Fetching & Recreating

Starting Tasks covers the handle you get back when you start a task. This page covers the two things you do with a task you did not just start — read it, or run it again.

Fetching a task by ID

TypeScript
Python

import { getTask } from '@requence/task'

const task = await getTask('some-task-id')
// or: getTask({ taskId: 'some-task-id', accessToken: '...' })

task.status        // 'SUCCESSFUL' | 'FAILED' | 'IDLE' | 'PENDING' | 'RUNNING'
                   // | 'STOPPED' | 'AWAITING_DELIVERY'
task.statusText    // Human-readable status description, if any
task.taskTemplate  // The task template name
task.name          // Human-readable task name
task.branch        // 'live' or a branch name

task.context.input                 // The task's input
task.context.result                // The task's result (if finished)
task.context.getNodeData(alias)    // Output from a specific node
task.context.getNodeError(alias)   // Error from a specific node

get_task returns a (status, context) tuple, and its statuses are lower-case:

from requence.task import get_task

status, context = get_task("some-task-id")
# or: get_task("some-task-id", access_token="...")

# status: 'successful' | 'failed' | 'idle' | 'pending' | 'running' | 'stopped'
print(status)
print(context["input"])
print(context["result"])
print(context["node_data"].get("my_alias"))
print(context["node_error"].get("my_alias"))

This is a one-shot read, not a subscription. To follow a task as it runs, use Streaming Updates or Watching Tasks.

Recreating a task

Re-run an existing task with the same template and input. Name and priority may be overridden; everything else is taken from the original.

TypeScript
Python

import { recreateTask } from '@requence/task'

const task = await recreateTask({
  taskId: 'some-task-id',
  name: 'My Retry',      // Optional — overrides the original name
  priority: 3,           // Optional — overrides the original priority
})

const result = await task

You get back a task handle, exactly as createTask returns one, so it streams and awaits the same way.