pyweb camera support (#1901)

* add media module

* add Device class to media

* add camera test example

* add snap, download and other convenience methods

* load devices automagically

* add draw method to canvas

* add docstring for download

* add docstrings to draw method

* add docstrings to snap

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* load devices as soon as the page loads

* solve conflict

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* remove display calls listing devices in camera example

* fix typos and other small errors

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix typo in docstring

* fix error message typo

* replace setAttribute on JS properties with accessors

* remove debug statement

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* add docstrings

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* add docstrings to camera example

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

pyscript.core/src/stdlib/pyweb/media.py

@@ -0,0 +1,95 @@
+from pyodide.ffi import to_js
+from pyscript import window
+
+
+class Device:
+    """Device represents a media input or output device, such as a microphone,
+    camera, or headset.
+    """
+
+    def __init__(self, device):
+        self._js = device
+
+    @property
+    def id(self):
+        return self._js.deviceId
+
+    @property
+    def group(self):
+        return self._js.groupId
+
+    @property
+    def kind(self):
+        return self._js.kind
+
+    @property
+    def label(self):
+        return self._js.label
+
+    def __getitem__(self, key):
+        return getattr(self, key)
+
+    @classmethod
+    async def load(cls, audio=False, video=True):
+        """Load the device stream."""
+        options = window.Object.new()
+        options.audio = audio
+        if isinstance(video, bool):
+            options.video = video
+        else:
+            # TODO: Think this can be simplified but need to check it on the pyodide side
+
+            # TODO: this is pyodide specific. shouldn't be!
+            options.video = window.Object.new()
+            for k in video:
+                setattr(
+                    options.video,
+                    k,
+                    to_js(video[k], dict_converter=window.Object.fromEntries),
+                )
+
+        stream = await window.navigator.mediaDevices.getUserMedia(options)
+        return stream
+
+    async def get_stream(self):
+        key = self.kind.replace("input", "").replace("output", "")
+        options = {key: {"deviceId": {"exact": self.id}}}
+
+        return await self.load(**options)
+
+
+async def list_devices() -> list[dict]:
+    """
+    Return the list of the currently available media input and output devices,
+    such as microphones, cameras, headsets, and so forth.
+
+    Output:
+
+        list(dict) - list of dictionaries representing the available media devices.
+            Each dictionary has the following keys:
+            * deviceId: a string that is an identifier for the represented device
+                that is persisted across sessions. It is un-guessable by other
+                applications and unique to the origin of the calling application.
+                It is reset when the user clears cookies (for Private Browsing, a
+                different identifier is used that is not persisted across sessions).
+
+            * groupId: a string that is a group identifier. Two devices have the same
+                group identifier if they belong to the same physical device — for
+                example a monitor with both a built-in camera and a microphone.
+
+            * kind: an enumerated value that is either "videoinput", "audioinput"
+                or "audiooutput".
+
+            * label: a string describing this device (for example "External USB
+                Webcam").
+
+    Note: the returned list will omit any devices that are blocked by the document
+    Permission Policy: microphone, camera, speaker-selection (for output devices),
+    and so on. Access to particular non-default devices is also gated by the
+    Permissions API, and the list will omit devices for which the user has not
+    granted explicit permission.
+    """
+    # https://developer.mozilla.org/en-US/docs/Web/API/MediaDevices/enumerateDevices
+    return [
+        Device(obj) for obj in await window.navigator.mediaDevices.enumerateDevices()
+    ]

pyscript.core/src/stdlib/pyweb/pydom.py

@@ -204,6 +204,91 @@ class Element(BaseElement):
     def show_me(self):
         self._js.scrollIntoView()
 
+    def snap(
+        self,
+        to: BaseElement | str = None,
+        width: int | None = None,
+        height: int | None = None,
+    ):
+        """
+        Captures a snapshot of a video element. (Only available for video elements)
+
+        Inputs:
+
+            * to: element where to save the snapshot of the video frame to
+            * width: width of the image
+            * height: height of the image
+
+        Output:
+            (Element) canvas element where the video frame snapshot was drawn into
+        """
+        if self._js.tagName != "VIDEO":
+            raise AttributeError("Snap method is only available for video Elements")
+
+        if to is None:
+            canvas = self.create("canvas")
+            if width is None:
+                width = self._js.width
+            if height is None:
+                height = self._js.height
+            canvas._js.width = width
+            canvas._js.height = height
+
+        elif isistance(to, Element):
+            if to._js.tagName != "CANVAS":
+                raise TypeError("Element to snap to must a canvas.")
+            canvas = to
+        elif getattr(to, "tagName", "") == "CANVAS":
+            canvas = Element(to)
+        elif isinstance(to, str):
+            canvas = pydom[to][0]
+            if canvas._js.tagName != "CANVAS":
+                raise TypeError("Element to snap to must a be canvas.")
+
+        canvas.draw(self, width, height)
+
+        return canvas
+
+    def download(self, filename: str = "snapped.png") -> None:
+        """Download the current element (only available for canvas elements) with the filename
+        provided in input.
+
+        Inputs:
+            * filename (str): name of the file being downloaded
+
+        Output:
+            None
+        """
+        if self._js.tagName != "CANVAS":
+            raise AttributeError(
+                "The download method is only available for canvas Elements"
+            )
+
+        link = self.create("a")
+        link._js.download = filename
+        link._js.href = self._js.toDataURL()
+        link._js.click()
+
+    def draw(self, what, width, height):
+        """Draw `what` on the current element  (only available for canvas elements).
+
+        Inputs:
+
+            * what (canvas image source): An element to draw into the context. The specification permits any canvas
+                image source, specifically, an HTMLImageElement, an SVGImageElement, an HTMLVideoElement,
+                an HTMLCanvasElement, an ImageBitmap, an OffscreenCanvas, or a VideoFrame.
+        """
+        if self._js.tagName != "CANVAS":
+            raise AttributeError(
+                "The draw method is only available for canvas Elements"
+            )
+
+        if isinstance(what, Element):
+            what = what._js
+
+        # https://developer.mozilla.org/en-US/docs/Web/API/CanvasRenderingContext2D/drawImage
+        self._js.getContext("2d").drawImage(what, 0, 0, width, height)
+
 
 class OptionsProxy:
     """This class represents the options of a select element. It

pyscript.core/test/camera.html

@@ -0,0 +1,24 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  <title>PyScript Media Example</title>
+  <link rel="stylesheet" href="../dist/core.css">
+  <script type="module" src="../dist/core.js"></script>
+</head>
+<body>
+  <script type="py" src="camera.py" async></script>
+
+  <label for="cars">Choose a device:</label>
+
+  <select name="devices" id="devices"></select>
+
+  <button id="pick-device">Select the device</button>
+  <button id="snap">Snap</button>
+
+  <div id="result"></div>
+
+  <video id="video" width="600" height="400" autoplay></video>
+</body>
+</html>

pyscript.core/test/camera.py

@@ -0,0 +1,32 @@
+from pyodide.ffi import create_proxy
+from pyscript import display, document, when, window
+from pyweb import media, pydom
+
+devicesSelect = pydom["#devices"][0]
+video = pydom["video"][0]
+devices = {}
+
+
+async def list_media_devices(event=None):
+    """List the available media devices."""
+    global devices
+    for i, device in enumerate(await media.list_devices()):
+        devices[device.id] = device
+        label = f"{i} - ({device.kind}) {device.label} [{device.id}]"
+        devicesSelect.options.add(value=device.id, html=label)
+
+
+@when("click", "#pick-device")
+async def connect_to_device(e):
+    """Connect to the selected device."""
+    device = devices[devicesSelect.value]
+    video._js.srcObject = await device.get_stream()
+
+
+@when("click", "#snap")
+async def camera_click(e):
+    """Take a picture and download it."""
+    video.snap().download()
+
+
+await list_media_devices()

pyscript.core/types/stdlib/pyscript.d.ts

@@ -7,6 +7,7 @@ declare namespace _default {
         "util.py": string;
     };
     let pyweb: {
+        "media.py": string;
         "pydom.py": string;
     };
 }