threaded-order 1.5.3

A lightweight framework for running functions concurrently across multiple threads while maintaining a defined execution order.

threaded-order

threaded-order is a lightweight Python framework for running functions in parallel while honoring explicit dependency order. You declare dependencies; the scheduler handles sequencing, concurrency, and correctness.

Great for dependency-aware test runs, build steps, pipelines, and automation flows that need structure without giving up speed.

Why threaded-order?

Use it when you want:

• Parallel execution with strict upstream → downstream ordering
• A simple, declarative way to express dependencies (after=['a', 'b'])
• Deterministic behavior even under concurrency
• A DAG-driven execution model without heavyweight tooling
• A clean decorator-based API for organizing tasks
• A CLI (tdrun) for running functions as parallel tests

Key Features

• Parallel execution using Python threads backed by a dependency DAG
• Deterministic ordering based on after=[...] relationships
• Decorator-based API (@dmark, @dregister) for clean task definitions
• Shared state (opt-in) with a thread-safe, built-in lock
• Thread-safe logging via ThreadProxyLogger
• Graceful interrupt handling and clear run summaries
• CLI: tdrun — dependency-aware test runner with tag filtering
• DAG visualization — inspect your dependency graph with --graph
• Simple, extensible design — no external dependencies

About the DAG

threaded-order schedules work using a Directed Acyclic Graph (DAG) — this structure defines which tasks must run before others.
If you’re new to DAGs or want a quick refresher, this short primer is helpful: https://en.wikipedia.org/wiki/Directed_acyclic_graph

Installation

pip install threaded-order

API Overview

class Scheduler(
    workers=None,                 # max number of worker threads
    state=None,                   # shared state dict passed to @dmark functions
    store_results=True,           # save return values into state["results"]
    clear_results_on_start=True,  # wipe previous results
    setup_logging=False,          # enable built-in logging config
    add_stream_handler=True,      # attach stream handler to logger
    verbose=False                 # enable extra debug logging
)

Runs registered callables across multiple threads while respecting declared dependencies.

Core Methods

Method	Description
register(obj, name, after=None, with_state=False)	Register a callable for execution. after defines dependencies by name, specify if function is to receive the shared state.
dregister(after=None, with_state=False)	Decorator variant of register() for inline task definitions.
start()	Start execution, respecting dependencies. Returns a summary dictionary.
dmark(after=None, with_state=False, tags=None)	Decorator that marks a function for deferred registration by the scheduler, allowing you to declare dependencies (after) and whether the function should receive the shared state (with_state), and optionally add tags to the function (tags) for execution filtering.

Callbacks

All are optional and run on the scheduler thread (never worker threads).

Callback	When Fired	Signature
on_task_start(fn)	Before a task starts	(name)
on_task_run(fn)	When tasks starts running on a thread	(name, thread)
on_task_done(fn)	After a task finishes	(name, ok)
on_scheduler_start(fn)	Before scheduler starts running tasks	(meta)
on_scheduler_done(fn)	After all tasks complete	(summary)

Shared state and _state_lock

If with_state=True, tasks receive the shared state dict. Threaded-order inserts a re-entrant lock at state['_state_lock'] you can use when modifying shared values.

For more information refer to Shared State Guidelines

Interrupt Handling

Press Ctrl-C during execution to gracefully cancel outstanding work:

• Running tasks finish naturally or are marked as cancelled
• Remaining queued tasks are discarded
• Final summary reflects all results

CLI Overview (tdrun)

tdrun is a DAG-aware, parallel test runner built on top of the threaded-order scheduler.

It discovers @dmark functions inside a module, builds a dependency graph, and executes everything in parallel while preserving deterministic order.

You get:

• Parallel execution based on the Scheduler
• Predictable, DAG-driven ordering
• Tag filtering (--tags=tag1,tag2)
• Arbitrary state injection via --key=value
• Mock upstream results for single-function runs
• Graph inspection (--graph) to validate ordering and parallelism
• Clean pass/fail summary

CLI usage

usage: tdrun [-h] [--workers WORKERS] [--tags TAGS] [--log] [--verbose] [--graph] target

A threaded-order CLI for dependency-aware, parallel function execution.

positional arguments:
  target             Python file containing @dmark functions

options:
  -h, --help         show this help message and exit
  --workers WORKERS  Number of worker threads (default: Scheduler default)
  --tags TAGS        Comma-separated list of tags to filter functions by
  --log              enable logging output
  --verbose          enable verbose logging output
  --graph            show dependency graph and exit

Run all marked functions in a module:

tdrun path/to/module.py

Run a single function:

tdrun module.py::test_b

This isolates the function and ignores its upstream dependencies.

You can provide mocked results:

tdrun module.py::test_b --result-test_a=mock_value

Inject arbitrary state parameters

tdrun module.py --env=dev --region=us-west

These appear in initial_state and can be processed in your module’s setup_state(state).

This allows your module to compute initial state based on CLI parameters.

DAG Inspection

Use graph-only mode to inspect dependency structure:

tdrun examples/example4c.py --graph

Example output:

Graph: 6 nodes, 6 edges
Roots: [0]
Leaves: [4], [5]
Levels: 4

Nodes:
  [0] test_a
  [1] test_b
  [2] test_c
  [3] test_d
  [4] test_e
  [5] test_f

Edges:
  [0] -> [1], [2]
  [1] -> [5]
  [2] -> [3], [4]
  [3] -> [5]
  [4] -> (none)
  [5] -> (none)

Stats:
  Longest chain length (edges): 3
  Longest chains:
    test_a -> test_c -> test_d -> test_f
  High fan-in nodes (many dependencies):
    test_f (indegree=2)
  High fan-out nodes (many dependents):
    test_a (children=2)
    test_c (children=2)

Examples

See the examples/ folder for runnable demos.

Basic usage Example

Code

from threaded_order import Scheduler, ThreadProxyLogger
import time
import random

s = Scheduler(workers=3, setup_logging=True)
logger = ThreadProxyLogger()

def run(name):
    time.sleep(random.uniform(.5, 3.5))
    logger.info(f'{name} completed')

@s.dregister()
def a(): run('a')

@s.dregister(after=['a'])
def b(): run('b')

@s.dregister(after=['a'])
def c(): run('c')

@s.dregister(after=['c'])
def d(): run('d')

@s.dregister(after=['c'])
def e(): run('e')

@s.dregister(after=['b', 'd'])
def f(): run('f')

if __name__ == '__main__':
    s.on_scheduler_done(lambda s: print(f"Passed:{len(s['passed'])} Failed:{len(s['failed'])}"))
    s.start()

Shared State Example

Code

import json
from time import sleep
from threaded_order import Scheduler

s = Scheduler(workers=3, state={})

@s.dregister(with_state=True)
def load(state):
    with state['_state_lock']:
        state['counter'] = state.get('counter', 0) + 1
    state["x"] = 10; return "loaded"

@s.dregister(with_state=True)
def behave(state):
    with state['_state_lock']:
        state['counter'] = state.get('counter', 0) + 1
    sleep(3); return "behaved"

@s.dregister(after=["load"], with_state=True)
def compute(state):
    with state['_state_lock']:
        state['counter'] = state.get('counter', 0) + 1
    state["x"] += 5; return state["x"]

s.start()
print(json.dumps(s.state, indent=2, default=str))

{
  "_state_lock": "<unlocked _thread.RLock object owner=0 count=0 at 0x7ac9632852c0>",
  "results": {
    "load": "loaded",
    "compute": 15,
    "behave": "behaved"
  },
  "counter": 3,
  "x": 15
}

tdrun Example

Code

import time
import random
from faker import Faker
from threaded_order import dmark, ThreadProxyLogger

logger = ThreadProxyLogger()

def setup_state(state):
    state.update({
        'faker': Faker()
    })

def run(name, state, deps=None, fail=False):
    with state['_state_lock']:
        last_name = state['faker'].last_name()
    sleep = random.uniform(.5, 3.5)
    logger.debug(f'{name} \"{last_name}\" running - sleeping {sleep:.2f}s')
    time.sleep(sleep)
    if fail:
        assert False, 'Intentional Failure'
    else:
        results = []
        for dep in (deps or []):
            dep_result = state['results'].get(dep, '--no-result--')
            results.append(f'{name}.{dep_result}')
        if not results:
            results.append(name)
        logger.info(f'{name} passed')
        return '|'.join(results)

@dmark(with_state=True)
def test_a(state): return run('test_a', state)

@dmark(with_state=True, after=['test_a'])
def test_b(state): return run('test_b', state, deps=['test_a'])

@dmark(with_state=True, after=['test_a'])
def test_c(state): return run('test_c', state, deps=['test_a'])

@dmark(with_state=True, after=['test_c'])
def test_d(state): return run('test_d', state, deps=['test_c'], fail=True)
    
@dmark(with_state=True, after=['test_c'])
def test_e(state): return run('test_e', state, deps=['test_c'])

@dmark(with_state=True, after=['test_b', 'test_d'])
def test_f(state): return run('test_f', state, deps=['test_b', 'test_d'])

tdrun tag filtering Example

Code

import time
import random
from faker import Faker
from threaded_order import dmark, ThreadProxyLogger

logger = ThreadProxyLogger()

def setup_state(state):
    state.update({
        'faker': Faker()
    })

def run(name, state, deps=None, fail=False):
    with state['_state_lock']:
        last_name = state['faker'].last_name()
    sleep = random.uniform(.5, 3.5)
    logger.debug(f'{name} \"{last_name}\" running - sleeping {sleep:.2f}s')
    time.sleep(sleep)
    if fail:
        assert False, 'Intentional Failure'
    else:
        results = []
        for dep in (deps or []):
            dep_result = state.get(dep, '--no-result--')
            results.append(f'{name}.{dep_result}')
        if not results:
            results.append(name)
        logger.info(f'{name} passed')
        state[name] = '|'.join(results)

@dmark(with_state=True, tags='layer1')
def test_a(state): return run('test_a', state)

@dmark(with_state=True, after=['test_a'], tags='layer2')
def test_b(state): return run('test_b', state, deps=['test_a'])

@dmark(with_state=True, after=['test_a'], tags='layer2')
def test_c(state): return run('test_c', state, deps=['test_a'])

@dmark(with_state=True, after=['test_c'], tags='layer3')
def test_d(state): return run('test_d', state, deps=['test_c'], fail=True)
    
@dmark(with_state=True, after=['test_c'], tags='layer3')
def test_e(state): return run('test_e', state, deps=['test_c'])

@dmark(with_state=True, after=['test_b', 'test_d'], tags='layer4')
def test_f(state): return run('test_f', state, deps=['test_b', 'test_d'])

ProgressBar Integration Example

Can be done by using the on_task_done callback. See example5

Development

Clone the repository and ensure the latest version of Docker is installed on your development server.

Build the Docker image:

docker image build \
-t threaded-order:latest .

Run the Docker container:

docker container run \
--rm \
-it \
-v $PWD:/code \
threaded-order:latest \
bash

Execute the dev pipeline:

make dev