Python bindings for the Ophidia Data Analytics Platform
It is an alternative to Oph_Term, the Ophidia no-GUI interpreter component, and a convenient way to submit SOAP HTTPS requests to an Ophidia server or to develop your own application using Python.
It runs on Python 2.7, 3.3, 3.4 and 3.5, has no Python dependencies and is pure-Python code. It requires a running Ophidia instance for client-server interactions. The latest PyOphidia version (v1.5.0) is compatible with Ophidia v1.2.0.
It provides 2 main modules:
client.py: generic low level class to submit any type of requests (simple tasks and workflows), using SSL and SOAP with the client ophsubmit.py;
cube.py: high level cube-oriented class to interact directly with cubes, with several methods wrapping the operators.
To install PyOphidia package run the following command:
pip install pyophidia
Install with conda
To install PyOphidia with conda run the following command:
conda install -c conda-forge pyophidia
Installation from sources
To install the latest developement version run the following commands:
git clone https://github.com/OphidiaBigData/PyOphidia cd PyOphidia python setup.py install
Import client module from PyOphidia package:
from PyOphidia import client
Instantiate a client
Create a new Client() using the login parameters username, password, host and port. It will also try to resume the last session the user was connected to, as well as the last working directory and the last produced cube.
ophclient = client.Client(username="oph-user",password="oph-passwd",server="127.0.0.1",port="11732")
In case of authentication token is used:
ophclient = client.Client(token="token",server="127.0.0.1",port="11732")
username: Ophidia username
password: Ophidia password
server: Ophidia server address
port: Ophidia server port (default is 11732)
session: ID of the current session
cwd: Current Working Directory
cdd: Current Data Directory
cube: Last produced cube PID
exec_mode: Execution mode, ‘sync’ for synchronous mode (default), ‘async’ for asynchronous mode
ncores: Number of cores for each operation (default is 1)
last_request: Last submitted query
last_response: Last response received from the server (JSON string)
last_jobid: Job ID associated to the last request
last_return_value: Last return value associated to response
last_error: Last error value associated to response
submit(query, display) -> self: Submit a query like ‘operator=myoperator;param1=value1;’ or ‘myoperator param1=value1;’ to the Ophidia server according to all login parameters of the Client and its state.
get_progress(id) -> dict : Get progress of a workflow, either by specifying the id or from the last submitted one.
deserialize_response() -> dict: Return the last_response JSON string attribute as a Python dictionary.
get_base_path(display) -> self : Get base path for data from the Ophidia server.
resume_session(display) -> self: Resume the last session the user was connected to.
resume_cwd(display) -> self: Resume the last cwd (current working directory) the user was located into.
resume_cube(display) -> self: Resume the last cube produced by the user.
wsubmit(workflow, *params) -> self: Submit an entire workflow passing a JSON string or the path of a JSON file and an optional series of parameters that will replace $1, $2 etc. in the workflow. The workflow will be validated against the Ophidia Workflow JSON Schema.
wisvalid(workflow) -> bool: Return True if the workflow (a JSON string or a Python dict) is valid against the Ophidia Workflow JSON Schema or False.
pretty_print(response, response_i) -> self: Prints the last_response JSON string attribute as a formatted response.
To display the command output set “display=True”
Submit a request
Execute the request oph_list level=2:
ophclient.submit("oph_list level=2", display=True)
Set a Client for the Cube class
Instantiate a new Client common to all Cube instances:
from PyOphidia import cube cube.Cube.setclient(username="oph-user",password="oph-passwd",server="127.0.0.1",port="11732")
pid: Cube PID
creation_date: Creation date of the cube
measure: Name of the variable imported into the cube
measure_type: Measure data type
level: Number of operations between the original imported cube and the actual cube
nfragments: Total number of fragments
source_file: Parent of the actual cube
hostxcube: Number of hosts on which the cube is stored
dbmsxhost: Number of DBMS instances on each host
dbxdbms: Number of databases for each DBMS
fragxdb: Number of fragments for each database
rowsxfrag: Number of rows for each fragment
elementsxrow: Number of elements for each row
compressed: If the cube is compressed or not
size: Size of the cube
nelements: Total number of elements
dim_info: List of dict with information on each cube dimension
client: instance of class Client through which it is possible to submit all requests
Create a new container
Create a new container to contain our cubes called test, with 3 double dimensions (lat, lon and time):
Import a new cube
Import the variable T2M from the NetCDF file /path/to/file.nc into a new cube inside the test container. Use lat and lon as explicit dimensions and time as implicit dimension expressed in days:
mycube = cube.Cube(container='test',exp_dim='lat|lon',imp_dim='time',measure='T2M',src_path='/path/to/file.nc',exp_concept_level='c|c',imp_concept_level='d')
Create a Cube object from an existing cube identifier
Instantiate a new Cube using the PID of an existing cube:
mycube2 = cube.Cube(pid='http://127.0.0.1/1/2')
Show a Cube structure and info
To shows metadata information about a data cube, its size and the dimensions related to it:
For the operators such as “cubeschema”, “cubesize”, “cubeelements”, “explore”, “hierarchy”, “info”, “list”, “loggingbk”, “operators”, “search”, “showgrid”, “man”, “metadata”, “primitives”, “provenance”, “search”, “showgrid”, “tasks” and other operators that provide verbose output, the display parameter by default is “True”. For the rest of operators, to display the result, “dispay=True” should be set.
Subset a Cube
To perform a subsetting operation along dimensions of a data cube (dimension values are used as input filters):
mycube3 = mycube2.subset(subset_dims='lat|lon',subset_filter='1:10|20:30',subset_type='coord')
To explore a data cube filtering the data along its dimensions:
Export to NetCDF file
To export data into a single NetCDF file:
Export to Python array
To exports data in a python-friendly format:
data = mycube3.export_array(show_time='yes')
Release history Release notifications | RSS feed
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Hashes for PyOphidia-1.5.0-py2.py3-none-any.whl