卡布图皮 1.0.0

示例用法

import time
from kabutopy.client import Client

# make a client instance with the base url of the kabuto service
client = Client("https://:5000")

# registering a user
client.register("user", "password", "email@email.com")

# loging in with a user, the client keeps track of the cookies
client.login("user", "password")

# creating an image
# name: string
# dockerfile: raw string / open file / path to a file
# repo_url: string
# nocache: boolean
image = client.create_image("my_image", dockerfile="some file", repo_url=None, nocache=False)

# requesting the image status and image id will poll the server
# if the id is still empty, or is the status is PENDING
# it will update id, state and output if the state is SUCCESS
# it will update state, error and output if the state is FAIL

while image.status == 'PENDING':
    sleep(1)

# creating a pipeline
# name: string
pipeline = client.create_pipeline("my_pipeline")

# creating a job
# command: string
# image: Image / int / string that is castable to int
# attachment: open file / path to a file (a list of these things also works when multiple files need to be uploaded)
# pipeline: Pipeline / int / string that is castable to int
job = client.create_job(command, image, attachments, pipeline)

# getting logs
# job: Job / int / string that is castable to int
# job_id: int / string that is castable to int
# you are able to get the logs trough the client
logs = client.get_job_logs(job)

# trough the job itself
# job_id: int / string that is castable to int
logs = job.get_logs()

# getting results
# job: Job / int / string that is castable to int
# trough the client
# will return None if the job has not yet started, is still running or has failed
result = client.get_job_results(job)

# trough the job itself
result = job.get_results()

类和方法