(Python) client for the ChRIS API

Navigation

Verified details

These details have been verified by PyPI
Maintainers
Avatar for debio from gravatar.com debio Avatar for jbernal from gravatar.com jbernal

Unverified details

These details have not been verified by PyPI
Project links
Meta
  • License: MIT
  • Author: FNNDSC
  • Requires: Python >=3.8
Report project as malware

Project description

A python client for the ChRIS API.

https://travis-ci.org/FNNDSC/python-chrisclient.svg?branch=master

Quick Overview

This repository contains various python scripts and modules that provide a rich client experience for interacting with a CUBE backend instance. Interaction is either via the command line interface (CLI) or through a Python programmatic interface using relevant modules.

Overview

At time of writing (early 2022), three scripts/modules are in production:

  • a general CUBE client

  • a more specific plugin search utility

  • a more specific plugin run schedule utility

Installation

Use the PyPI, Luke!

pip install -U python-chrisclient

ChRIS server preconditions

These preconditions are only necessary to be able to test the client against an actual instance of a ChRIS server both during development and for the automated tests.

Install latest just command runner.

Fire up the full set of ChRIS services:

Open a terminal and run the following commands in any working directory:

$> git clone https://github.com/FNNDSC/ChRIS_ultron_backEnd.git
$> cd ChRIS_ultron_backEnd
$> just

Create a test feed and two pipelines:

Run the following shell script on the directory of this README file. NOTE: The script depends on the popular curl and jq command line tools.

$> ./pre_test.sh

Run the automated tests:

Run the following shell script on the directory of this README file. NOTE: The script depends on the popular curl and jq command line tools.

$> nosetests

Tear down the full set of ChRIS services:

You can later remove all the backend containers and release storage volumes with:

$> cd ChRIS_ultron_backEnd
$> just nuke

General CUBE client usage

Python programmatic interface

Instantiate the client:

from chrisclient import client

cl = client.Client('http://localhost:8000/api/v1/', 'cube', 'cube1234')

Alternatively get a valid token for the user and instantiate the client:

token = client.Client.get_auth_token('http://localhost:8000/api/v1/auth-token/', 'cube', 'cube1234' )
cl = client.Client('http://localhost:8000/api/v1/', token=token)

Upload and create a new plugin (only works for ChRIS admins):

cl_admin = client.Client('http://localhost:8000/api/v1/', 'chris', 'chris1234')
response = cl_admin.admin_upload_plugin('host,moc', '~/simpledsapp.json')

Get plugins given search parameters:

search_params = {'name': 'pl-dircopy'}
response = cl.get_plugins(search_params)

Get a plugin by id:

plugin_id = 1
response = cl.get_plugin_by_id(plugin_id)

Get a plugin’s parameters:

plugin_id = 1
response = cl.get_plugin_parameters(plugin_id, {'limit': 50, 'offset':0})

These retrieving operations are supported for all other high level resources such as feeds, pipelines, plugin instances and workflows.

Get a pipeline’s default parameters and nodes data structure and then run a workflow from the pipeline:

pipeline_id = 2
# attempt to fetch all parameters in a single request by setting a very high limit
response = cl.get_pipeline_default_parameters(pipeline_id, {'limit': 100, 'offset':0})
nodes = cl.compute_workflow_nodes_info(response['data'])
response = cl.create_workflow(pipeline_id, {'previous_plugin_inst_id': 1, 'nodes_info': json.dumps(nodes)})

Please visit the wiki for more information about the client’s API and examples.

Standalone CLI client tool

List plugins:

chrisclient -u cube -p cube1234 http://localhost:8000/api/v1/ list plugin offset==0 limit==2 --verbose

List pipelines:

chrisclient -u cube -p cube1234 http://localhost:8000/api/v1/ list pipeline --verbose

List plugin instances:

chrisclient -u cube -p cube1234 http://localhost:8000/api/v1/ list plugininstance offset==0 limit==1

Upload and create plugin (only works for ChRIS admins):

chrisclient -u chris -p chris1234 http://localhost:8000/api/v1/ add plugin --computenames host,moc --fname ~/simpledsapp.json

Create plugin instance (run plugin):

chrisclient -u cube -p cube1234 http://localhost:8000/api/v1/ add plugininstance
 --pluginid 3 --instancedata '{"previous_id": 1, "dir": "home/cube/uploads"}'

Create pipeline:

chrisclient -u cube -p cube1234 http://localhost:8000/api/v1/ add pipeline --pipelinedata '{"name": "Pipeline1", "plugin_tree": "[{\"plugin_id\": 2, \"previous_index\": null}, {\"plugin_id\": 2, \"previous_index\": 0}]"}'

Create workflow (run pipeline):

chrisclient -u cube -p cube1234 http://localhost:8000/api/v1/ add workflow --pipelineid 2 --workflowdata '{"previous_plugin_inst_id": 1, "nodes_info": "[{\"piping_id\": 3, \"compute_resource_name\": \"host\"}, {\"piping_id\": 4, \"compute_resource_name\": \"host\"}, {\"piping_id\": 5, \"compute_resource_name\": \"host\"}]"}'

Run

Plugins can be run/scheduled on a CUBE instance using the chrispl-run script. The CLI parameters are broadly similar to chrispl-search with some semantic changes more pertinent to the run call – the for search is fixed to the plugin id and the search --pluginSpec becomes the --using CLI.

Run Examples

Run an FS plugin, pl-mri10yr06mo01da_normal

chrispl-run --plugin name=pl-mri10yr06mo01da_normal \
            --onCUBE '{
                "protocol":     "http",
                "port":         "8000",
                "address":      "%HOSTIP",
                "user":         "chris",
                "password":     "chris1234"}'

This plugin does not require any specific CLI args when run in the default state. Once posted to CUBE, a string is returned to the shell:

(name=pl-mri10yr06mo01da_normal) id 14

Indicating that the plugin instance ID of the plugin in CUBE is 14 (for example).

For convenience, let’s set:

CUBE='{
    "protocol":     "http",
    "port":         "8000",
    "address":      "%HOSTIP",
    "user":         "chris",
    "password":     "chris1234"
}'

This return construct lends itself easily to scripting:

ROOTNODE=$(./chrispl-run --plugin name=pl-mri10yr06mo01da_normal --onCUBE "$CUBE" | awk '{print $3}')

or with some formatting:

ROOTNODE=$(
    chrispl-run --plugin name=pl-mri10yr06mo01da_normal     \
                --onCUBE="$CUBE"                            |
                     awk '{print $3}'
)

Run a DS plugin, pl-freesurfer_pp, that builds on the previous node

In this manner, a workflow can be constructed. First construct the arguments for the next plugin:

ARGS="                              \
--ageSpec=10-06-01;                 \
--copySpec=sag,cor,tra,stats,3D;    \
--previous_id=$ROOTNODE             \
"

and now schedule the run:

chrispl-run --plugin name="pl-freesurfer_pp"    \
            --args="$ARGS"                      \
            --onCUBE="$CUBE"

which will return:

(name=pl-freesurfer_pp)        id 19

As before, this can be captured and used for subsequent chaining:

FSNODE=$(
    chrispl-run --plugin name=pl-freesurfer_pp  \
                --args="$ARGS"                  \
                --onCUBE="$CUBE"                |
                     awk '{print $3}'
)

Additional Reading

Consult the ChRIS_docs workflow directory for examples of workflows built using these tools.

-30-

Project details

Verified details

These details have been verified by PyPI
Maintainers
Avatar for debio from gravatar.com debio Avatar for jbernal from gravatar.com jbernal

Unverified details

These details have not been verified by PyPI
Project links
Meta
  • License: MIT
  • Author: FNNDSC
  • Requires: Python >=3.8

Release history Release notifications | RSS feed

This version

2.13.0

2.12.0

2.11.1

2.11.0

2.10.0

2.9.1

2.8.2

2.8.0

2.7.0

2.6.0

2.5.0

2.4.0

2.3.2

2.3.0

2.2.10

2.2.8

2.2.6

2.2.4

2.2.2

2.2.0

2.0.4

2.0.2

2.0.0

1.2.8

1.2.6

1.2.4

1.2.2

1.2.0

1.0.10

1.0.8

1.0.6

1.0.4

1.0.2

1.0.0

0.0.1

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

python_chrisclient-2.13.0.tar.gz (31.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

python_chrisclient-2.13.0-py3-none-any.whl (32.5 kB view details)

Uploaded Python 3

File details

Details for the file python_chrisclient-2.13.0.tar.gz.

File metadata

  • Download URL: python_chrisclient-2.13.0.tar.gz
  • Upload date:
  • Size: 31.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.13.7

File hashes

Hashes for python_chrisclient-2.13.0.tar.gz
Algorithm Hash digest
SHA256 6181212cb07dd93d884088005b7242986120d4df9af2d3df9572dddc8669001e
MD5 d0ac5a1b1097d53d496496782dc2d158
BLAKE2b-256 46cdff11c92b08aa3e8fb16697fa939b8c058a813479dd2c5d1168b42c3a7327

See more details on using hashes here.

File details

Details for the file python_chrisclient-2.13.0-py3-none-any.whl.

File metadata

File hashes

Hashes for python_chrisclient-2.13.0-py3-none-any.whl
Algorithm Hash digest
SHA256 2cd97a07b547b1e4dee6cba5669de33c803f27cc42ffbed44221aca6b9e4299f
MD5 c262d12447146520f7dafd2f23050713
BLAKE2b-256 295c46981b5b6aa6154d8d1d30fec16f858915f6c7fdd80bd4b73b1e8ca0d984

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page