Skip to main content

​
synth_ai.sdk.harbor.uploader

HarborDeploymentUploader - Upload deployments to Harbor API. This module handles the API interactions for creating and managing Harbor deployments. It provides both sync and async interfaces. Example: 
>>> from synth_ai.sdk.harbor import HarborBuildSpec, upload_harbor_deployment
>>>
>>> spec = HarborBuildSpec(
...     name="my-deployment",
...     dockerfile_path="./Dockerfile",
...     context_dir=".",
...     entrypoint="python run.py",
... )
>>>
>>> result = upload_harbor_deployment(spec)
>>> print(f"Created deployment: {result.deployment_id}")

​
Functions

​
upload_harbor_deployment

upload_harbor_deployment(spec: HarborBuildSpec, api_key: str | None = None, backend_url: str | None = None, auto_build: bool = True, wait_for_ready: bool = False, build_timeout_s: float = 600.0) -> HarborDeploymentResult
Upload a Harbor deployment from a build spec. This is the primary user-facing function for uploading deployments. It packages the build context, creates the deployment, and optionally waits for the build to complete. Args:
  • spec: HarborBuildSpec defining the deployment
  • api_key: Synth API key (default: from SYNTH_API_KEY env)
  • backend_url: Synth backend URL (default: from SYNTH_BACKEND_URL env)
  • auto_build: Whether to trigger build immediately (default: True)
  • wait_for_ready: Whether to wait for build completion (default: False)
  • build_timeout_s: Maximum time to wait for build (default: 600s)
Returns:
  • HarborDeploymentResult with deployment details
Raises:
  • HarborAPIError: If API request fails
  • FileNotFoundError: If Dockerfile or context_dir doesn’t exist
  • ValueError: If package exceeds size limit
  • TimeoutError: If wait_for_ready=True and build times out

​
upload_harbor_deployment_async

upload_harbor_deployment_async(spec: HarborBuildSpec, api_key: str | None = None, backend_url: str | None = None, auto_build: bool = True, wait_for_ready: bool = False, build_timeout_s: float = 600.0) -> HarborDeploymentResult
Upload a Harbor deployment asynchronously. Async version of upload_harbor_deployment(). See that function for full documentation.

​
Classes

​
HarborAPIError

Error from Harbor API.

​
HarborDeploymentUploader

Upload and manage Harbor deployments via API. Handles authentication, request formatting, and error handling for Harbor deployment operations. Attributes:
  • backend_url: Synth backend URL
  • api_key: Synth API key for authentication
  • timeout: Request timeout in seconds
Methods:

​
create_deployment_async

create_deployment_async(self, spec: HarborBuildSpec, auto_build: bool = True) -> HarborDeploymentResult
Create a new Harbor deployment asynchronously. Args:
  • spec: HarborBuildSpec defining the deployment
  • auto_build: Whether to trigger build immediately (default: True)
Returns:
  • HarborDeploymentResult with deployment details
Raises:
  • HarborAPIError: If API request fails
  • FileNotFoundError: If Dockerfile or context_dir doesn’t exist
  • ValueError: If package exceeds size limit

​
create_deployment

create_deployment(self, spec: HarborBuildSpec, auto_build: bool = True) -> HarborDeploymentResult
Create a new Harbor deployment synchronously. Args:
  • spec: HarborBuildSpec defining the deployment
  • auto_build: Whether to trigger build immediately (default: True)
Returns:
  • HarborDeploymentResult with deployment details
Raises:
  • HarborAPIError: If API request fails

​
get_deployment_status_async

get_deployment_status_async(self, deployment_id: str) -> dict[str, Any]
Get deployment status asynchronously. Args:
  • deployment_id: UUID of the deployment
Returns:
  • Status dictionary with deployment details and build history

​
get_deployment_status

get_deployment_status(self, deployment_id: str) -> dict[str, Any]
Get deployment status synchronously.

​
wait_for_build_async

wait_for_build_async(self, deployment_id: str, timeout_s: float = 600.0, poll_interval_s: float = 5.0) -> dict[str, Any]
Wait for deployment build to complete. Args:
  • deployment_id: UUID of the deployment
  • timeout_s: Maximum time to wait in seconds
  • poll_interval_s: Interval between status checks
Returns:
  • Final status dictionary
Raises:
  • TimeoutError: If build doesn’t complete within timeout
  • HarborAPIError: If build fails

​
wait_for_build

wait_for_build(self, deployment_id: str, timeout_s: float = 600.0, poll_interval_s: float = 5.0) -> dict[str, Any]
Wait for deployment build to complete synchronously.

​
trigger_build_async

trigger_build_async(self, deployment_id: str) -> dict[str, Any]
Trigger a new build for an existing deployment. Args:
  • deployment_id: UUID of the deployment
Returns:
  • Build information with build_id

​
trigger_build

trigger_build(self, deployment_id: str) -> dict[str, Any]
Trigger a new build synchronously.

​
list_deployments_async

list_deployments_async(self, status: str | None = None, limit: int = 50, offset: int = 0) -> list[dict[str, Any]]
List deployments with optional filtering. Args:
  • status: Filter by status (pending, building, ready, failed)
  • limit: Maximum number of results
  • offset: Pagination offset
Returns:
  • List of deployment dictionaries

​
list_deployments

list_deployments(self, status: str | None = None, limit: int = 50, offset: int = 0) -> list[dict[str, Any]]
List deployments synchronously.