pyflink.multimodal.operators.image_encode#

image_encode(*columns, format=None, quality=85, output='bytes', concurrency=None)[source]#

Encode an image into compressed bytes or a base64 data URL.

Parameters

*columns – Optional image columns. When provided, the UDF is applied directly instead of returning a factory.
format – Target image format, e.g. "JPEG", "PNG", "WEBP", "BMP". None (default) keeps the source format for encoded byte inputs; decoded image values default to "JPEG" because decoded images do not retain original container metadata.
quality – Compression quality (1-100). Only affects lossy formats such as JPEG and WebP. Default 85.
output – Output representation. "bytes" (default) preserves the existing binary contract. "data_url" returns a string such as data:image/jpeg;base64,... for AI functions that accept image URLs or inline data URLs.
concurrency – UDF concurrency. None uses the framework default.

Returns

A UDF producing encoded image bytes or a data URL string.

Example::

>>> # As a reusable variable
>>> encode = image_encode(format="JPEG")
>>> df = df.with_columns("jpg_bytes", encode(col("img")))
>>>
>>> # Inline
>>> df = df.with_columns("jpg_bytes", image_encode(col("img"), format="JPEG"))