cristhian
6f76074709
validate / validate (push) Successful in 59m15s
A source that documents itself through a specific downstream node depends on that node to be understood, and that node already depends on this one for frames. The README pointed at one by name and the CI workflow still carried a comment about the dataflow that wired it, which no longer exists. The output is documented by its type instead: wire `image` into anything that takes std/media/v1/Image. PREVIEW covers seeing the frames without naming anyone. Generic words like "a viewer" or "a consumer" stay, since they describe a role rather than a repository.
49 lines
1.3 KiB
Python
49 lines
1.3 KiB
Python
"""Packing a frame as std/media/v1/Image.
|
|
|
|
The struct is respelled here rather than imported: node repositories share no
|
|
source package, so the manifest and this schema are the whole contract a
|
|
consumer has.
|
|
"""
|
|
|
|
import numpy as np
|
|
import pyarrow as pa
|
|
|
|
IMAGE_TYPE = pa.struct([
|
|
("width", pa.uint32()),
|
|
("height", pa.uint32()),
|
|
("encoding", pa.utf8()),
|
|
("data", pa.large_binary()),
|
|
])
|
|
|
|
|
|
def pack(frame: np.ndarray, encoding: str, width: int, height: int) -> pa.Array:
|
|
"""Pack encoded frame bytes as a single-row std/media/v1/Image array.
|
|
|
|
Parameters
|
|
----------
|
|
frame : np.ndarray
|
|
The bytes to send. Either a raw pixel buffer or the output of an
|
|
encoder; either way only its bytes are read.
|
|
encoding : str
|
|
Name of the encoding the bytes are in, e.g. ``bgr8`` or ``jpeg``.
|
|
width, height : int
|
|
Dimensions of the *image*, which for a compressed encoding are not the
|
|
shape of ``frame``. They are recorded so a receiver can size a buffer
|
|
without decoding first.
|
|
|
|
Returns
|
|
-------
|
|
pa.Array
|
|
One-row struct array matching ``IMAGE_TYPE``.
|
|
|
|
"""
|
|
return pa.array(
|
|
[{
|
|
"width": width,
|
|
"height": height,
|
|
"encoding": encoding,
|
|
"data": frame.tobytes(),
|
|
}],
|
|
type=IMAGE_TYPE,
|
|
)
|