workflow

End-to-end Lambda layer workflow — Build → Package → Upload → Publish.

This module provides a single LayerDeploymentWorkflow class that orchestrates the complete 4-step pipeline for creating and deploying an AWS Lambda layer:

1. Build — install dependencies via a builder (duck-typed, e.g. uv, pip, poetry)

2. Package — create layer.zip from the build artifacts

3. Upload — upload layer.zip to S3 staging

4. Publish — create a versioned Lambda layer in AWS

Usage:

deployment = LayerDeploymentWorkflow(
    builder=my_builder,        # duck type: needs .run() and .path_layout
    path_manifest=Path("uv.lock"),
    s3dir_lambda=S3Path("s3://bucket/lambda/"),
    layer_name="my_layer",
    s3_client=bsm,
    lambda_client=lambda_client,
)
layer_deployment = deployment.run()

class aws_lbd_art_builder_core.layer.workflow.T_BUILDER(*args, **kwargs)

Protocol for Lambda layer builders.

Any builder that has a run() method and a path_layout attribute satisfies this protocol.

class aws_lbd_art_builder_core.layer.workflow.LayerDeploymentWorkflow(verbose: bool = True, printer: ~typing.Callable[[str], None] = <built-in function print>, builder: ~aws_lbd_art_builder_core.layer.workflow.T_BUILDER = REQ, path_manifest: ~pathlib.Path = REQ, s3dir_lambda: S3Path = REQ, layer_name: str = REQ, s3_client: S3Client = REQ, lambda_client: LambdaClient = REQ, ignore_package_list: list[str] | None = None, publish_layer_version_kwargs: dict[str, ~typing.Any] | None = None)

One-stop orchestrator for the complete Lambda layer lifecycle.

This class combines the 4 steps — Build, Package, Upload, Publish — into a single run() call. Each step can also be called individually.

Parameters:

• builder – A builder object that satisfies T_BUILDER protocol (i.e. has a .run() method and a .path_layout attribute). Downstream packages (uv, pip, poetry) each provide their own builder.

• path_manifest – Path to the dependency manifest file used for change detection (e.g. uv.lock, poetry.lock, requirements.txt).

• s3dir_lambda – S3 directory where Lambda artifacts are stored (e.g. S3Path("s3://my-bucket/projects/my_app/lambda/")).

• layer_name – Name of the Lambda layer to create / update (e.g. "my_app").

• s3_client – Boto3 S3 client (or BotoSesManager) for S3 operations.

• lambda_client – Boto3 Lambda client for publishing the layer version.

• ignore_package_list – Packages to exclude from layer.zip. Defaults to default_ignore_package_list (boto3, botocore, setuptools, pip, pytest, etc.).

• publish_layer_version_kwargs – Extra keyword arguments passed to lambda_client.publish_layer_version() (e.g. {"CompatibleRuntimes": ["python3.12"], "Description": "..."}).

• verbose – If True, log progress messages. Inherited from BaseLogger.

• printer – Callable for emitting log messages. Inherited from BaseLogger.

Example:

import aws_lbd_art_builder_uv.api as aws_lbd_art_builder_uv
from boto_session_manager import BotoSesManager
from s3pathlib import S3Path

from aws_lbd_art_builder_core.layer.workflow import LayerDeploymentWorkflow

bsm = BotoSesManager(region_name="us-east-1")

# 1. Create a builder (downstream package provides this)
builder = aws_lbd_art_builder_uv.layer_api.UvLambdaLayerLocalBuilder(
    path_pyproject_toml=Path("pyproject.toml"),
    skip_prompt=True,
)

# 2. Run the full pipeline in one call
workflow = LayerDeploymentWorkflow(
    builder=builder,
    path_manifest=Path("uv.lock"),
    s3dir_lambda=S3Path("s3://my-bucket/projects/my_app/lambda/"),
    layer_name="my_app",
    s3_client=bsm,
    lambda_client=bsm.lambda_client,
    publish_layer_version_kwargs={
        "CompatibleRuntimes": ["python3.12"],
        "Description": "my_app dependencies layer",
    },
)
result = workflow.run()
print(result.layer_version_arn)

run() → LayerDeployment

Execute the full Build → Package → Upload → Publish pipeline.

Returns:

LayerDeployment record with layer version details

step_1_build()

Step 1 — Build: install dependencies using the provided builder.

step_2_package()

Step 2 — Package: create layer.zip from build artifacts.

step_3_upload()

Step 3 — Upload: send layer.zip to S3 staging location.

step_4_publish() → LayerDeployment

Step 4 — Publish: create a new Lambda layer version from the uploaded zip.

Returns:

LayerDeployment record with layer version details