python - Parallel execution using Canvas Workflow -

i've been working on project requires execution of tasks in parallel way as possible. code below exemplifies problem:

# -*- coding: utf-8 -*- __future__ import absolute_import celery.app import celery celery.utils.log import get_task_logger  logger = get_task_logger(__name__)  app = celery('celeryproject', backend="redis://localhost:6379/0", broker="amqp://guest:guest@localhost:5672//")   @app.task def step1(x):     # list of str webservice using x     return ["step1-%s-%s" % (x, i) in range(5)]   @app.task def step2(y):     # list of str webservice using x     return ["step2-%s-%s" % (x, i) in range(5, 10)]   @app.task def step3(z):     # process z     logger.info(z)  @app.task def pipeline(xs):     x in xs:         r1_list = step1.delay(x).get()         r1 in r1_list:             r2_list = step2.delay(x, r1).get()             r2 in r2_list:                 step3.delay(r2) 

i need execute task.pipeline start process. tried canvas workflow examples (http://docs.celeryproject.org/en/latest/userguide/canvas.html) got problems along way. you, guys, me?

edit: step1 returns list , want deal each element of list inside step2. this, returns list , want deal each element of list inside step3.

i apply step2 each element returned step1 , apply step3 each element returned map before.

i tried, example: chord(step1.s('xxx'))(step2.map())

but got error:

typeerror('map() takes 2 arguments (1 given)',) 

also tried use groups didn't work.