All URIs are relative to http://localhost:8000
| Method | HTTP request | Description |
|---|---|---|
| cancel_workflow | POST /api/v1/workflows/{id}/cancel | |
| create_workflow | POST /api/v1/workflows | |
| get_workflow | GET /api/v1/workflows/{id} | |
| get_workflow_diagram | GET /api/v1/workflows/{id}/diagram | |
| get_workflow_step_states | GET /api/v1/workflows/{id}/steps | |
| list_workflows | GET /api/v1/workflows | |
| retry_workflow | POST /api/v1/workflows/{id}/retry | |
| stream_workflow_events | GET /api/v1/workflows/{id}/events | Stream a workflow's progress as Server-Sent Events: one `status` event whenever the workflow or any step changes, until the workflow is terminal. |
cancel_workflow(id)
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
id = 'id_example' # str | Workflow id
try:
api_instance.cancel_workflow(id)
except Exception as e:
print("Exception when calling WorkflowsApi->cancel_workflow: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| id | str | Workflow id |
void (empty response body)
- Content-Type: Not defined
- Accept: application/json
| Status code | Description | Response headers |
|---|---|---|
| 204 | Cancelled | - |
| 401 | Unauthorized | - |
| 403 | Forbidden | - |
| 404 | Not found | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
CreateWorkflowResponse create_workflow(create_workflow_request)
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.models.create_workflow_request import CreateWorkflowRequest
from queueflow.models.create_workflow_response import CreateWorkflowResponse
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
create_workflow_request = queueflow.CreateWorkflowRequest() # CreateWorkflowRequest |
try:
api_response = api_instance.create_workflow(create_workflow_request)
print("The response of WorkflowsApi->create_workflow:\n")
pprint(api_response)
except Exception as e:
print("Exception when calling WorkflowsApi->create_workflow: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| create_workflow_request | CreateWorkflowRequest |
- Content-Type: application/json
- Accept: application/json
| Status code | Description | Response headers |
|---|---|---|
| 201 | Workflow created | - |
| 400 | Invalid workflow (e.g. dependency cycle) | - |
| 401 | Unauthorized | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
Workflow get_workflow(id)
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.models.workflow import Workflow
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
id = 'id_example' # str | Workflow id
try:
api_response = api_instance.get_workflow(id)
print("The response of WorkflowsApi->get_workflow:\n")
pprint(api_response)
except Exception as e:
print("Exception when calling WorkflowsApi->get_workflow: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| id | str | Workflow id |
- Content-Type: Not defined
- Accept: application/json
| Status code | Description | Response headers |
|---|---|---|
| 200 | Workflow | - |
| 403 | Forbidden | - |
| 404 | Not found | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
WorkflowDiagramResponse get_workflow_diagram(id)
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.models.workflow_diagram_response import WorkflowDiagramResponse
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
id = 'id_example' # str | Workflow id
try:
api_response = api_instance.get_workflow_diagram(id)
print("The response of WorkflowsApi->get_workflow_diagram:\n")
pprint(api_response)
except Exception as e:
print("Exception when calling WorkflowsApi->get_workflow_diagram: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| id | str | Workflow id |
- Content-Type: Not defined
- Accept: application/json
| Status code | Description | Response headers |
|---|---|---|
| 200 | Mermaid diagram of the workflow DAG | - |
| 401 | Unauthorized | - |
| 403 | Forbidden | - |
| 404 | Not found | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
WorkflowStepStatesResponse get_workflow_step_states(id)
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.models.workflow_step_states_response import WorkflowStepStatesResponse
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
id = 'id_example' # str | Workflow id
try:
api_response = api_instance.get_workflow_step_states(id)
print("The response of WorkflowsApi->get_workflow_step_states:\n")
pprint(api_response)
except Exception as e:
print("Exception when calling WorkflowsApi->get_workflow_step_states: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| id | str | Workflow id |
- Content-Type: Not defined
- Accept: application/json
| Status code | Description | Response headers |
|---|---|---|
| 200 | Runtime status of every step, in declaration order. The workflow record carries only step definitions; this is the live progress view. | - |
| 401 | Unauthorized | - |
| 403 | Forbidden | - |
| 404 | Not found | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
ListWorkflowsResponse list_workflows(status=status, queue=queue, task_name=task_name, replayed=replayed, limit=limit, offset=offset, order_by=order_by, include_total=include_total, cursor=cursor, created_after=created_after, created_before=created_before)
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.models.list_workflows_response import ListWorkflowsResponse
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
status = 'status_example' # str | Filter by status (e.g. `pending`, `completed`). (optional)
queue = 'queue_example' # str | Filter by queue name (jobs only). (optional)
task_name = 'task_name_example' # str | Filter by task name (jobs and dead letters). (optional)
replayed = True # bool | Dead letters only: `false` lists entries not yet replayed, `true` the replayed ones; both when omitted. (optional)
limit = 56 # int | Page size, 1..=100 (default 50). (optional)
offset = 56 # int | Number of records to skip (default 0). (optional)
order_by = 'order_by_example' # str | `created_at ASC` or `created_at DESC` (default DESC). (optional)
include_total = True # bool | Include the exact `total` count in the response (default false; the count is an extra full scan over the filtered set). (optional)
cursor = 'cursor_example' # str | Opaque keyset cursor from a previous page's `next_cursor`. When set, `offset` is ignored and listing continues where that page ended. (optional)
created_after = '2013-10-20T19:20:30+01:00' # datetime | Only rows created at or after this instant (RFC 3339, inclusive). With `created_before` this forms the half-open range `[after, before)` — the natural shape for walking history period by period. (optional)
created_before = '2013-10-20T19:20:30+01:00' # datetime | Only rows created strictly before this instant (RFC 3339, exclusive). (optional)
try:
api_response = api_instance.list_workflows(status=status, queue=queue, task_name=task_name, replayed=replayed, limit=limit, offset=offset, order_by=order_by, include_total=include_total, cursor=cursor, created_after=created_after, created_before=created_before)
print("The response of WorkflowsApi->list_workflows:\n")
pprint(api_response)
except Exception as e:
print("Exception when calling WorkflowsApi->list_workflows: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| status | str | Filter by status (e.g. `pending`, `completed`). | [optional] |
| queue | str | Filter by queue name (jobs only). | [optional] |
| task_name | str | Filter by task name (jobs and dead letters). | [optional] |
| replayed | bool | Dead letters only: `false` lists entries not yet replayed, `true` the replayed ones; both when omitted. | [optional] |
| limit | int | Page size, 1..=100 (default 50). | [optional] |
| offset | int | Number of records to skip (default 0). | [optional] |
| order_by | str | `created_at ASC` or `created_at DESC` (default DESC). | [optional] |
| include_total | bool | Include the exact `total` count in the response (default false; the count is an extra full scan over the filtered set). | [optional] |
| cursor | str | Opaque keyset cursor from a previous page's `next_cursor`. When set, `offset` is ignored and listing continues where that page ended. | [optional] |
| created_after | datetime | Only rows created at or after this instant (RFC 3339, inclusive). With `created_before` this forms the half-open range `[after, before)` — the natural shape for walking history period by period. | [optional] |
| created_before | datetime | Only rows created strictly before this instant (RFC 3339, exclusive). | [optional] |
- Content-Type: Not defined
- Accept: application/json
| Status code | Description | Response headers |
|---|---|---|
| 200 | Page of workflows | - |
| 401 | Unauthorized | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
retry_workflow(id)
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
id = 'id_example' # str | Workflow id
try:
api_instance.retry_workflow(id)
except Exception as e:
print("Exception when calling WorkflowsApi->retry_workflow: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| id | str | Workflow id |
void (empty response body)
- Content-Type: Not defined
- Accept: application/json
| Status code | Description | Response headers |
|---|---|---|
| 204 | The workflow is running again: every step that did not complete went back to pending and is re-scheduled; completed steps keep their results. The deadline, if any, is re-armed from the original timeout. | - |
| 401 | Unauthorized | - |
| 403 | Forbidden | - |
| 404 | Not found | - |
| 409 | The workflow is running or completed | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
WorkflowEvent stream_workflow_events(id)
Stream a workflow's progress as Server-Sent Events: one status event whenever the workflow or any step changes, until the workflow is terminal.
- Bearer (API Key) Authentication (bearerAuth):
import queueflow
from queueflow.models.workflow_event import WorkflowEvent
from queueflow.rest import ApiException
from pprint import pprint
# Defining the host is optional and defaults to http://localhost:8000
# See configuration.py for a list of all supported configuration parameters.
configuration = queueflow.Configuration(
host = "http://localhost:8000"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure Bearer authorization (API Key): bearerAuth
configuration = queueflow.Configuration(
access_token = os.environ["BEARER_TOKEN"]
)
# Enter a context with an instance of the API client
with queueflow.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = queueflow.WorkflowsApi(api_client)
id = 'id_example' # str | Workflow id
try:
# Stream a workflow's progress as Server-Sent Events: one `status` event whenever the workflow or any step changes, until the workflow is terminal.
api_response = api_instance.stream_workflow_events(id)
print("The response of WorkflowsApi->stream_workflow_events:\n")
pprint(api_response)
except Exception as e:
print("Exception when calling WorkflowsApi->stream_workflow_events: %s\n" % e)| Name | Type | Description | Notes |
|---|---|---|---|
| id | str | Workflow id |
- Content-Type: Not defined
- Accept: text/event-stream, application/json
| Status code | Description | Response headers |
|---|---|---|
| 200 | SSE stream; each `status` event carries a WorkflowEvent (the workflow status and every step's state). The first event is the current state; the stream closes after the terminal one, or after 15 minutes (reconnect to keep following). | - |
| 401 | Unauthorized | - |
| 403 | Forbidden | - |
| 404 | Not found | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]