parade-manage 0.0.1.4

A manage module of parade

Parade-manage

Parade-manage is moudle for parade.

Note: You should install parade first.

Install

Install is simple:

> pip install parade-manage

Usage

Tasks dag:

 t1  a   b   c
  \ / \ / \ / 
   d   e   f
    \ / \
     g   h

Note: t1 is table name, other are task name Enter your project first

> cd your_project

Initialize the class

from parade_manage import ParadeManage
	
manage = ParadeManage()

If task a is failed, you can get task a's subtask and deps, like

> manage.get_task('a') # get task a
([a, d, e, g, h], {'a': [], 'd': [a], 'e': ['a'], 'g': ['d', 'e'], 'h': [e]})

The first result is tasks, other is deps. And also, you can get mutil failed tasks.

> manage.get_task(['d', 'e']) # get task d and e
(['d', 'e', 'g', 'h'], {'d':[], 'e': [], 'g': ['d', 'e'], 'h': ['e']})

Then, create flow and store flow, generate a yaml file

> manage.store_task_flow('a') # args: 'a' or ['d', 'e']  

Or, remove flow by task name(s)

> manage.rm_task_flow('a')

Run the flow, execute failed tasks

manage.run_taskflow('a')

Sometimes, A task does not depend on other tasks, but uses some tables. Now you can get the tables and tasks.

> manage.get_source('d')
['t1', 'a']  # return table 't1' and task name 'a'

And if table t1 is failed, you can use store_source_flow to store flow

> manage.store_source_flow('t1')
# return a flow related to etl task only