamp_current STREQUAL _stamp_content)
+  message(STATUS "Fetching libwebrtc's simulcast encoder adapter, which the Windows prebuilt lacks")
+  file(REMOVE_RECURSE "${_dir}")
+  set(_fetch TRUE)
+endif()
+
+list(LENGTH _sources _count)
+math(EXPR _last "${_count} - 1")
+foreach(_i RANGE 0 ${_last} 2)
+  math(EXPR _j "${_i} + 1")
+  list(GET _sources ${_i} _path)
+  list(GET _sources ${_j} _sha)
+  if(_fetch)
+    _fetch_webrtc("${_path}" "${_sha}" "${_dir}/${_path}")
+  endif()
+  list(APPEND LIBWEBRTC_WINDOWS_SRC "${_dir}/${_path}")
+endforeach()
+
+if(_fetch)
+  file(WRITE "${_stamp}" "${_stamp_content}")
+endif()

{wrtc-0.0.2 → wrtc-0.2.0}/cmake/openh264.cmake

@@ -16,37 +16,6 @@ endif()
 set(OPENH264_INCLUDE_DIR "${_dir}/include")
 set(OPENH264_ENCODER_SRC "${_dir}/webrtc/${_encoder}.cc")
 
-# Downloads a file of libwebrtc at LIBWEBRTC_COMMIT; googlesource serves files only base64-encoded
-function(_fetch_webrtc path sha256 dest)
-  set(_encoded "${dest}.b64")
-  foreach(_attempt RANGE 1 4)
-    file(DOWNLOAD "https://webrtc.googlesource.com/src/+/${LIBWEBRTC_COMMIT}/${path}?format=TEXT"
-        "${_encoded}" STATUS _status TLS_VERIFY ON)
-    list(GET _status 0 _code)
-    if(_code EQUAL 0)
-      execute_process(
-          COMMAND "${Python_EXECUTABLE}" -c
-              "import base64, pathlib, sys; p = pathlib.Path(sys.argv[2]); p.parent.mkdir(parents=True, exist_ok=True); p.write_bytes(base64.b64decode(pathlib.Path(sys.argv[1]).read_bytes()))"
-              "${_encoded}" "${dest}"
-          RESULT_VARIABLE _decoded)
-      if(_decoded EQUAL 0)
-        file(SHA256 "${dest}" _actual)
-        if(_actual STREQUAL sha256)
-          file(REMOVE "${_encoded}")
-          return()
-        endif()
-        set(_status "SHA256 mismatch: ${_actual}")
-      else()
-        set(_status "can't decode it")
-      endif()
-    endif()
-    if(_attempt EQUAL 4)
-      message(FATAL_ERROR "Failed to fetch ${path} at ${LIBWEBRTC_COMMIT}: ${_status}")
-    endif()
-    execute_process(COMMAND "${CMAKE_COMMAND}" -E sleep ${_attempt})
-  endforeach()
-endfunction()
-
 file(SHA256 "${CMAKE_CURRENT_LIST_DIR}/openh264.sha256" _manifest_hash)
 set(_stamp "${_dir}/.complete")
 set(_stamp_content