✨ First commit

.gitattributes

@@ -0,0 +1,2 @@
+# Auto detect text files and perform LF normalization
+* text=auto

.gitignore

@@ -0,0 +1,11 @@
+# Aulavirtuale-dl ignores
+VideoLezioni/
+
+# Python ignores
+**/__pycache__/
+dist/
+*.egg-info/
+**/*.pyc
+
+# PyCharm ignores
+.idea/

README.md

@@ -0,0 +1,3 @@
+# bbbdl
+
+A tool for downloading BigBlueButton meetings

bbbdl/__main__.py

@@ -0,0 +1,20 @@
+import ffmpeg
+import click
+from .resources import Meeting
+from .composer import compose_lesson
+
+
+@click.command()
+@click.option("-i", "--input-url", type=str, prompt=True,
+              help="The URL of the meeting to download.")
+@click.option("-o", "--output-file", type=str, prompt=True,
+              help="The file the video should be written to.")
+def download(input_url, output_file):
+    meeting = Meeting.from_url(input_url)
+    streams = compose_lesson(meeting)
+    output = ffmpeg.output(*streams, output_file)
+    output.run()
+
+
+if __name__ == "__main__":
+    download()

bbbdl/composer.py

@@ -0,0 +1,24 @@
+from typing import *
+import ffmpeg
+from .resources import Meeting
+
+
+def compose_screensharing(meeting: Meeting) -> Tuple[ffmpeg.Stream, ffmpeg.Stream]:
+    """Keep the deskshare video and the webcam audio, while discarding the rest."""
+
+    return (
+        meeting.deskshare.as_stream().video,
+        meeting.webcams.as_stream().audio,
+    )
+
+
+def compose_lesson(meeting: Meeting) -> Tuple[ffmpeg.Stream, ffmpeg.Stream]:
+    """Keep slides, deskshare video and webcam audio, while discarding the rest."""
+
+    video_stream, audio_stream = compose_screensharing(meeting)
+
+    for shape in meeting.shapes:
+        video_stream = ffmpeg.overlay(video_stream, shape.resource.as_stream().video,
+                                      enable=f"between(t, {shape.start}, {shape.end})")
+
+    return video_stream, audio_stream

bbbdl/errors.py

@@ -0,0 +1,6 @@
+class BBBDLError(Exception):
+    pass
+
+
+class DownloadError(BBBDLError):
+    pass

bbbdl/resources.py

@@ -0,0 +1,69 @@
+from __future__ import annotations
+from typing import *
+import dataclasses
+import requests
+import bs4
+import ffmpeg
+from .urlhandler import playback_to_data
+
+
+@dataclasses.dataclass()
+class Resource:
+    href: str
+
+    def as_stream(self, **kwargs) -> ffmpeg.Stream:
+        return ffmpeg.input(self.href, **kwargs)
+
+
+@dataclasses.dataclass()
+class Shape:
+    resource: Resource
+    start: float
+    end: float
+
+    @classmethod
+    def from_tag(cls, tag: bs4.Tag, *, base_url: str) -> Shape:
+        # No, `"in" not in tag` does not work
+        if not tag["in"]:
+            raise ValueError("Tag has no 'in' parameter")
+        if not tag["out"]:
+            raise ValueError("Tag has no 'out' parameter")
+        if not tag["xlink:href"]:
+            raise ValueError("Tag has no 'xlink:href' parameter")
+
+        return cls(
+            resource=Resource(href=f"{base_url}/{tag['xlink:href']}"),
+            start=float(tag["in"]),
+            end=float(tag["out"]),
+        )
+
+
+@dataclasses.dataclass()
+class Meeting:
+    deskshare: Resource
+    webcams: Resource
+    shapes: List[Shape]
+
+    @classmethod
+    def from_base_url(cls, base_url: str, meeting_id: str) -> Meeting:
+        deskshare = Resource(href=f"{base_url}/presentation/{meeting_id}/deskshare/deskshare.webm")
+        webcams = Resource(href=f"{base_url}/presentation/{meeting_id}/video/webcams.mp4")
+
+        shape_soup = bs4.BeautifulSoup(requests.get(f"{base_url}/presentation/{meeting_id}/shapes.svg").text,
+                                       "lxml")
+        shapes: List[Shape] = []
+        for tag in shape_soup.find_all("image"):
+            try:
+                shapes.append(Shape.from_tag(tag, base_url=f"{base_url}/presentation/{meeting_id}"))
+            except ValueError:
+                continue
+
+        return cls(
+            deskshare=deskshare,
+            webcams=webcams,
+            shapes=shapes
+        )
+
+    @classmethod
+    def from_url(cls, url: str) -> Meeting:
+        return cls.from_base_url(*playback_to_data(url))

bbbdl/urlhandler.py

@@ -0,0 +1,12 @@
+from typing import *
+import re
+
+
+split_regex = re.compile(r"^(https?://.+)/playback/presentation/2\.0/playback\.html\?meetingId=([0-9a-f-]+)$")
+
+
+def playback_to_data(url) -> Sequence[str]:
+    match = split_regex.match(url)
+    if not match:
+        raise ValueError("Could not split URL in base_url and meeting_id")
+    return match.groups()

poetry.lock

@@ -0,0 +1,209 @@
+[[package]]
+name = "beautifulsoup4"
+version = "4.9.3"
+description = "Screen-scraping library"
+category = "main"
+optional = false
+python-versions = "*"
+
+[package.extras]
+html5lib = ["html5lib"]
+lxml = ["lxml"]
+
+[package.dependencies]
+[package.dependencies.soupsieve]
+version = ">1.2"
+python = ">=3.0"
+
+[[package]]
+name = "certifi"
+version = "2020.6.20"
+description = "Python package for providing Mozilla's CA Bundle."
+category = "main"
+optional = false
+python-versions = "*"
+
+[[package]]
+name = "chardet"
+version = "3.0.4"
+description = "Universal encoding detector for Python 2 and 3"
+category = "main"
+optional = false
+python-versions = "*"
+
+[[package]]
+name = "click"
+version = "7.1.2"
+description = "Composable command line interface toolkit"
+category = "main"
+optional = false
+python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
+
+[[package]]
+name = "ffmpeg-python"
+version = "0.2.0"
+description = "Python bindings for FFmpeg - with complex filtering support"
+category = "main"
+optional = false
+python-versions = "*"
+
+[package.extras]
+dev = ["future (0.17.1)", "numpy (1.16.4)", "pytest-mock (1.10.4)", "pytest (4.6.1)", "Sphinx (2.1.0)", "tox (3.12.1)"]
+
+[package.dependencies]
+future = "*"
+
+[[package]]
+name = "future"
+version = "0.18.2"
+description = "Clean single-source support for Python 3 and 2"
+category = "main"
+optional = false
+python-versions = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*"
+
+[[package]]
+name = "idna"
+version = "2.10"
+description = "Internationalized Domain Names in Applications (IDNA)"
+category = "main"
+optional = false
+python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*"
+
+[[package]]
+name = "lxml"
+version = "4.6.1"
+description = "Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API."
+category = "main"
+optional = false
+python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, != 3.4.*"
+
+[package.extras]
+cssselect = ["cssselect (>=0.7)"]
+html5 = ["html5lib"]
+htmlsoup = ["beautifulsoup4"]
+source = ["Cython (>=0.29.7)"]
+
+[[package]]
+name = "requests"
+version = "2.24.0"
+description = "Python HTTP for Humans."
+category = "main"
+optional = false
+python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
+
+[package.extras]
+security = ["pyOpenSSL (>=0.14)", "cryptography (>=1.3.4)"]
+socks = ["PySocks (>=1.5.6,<1.5.7 || >1.5.7)", "win-inet-pton"]
+
+[package.dependencies]
+certifi = ">=2017.4.17"
+chardet = ">=3.0.2,<4"
+idna = ">=2.5,<3"
+urllib3 = ">=1.21.1,<1.25.0 || >1.25.0,<1.25.1 || >1.25.1,<1.26"
+
+[[package]]
+name = "soupsieve"
+version = "2.0.1"
+description = "A modern CSS selector implementation for Beautiful Soup."
+category = "main"
+optional = false
+python-versions = ">=3.5"
+marker = "python_version >= \"3.0\""
+
+[[package]]
+name = "urllib3"
+version = "1.25.11"
+description = "HTTP library with thread-safe connection pooling, file post, and more."
+category = "main"
+optional = false
+python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*, <4"
+
+[package.extras]
+brotli = ["brotlipy (>=0.6.0)"]
+secure = ["pyOpenSSL (>=0.14)", "cryptography (>=1.3.4)", "idna (>=2.0.0)", "certifi", "ipaddress"]
+socks = ["PySocks (>=1.5.6,<1.5.7 || >1.5.7,<2.0)"]
+
+[metadata]
+lock-version = "1.0"
+python-versions = "^3.8"
+content-hash = "8658967e9571ffdc038fe220a009684044442d130b1da057b7a099aa4af53244"
+
+[metadata.files]
+beautifulsoup4 = [
+    {file = "beautifulsoup4-4.9.3-py2-none-any.whl", hash = "sha256:4c98143716ef1cb40bf7f39a8e3eec8f8b009509e74904ba3a7b315431577e35"},
+    {file = "beautifulsoup4-4.9.3-py3-none-any.whl", hash = "sha256:fff47e031e34ec82bf17e00da8f592fe7de69aeea38be00523c04623c04fb666"},
+    {file = "beautifulsoup4-4.9.3.tar.gz", hash = "sha256:84729e322ad1d5b4d25f805bfa05b902dd96450f43842c4e99067d5e1369eb25"},
+]
+certifi = [
+    {file = "certifi-2020.6.20-py2.py3-none-any.whl", hash = "sha256:8fc0819f1f30ba15bdb34cceffb9ef04d99f420f68eb75d901e9560b8749fc41"},
+    {file = "certifi-2020.6.20.tar.gz", hash = "sha256:5930595817496dd21bb8dc35dad090f1c2cd0adfaf21204bf6732ca5d8ee34d3"},
+]
+chardet = [
+    {file = "chardet-3.0.4-py2.py3-none-any.whl", hash = "sha256:fc323ffcaeaed0e0a02bf4d117757b98aed530d9ed4531e3e15460124c106691"},
+    {file = "chardet-3.0.4.tar.gz", hash = "sha256:84ab92ed1c4d4f16916e05906b6b75a6c0fb5db821cc65e70cbd64a3e2a5eaae"},
+]
+click = [
+    {file = "click-7.1.2-py2.py3-none-any.whl", hash = "sha256:dacca89f4bfadd5de3d7489b7c8a566eee0d3676333fbb50030263894c38c0dc"},
+    {file = "click-7.1.2.tar.gz", hash = "sha256:d2b5255c7c6349bc1bd1e59e08cd12acbbd63ce649f2588755783aa94dfb6b1a"},
+]
+ffmpeg-python = [
+    {file = "ffmpeg-python-0.2.0.tar.gz", hash = "sha256:65225db34627c578ef0e11c8b1eb528bb35e024752f6f10b78c011f6f64c4127"},
+    {file = "ffmpeg_python-0.2.0-py3-none-any.whl", hash = "sha256:ac441a0404e053f8b6a1113a77c0f452f1cfc62f6344a769475ffdc0f56c23c5"},
+]
+future = [
+    {file = "future-0.18.2.tar.gz", hash = "sha256:b1bead90b70cf6ec3f0710ae53a525360fa360d306a86583adc6bf83a4db537d"},
+]
+idna = [
+    {file = "idna-2.10-py2.py3-none-any.whl", hash = "sha256:b97d804b1e9b523befed77c48dacec60e6dcb0b5391d57af6a65a312a90648c0"},
+    {file = "idna-2.10.tar.gz", hash = "sha256:b307872f855b18632ce0c21c5e45be78c0ea7ae4c15c828c20788b26921eb3f6"},
+]
+lxml = [
+    {file = "lxml-4.6.1-cp27-cp27m-macosx_10_9_x86_64.whl", hash = "sha256:4b7572145054330c8e324a72d808c8c8fbe12be33368db28c39a255ad5f7fb51"},
+    {file = "lxml-4.6.1-cp27-cp27m-manylinux1_i686.whl", hash = "sha256:302160eb6e9764168e01d8c9ec6becddeb87776e81d3fcb0d97954dd51d48e0a"},
+    {file = "lxml-4.6.1-cp27-cp27m-manylinux1_x86_64.whl", hash = "sha256:d4ad7fd3269281cb471ad6c7bafca372e69789540d16e3755dd717e9e5c9d82f"},
+    {file = "lxml-4.6.1-cp27-cp27m-win32.whl", hash = "sha256:189ad47203e846a7a4951c17694d845b6ade7917c47c64b29b86526eefc3adf5"},
+    {file = "lxml-4.6.1-cp27-cp27m-win_amd64.whl", hash = "sha256:56eff8c6fb7bc4bcca395fdff494c52712b7a57486e4fbde34c31bb9da4c6cc4"},
+    {file = "lxml-4.6.1-cp27-cp27mu-manylinux1_i686.whl", hash = "sha256:23c83112b4dada0b75789d73f949dbb4e8f29a0a3511647024a398ebd023347b"},
+    {file = "lxml-4.6.1-cp27-cp27mu-manylinux1_x86_64.whl", hash = "sha256:0e89f5d422988c65e6936e4ec0fe54d6f73f3128c80eb7ecc3b87f595523607b"},
+    {file = "lxml-4.6.1-cp35-cp35m-manylinux1_i686.whl", hash = "sha256:2358809cc64394617f2719147a58ae26dac9e21bae772b45cfb80baa26bfca5d"},
+    {file = "lxml-4.6.1-cp35-cp35m-manylinux1_x86_64.whl", hash = "sha256:be1ebf9cc25ab5399501c9046a7dcdaa9e911802ed0e12b7d620cd4bbf0518b3"},
+    {file = "lxml-4.6.1-cp35-cp35m-manylinux2014_aarch64.whl", hash = "sha256:4fff34721b628cce9eb4538cf9a73d02e0f3da4f35a515773cce6f5fe413b360"},
+    {file = "lxml-4.6.1-cp35-cp35m-win32.whl", hash = "sha256:475325e037fdf068e0c2140b818518cf6bc4aa72435c407a798b2db9f8e90810"},
+    {file = "lxml-4.6.1-cp35-cp35m-win_amd64.whl", hash = "sha256:f98b6f256be6cec8dd308a8563976ddaff0bdc18b730720f6f4bee927ffe926f"},
+    {file = "lxml-4.6.1-cp36-cp36m-macosx_10_9_x86_64.whl", hash = "sha256:be7c65e34d1b50ab7093b90427cbc488260e4b3a38ef2435d65b62e9fa3d798a"},
+    {file = "lxml-4.6.1-cp36-cp36m-manylinux1_i686.whl", hash = "sha256:d18331ea905a41ae71596502bd4c9a2998902328bbabd29e3d0f5f8569fabad1"},
+    {file = "lxml-4.6.1-cp36-cp36m-manylinux1_x86_64.whl", hash = "sha256:3d9b2b72eb0dbbdb0e276403873ecfae870599c83ba22cadff2db58541e72856"},
+    {file = "lxml-4.6.1-cp36-cp36m-manylinux2014_aarch64.whl", hash = "sha256:d20d32cbb31d731def4b1502294ca2ee99f9249b63bc80e03e67e8f8e126dea8"},
+    {file = "lxml-4.6.1-cp36-cp36m-win32.whl", hash = "sha256:d182eada8ea0de61a45a526aa0ae4bcd222f9673424e65315c35820291ff299c"},
+    {file = "lxml-4.6.1-cp36-cp36m-win_amd64.whl", hash = "sha256:c0dac835c1a22621ffa5e5f999d57359c790c52bbd1c687fe514ae6924f65ef5"},
+    {file = "lxml-4.6.1-cp37-cp37m-macosx_10_9_x86_64.whl", hash = "sha256:d84d741c6e35c9f3e7406cb7c4c2e08474c2a6441d59322a00dcae65aac6315d"},
+    {file = "lxml-4.6.1-cp37-cp37m-manylinux1_i686.whl", hash = "sha256:8862d1c2c020cb7a03b421a9a7b4fe046a208db30994fc8ff68c627a7915987f"},
+    {file = "lxml-4.6.1-cp37-cp37m-manylinux1_x86_64.whl", hash = "sha256:3a7a380bfecc551cfd67d6e8ad9faa91289173bdf12e9cfafbd2bdec0d7b1ec1"},
+    {file = "lxml-4.6.1-cp37-cp37m-manylinux2014_aarch64.whl", hash = "sha256:2d6571c48328be4304aee031d2d5046cbc8aed5740c654575613c5a4f5a11311"},
+    {file = "lxml-4.6.1-cp37-cp37m-win32.whl", hash = "sha256:803a80d72d1f693aa448566be46ffd70882d1ad8fc689a2e22afe63035eb998a"},
+    {file = "lxml-4.6.1-cp37-cp37m-win_amd64.whl", hash = "sha256:24e811118aab6abe3ce23ff0d7d38932329c513f9cef849d3ee88b0f848f2aa9"},
+    {file = "lxml-4.6.1-cp38-cp38-macosx_10_9_x86_64.whl", hash = "sha256:2e311a10f3e85250910a615fe194839a04a0f6bc4e8e5bb5cac221344e3a7891"},
+    {file = "lxml-4.6.1-cp38-cp38-manylinux1_i686.whl", hash = "sha256:a71400b90b3599eb7bf241f947932e18a066907bf84617d80817998cee81e4bf"},
+    {file = "lxml-4.6.1-cp38-cp38-manylinux1_x86_64.whl", hash = "sha256:211b3bcf5da70c2d4b84d09232534ad1d78320762e2c59dedc73bf01cb1fc45b"},
+    {file = "lxml-4.6.1-cp38-cp38-manylinux2014_aarch64.whl", hash = "sha256:e65c221b2115a91035b55a593b6eb94aa1206fa3ab374f47c6dc10d364583ff9"},
+    {file = "lxml-4.6.1-cp38-cp38-win32.whl", hash = "sha256:d6f8c23f65a4bfe4300b85f1f40f6c32569822d08901db3b6454ab785d9117cc"},
+    {file = "lxml-4.6.1-cp38-cp38-win_amd64.whl", hash = "sha256:573b2f5496c7e9f4985de70b9bbb4719ffd293d5565513e04ac20e42e6e5583f"},
+    {file = "lxml-4.6.1-cp39-cp39-manylinux1_i686.whl", hash = "sha256:1d87936cb5801c557f3e981c9c193861264c01209cb3ad0964a16310ca1b3301"},
+    {file = "lxml-4.6.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:2d5896ddf5389560257bbe89317ca7bcb4e54a02b53a3e572e1ce4226512b51b"},
+    {file = "lxml-4.6.1-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:9b06690224258db5cd39a84e993882a6874676f5de582da57f3df3a82ead9174"},
+    {file = "lxml-4.6.1-cp39-cp39-win32.whl", hash = "sha256:bb252f802f91f59767dcc559744e91efa9df532240a502befd874b54571417bd"},
+    {file = "lxml-4.6.1-cp39-cp39-win_amd64.whl", hash = "sha256:7ecaef52fd9b9535ae5f01a1dd2651f6608e4ec9dc136fc4dfe7ebe3c3ddb230"},
+    {file = "lxml-4.6.1.tar.gz", hash = "sha256:c152b2e93b639d1f36ec5a8ca24cde4a8eefb2b6b83668fcd8e83a67badcb367"},
+]
+requests = [
+    {file = "requests-2.24.0-py2.py3-none-any.whl", hash = "sha256:fe75cc94a9443b9246fc7049224f75604b113c36acb93f87b80ed42c44cbb898"},
+    {file = "requests-2.24.0.tar.gz", hash = "sha256:b3559a131db72c33ee969480840fff4bb6dd111de7dd27c8ee1f820f4f00231b"},
+]
+soupsieve = [
+    {file = "soupsieve-2.0.1-py3-none-any.whl", hash = "sha256:1634eea42ab371d3d346309b93df7870a88610f0725d47528be902a0d95ecc55"},
+    {file = "soupsieve-2.0.1.tar.gz", hash = "sha256:a59dc181727e95d25f781f0eb4fd1825ff45590ec8ff49eadfd7f1a537cc0232"},
+]
+urllib3 = [
+    {file = "urllib3-1.25.11-py2.py3-none-any.whl", hash = "sha256:f5321fbe4bf3fefa0efd0bfe7fb14e90909eb62a48ccda331726b4319897dd5e"},
+    {file = "urllib3-1.25.11.tar.gz", hash = "sha256:8d7eaa5a82a1cac232164990f04874c594c9453ec55eef02eab885aa02fc17a2"},
+]

pyproject.toml

@@ -0,0 +1,24 @@
+[tool.poetry]
+name = "bbbdl"
+version = "0.1.0"
+description = "A downloader for BigBlueButton meetings"
+authors = [
+    "g.minoccari <g.minoccari@gmail.com>",
+    "Stefano Pigozzi <ste.pigozzi@gmail.com>"
+]
+license = "AGPL-3.0-or-later"
+
+[tool.poetry.dependencies]
+python = "^3.8"
+click = "^7.1.2"
+beautifulsoup4 = "^4.9.3"
+requests = "^2.24.0"
+ffmpeg-python = "^0.2.0"
+lxml = "^4.6.1"
+
+[tool.poetry.scripts]
+bbbdl = 'bbbdl.__main__:download'
+
+[build-system]
+requires = ["poetry>=0.12"]
+build-backend = "poetry.masonry.api"