Skip to main content

Python library for EPICS PV Access

Project description

PvaPy - PvAccess for Python

The PvaPy package is a Python API for EPICS4. It supports both PVA and CA providers, all standard EPICS4 types (structures, scalars, unions, etc), standard set of channel operations (put/get, monitor), RPC client/server, PVA server, etc. It uses Boost/Python to wrap EPICS4 C++ libraries and APIs, and is currently supported on the Unix-like operating systems (Linux, OSX).

Installation

At the moment prebuilt PvaPy packages exist for the Conda and PyPI package management system. If none of the prebuilt packages work for you, or if you need 32-bit Linux packages, you will have to build PvaPy from sources.

Conda

  • Supported platforms: Linux (64-bit), OSX
  • Supported Python versions: Python2, Python3

Make sure conda command is in your path, and execute the following command:

  $ conda install -c epics pvapy

This should result in installation of pvapy, pvapy-boost and epics-base Conda packages.

PyPI

  • Supported platforms: Linux (64-bit)
  • Supported Python versions: Python2, Python3

Make sure pip command is in your path, and execute the following command:

  $ pip install pvapy

PvaPy pip packages contain all needed epics base and boost libraries.

Build

This section is intended for those users that would like to build PvaPy from the source code.

Prerequisites

Building PvaPy from source requires recent versions of the following software:

  1. Standard development tools (gcc, make, autoconf, etc.)
  2. Python development header files/libraries (>=v2.6.6)
  3. Sphinx/Sphinx RTD Theme (optional, for generating documentation)

There are two ways of building and installing PvaPy from sources: automated and custom. The automated local install will download and build predefined versions of EPICS Base, Boost and PvaPy. Custom local install relies on already existing EPICS Base and Boost libraries, and builds only PvaPy.

Automated Build

Make sure the desired version of Python is in your path, and execute the following command from the top of the PvaPy source directory:

  $ PREFIX=/local/install/path make local

This should build and install predefined versions of EPICS Base, Boost Python and PvaPy libraries under the directory specified by the PREFIX variable. If PREFIX is not specified, all software will be installed under the tools/local subdirectory of the PvaPy distribution.

Custom Build

In addition to the Python-related prerequsites, the custom build requires the following software already built:

  1. EPICS Base (>=v3.14.12.3)
  2. EPICS4 CPP release (>=v4.4.0); note that EPICS7 release includes all needed v4 modules
  3. Boost (>=v1.41.0); must have the boost_python library built
  4. Boost.NumPy (optional, for retrieving numeric scalar arrays as numpy arrays; this is not needed for Boost v1.63.0 and later)

The version numbers mentioned above are the earliest that PvaPy was tested with. Any recent version of Python and the Boost libraries (such as those that come with current linux distributions) should work. Similarly, any recent version of EPICS Base that is supported by EPICS4 CPP should be fine.

Nothing special needs to be done when building the EPICS4 CPP modules. Ensure that the EPICS Base installation you use for this module is the same one that was used to build the EPICS4 modules.

This module has not been adapted for use on Microsoft Windows. Only Unix-like operating systems (e.g. Linux, MacOS, Solaris) are currently supported.

1. Configure PvaPy

This can be done manually, or using autoconf.

For manual configuration: Read the comments in both the configure/RELEASE and configure/CONFIG_SITE files and follow the instructions given there.

For automatic configuration: In the top level directory run

  $ make configure EPICS_BASE=/epics/base/path [EPICS4_DIR=/epics/v4/path] [BOOST_NUM_PY_DIR=/boost.numpy/path] [BOOST_ROOT=/boost/path] [PYTHON_VERSION=3]

Note that you can only use the automatic configuration if the v4 modules have not been renamed. In the above command replace /epics/base/path with the full path to your EPICS Base directory, and /epics/v4/path with the full path to your top level directory containing the v4 modules pvDataCPP, pvAccessCPP, etc. If you are using EPICS7 release, you can omit EPICS4_DIR argument, as configuration scripts will find the required v4 libraries and header files in the EPICS Base directory.

The optional BOOST_NUM_PY_DIR argument enables NumPy array support for older Boost versions, as v1.63.0 and later releases already include NumPy libraries. The optional BOOST_ROOT argument is typically used for custom Boost installation directory. In case you are using custom python installation that uses shared object libraries, you may need to set PATH and LD_LIBRARY_PATH environment variables before running the above make configure command. Also, note that building with python 3.x requires PYTHON_VERSION=3 argument.

The make configure command will check for your Boost/Python libraries, and create suitable configure/RELEASE.local and configure/CONFIG_SITE.local files that are used by the build process. They should look roughly like the examples below:

  $ cat RELEASE.local
  PVACLIENT = /home/epics/v4/pvaClientCPP
  PVDATABASE = /home/epics/v4/pvDatabaseCPP
  NORMATIVETYPES = /home/epics/v4/normativeTypesCPP
  PVACCESS = /home/epics/v4/pvAccessCPP
  PVDATA = /home/epics/v4/pvDataCPP
  EPICS_BASE = /home/epics/base-3.15.5

  $ cat CONFIG_SITE.local
  PVA_PY_CPPFLAGS = -I/usr/include -I/usr/include/python2.7
  PVA_PY_LDFLAGS = -L/usr/lib64 -lpython2.7
  PVA_PY_SYS_LIBS = boost_python  
  PVA_API_VERSION = 480
  PVA_RPC_API_VERSION = 480
  HAVE_BOOST_NUM_PY = 0
  HAVE_BOOST_PYTHON_NUM_PY = 0
  PYTHON_VERSION = 2.7
  PVA_PY_PYTHON = /bin/python
  PVA_PY_PYTHONPATH = /home/epics/v4/pvaPy/lib/python/2.7/linux-x86_64
  PVA_PY_LD_LIBRARY_PATH = /usr/lib64
  PVA_PY_EPICS_BASE = /home/epics/base-3.15.5
  PVA_PY_EPICS4_DIR = /home/epics/v4
  PVA_PY_HOST_ARCH = linux-x86_64
  PVA_PY_SETUP_SH = /home/epics/pvaPy/bin/linux-x86_64/pvapy_setup_full.2.7.sh

The above files were created automatically on a 64-bit RHEL 7.4 machine, with the following boost/python packages installed:

  $ rpm -q boost-python python-devel
  boost-python-1.53.0-27.el7.x86_64
  python-devel-2.7.5-58.el7.x86_64

Note that the automatic configuration process also creates pvapy_setup_full.$PYTHON_VERSION.(c)sh and pvapy_setup_pythonpath.$PYTHON_VERSION.(c)sh files in the bin/$EPICS_HOST_ARCH directory. The full setup files modify PATH, LD_LIBRARY_PATH and PYTHONPATH environment variables, while the pythonpath setup files modify only PYTHONPATH variable. For example,

  $ cat pvapy_setup_pythonpath.2.7.sh
  #!/bin/sh
  #
  # modifies PYTHONPATH environment variable
  #
  if test -z "$PYTHONPATH" ; then
      export PYTHONPATH=/home/epics/v4/pvaPy/lib/python/2.7/linux-x86_64
  else
      export PYTHONPATH=/home/epics/v4/pvaPy/lib/python/2.7/linux-x86_64:$PYTHONPATH
  fi

After building pvaPy, the environment setup files can be sourced to use the built python module, e.g.:

  $ . /home/epics/v4/pvaPy/bin/linux-x86_64/pvapy_setup_pythonpath.2.7.sh
  $ echo $PYTHONPATH
  /home/epics/v4/pvaPy/lib/python/2.7/linux-x86_64

or for csh users:

  % source /home/epics/v4/pvaPy/bin/linux-x86_64/pvapy_setup_pythonpath.2.7.csh
  % echo $PYTHONPATH
  /home/epics/v4/pvaPy/lib/python/2.7/linux-x86_64

2. Compile PvaPy Source

In the top level package directory run:

  $ make

The above command will create and install a loadable library pvaccess.so under the lib/python directory which can be imported directly by Python.

3. Generate Documentation

This step is optional and requires Sphinx to be installed:

  $ make doc

If a sphinx-build script is present on the system, html pages will be generated in the documentation/sphinx/_build/html directory.

Basic Usage: PV put/get

For simple testing, do the following:

  1. In a separate terminal, start the testDbPv IOC:
  $ cd $EPICS4_DIR/pvaSrv/testTop/iocBoot/testDbPv
  $ ../../bin/$EPICS_HOST_ARCH/testDbPv st.cmd
  1. Source the appropriate setup file from pvaPy's bin/$EPICS_HOST_ARCH directory and start python (the Python PVA module is called pvaccess):
  $ python
  >>> import pvaccess
  >>> dir (pvaccess)
  ['BOOLEAN', 'BYTE', 'CA', 'Channel', 'DOUBLE', 'FLOAT', 'FieldNotFound',
  'INT', 'InvalidArgument', 'InvalidDataType', 'InvalidRequest', 'LONG',
  'NtTable', 'NtType', 'PVA', 'ProviderType', 'PvAlarm', 'PvBoolean', 'PvByte',
  'PvDouble', 'PvFloat', 'PvInt', 'PvLong', 'PvObject', 'PvScalar',
  'PvScalarArray', 'PvShort', 'PvString', 'PvTimeStamp', 'PvType', 'PvUByte',
  'PvUInt', 'PvULong', 'PvUShort', 'PvUnion', 'PvaException', 'RpcClient',
  'RpcServer', 'SHORT', 'STRING', 'UBYTE', 'UINT', 'ULONG', 'USHORT', '__doc__',
  '__file__', '__name__', '__package__']
  >>> c = pvaccess.Channel('int01')
  >>> print c.get()
  epics:nt/NTScalar:1.0
      int value 0
  >>> c.putInt(7)
  >>> print c.get()
  epics:nt/NTScalar:1.0
      int value 7
  >>> c.put(pvaccess.PvInt(5))
  >>> print c.get()
  epics:nt/NTScalar:1.0
      int value 5

In the above, note that in addition to PV object classes like PvInt, one can also use standard Python types as arguments for channel puts.

Basic Usage: PV monitor

  1. In a separate terminal, start the testDbPv IOC:
  $ cd $EPICS4_DIR/pvaSrv/testTop/iocBoot/testDbPv
  $ ../../bin/$EPICS_HOST_ARCH/testDbPv st.cmd
  1. PV values can be changed using the IOC shell command dbpf, e.g:
  epics> dbpr 'float01'
  ASG:                DESC:               DISA: 0             DISP: 0
  DISV: 1             NAME: float01       SEVR: MAJOR         STAT: LOLO
  TPRO: 0             VAL: 0
  epics> dbpf 'float01' 11.1
  DBR_FLOAT:          11.1
  1. Monitor a channel in Python, passing in a subscriber object (function that processes PvObject instance):
  >>> c = pvaccess.Channel('float01')
  >>> def echo(x):
  ...     print 'New PV value:', x
  ...
  >>> c.subscribe('echo', echo)
  >>> c.startMonitor()
  >>> New PV value: epics:nt/NTScalar:1.0
      float value 11.1

  New PV value: epics:nt/NTScalar:1.0
      float value 11.2

  New PV value: epics:nt/NTScalar:1.0
      float value 11.3

  >>> c.stopMonitor()

Advanced Usage: PVA Server Class

  1. In terminal 1, create a simple 'pair' channel:
  $ python 
  >>> pv = PvObject({'x': INT, 'y' : INT})
  >>> pvaServer = PvaServer('pair', pv)
  1. In terminal 2, start monitoring this channel:
  $ pvget -m pair
  1. In terminal 1, update one of the object's fields:
  >>> pv['x'] = 1

This change should appear in terminal 2.

Advanced Usage: RPC Client Class

  1. In a separate terminal, start the v4 test RPC service:
  $ cd $EPICS4_DIR/pvAccessCPP/bin/$EPICS_HOST_ARCH
  $ ./rpcServiceExample # in terminal 2
  1. RPC test channel is 'sum':
  >>> rpc = pvaccess.RpcClient('sum')
  >>> request = pvaccess.PvObject({'a': pvaccess.STRING, 'b': pvaccess.STRING})
  >>> request.set({'a': '11', 'b': '22' })
  >>> print request
  structure
      string a 11
      string b 22
  >>> response = rpc.invoke(request)
  >>> print response
  structure
      double c 33

Advanced Usage: RPC Server Class

Example 1

  1. In a separate terminal, source the environment file and start python:
  $ python # in terminal 2
  >>> import pvaccess
  >>> srv = pvaccess.RpcServer()
  >>> def echo(x):    # x is an instance of PvObject
  ...     print 'Got object: ', x
  ...     return x    # service must return an instance of PvObject
  >>> srv.registerService('echo', echo)
  >>> srv.listen()
  1. In terminal 1, reuse previous request object
  >>> rpc = pvaccess.RpcClient('echo')
  >>> response = rpc.invoke(request)
  >>> print response
  structure
      string a 11
      string b 22

Example 2

  1. In terminal 2:
  $ python
  >>> import pvaccess
  >>> srv = pvaccess.RpcServer()
  >>> def sum(x):
  ...     a = x.getInt('a')
  ...     b = x.getInt('b')
  ...     return pvaccess.PvInt(a+b)
  >>> srv.registerService('sum', sum)
  >>> srv.listen()
  1. In terminal 1:
  >>> rpc = pvaccess.RpcClient('sum')
  >>> request = pvaccess.PvObject({'a': pvaccess.INT, 'b': pvaccess.INT})
  >>> request.set({'a': 11, 'b': 22})
  >>> print request
  structure
      int a 11
      int b 22
  >>> response = rpc.invoke(request)
  >>> print response
  structure
      int value 33

Example 3

  1. In terminal 2:
  >>> import pvaccess
  >>> srv = pvaccess.RpcServer()
  >>> def hash(x):
  ...     import hashlib
  ...     md5 = hashlib.md5()
  ...     md5.update(str(x))
  ...     h = md5.hexdigest()
  ...     dict = x.getStructureDict()
  ...     dict['hash'] = pvaccess.STRING
  ...     response = pvaccess.PvObject(dict)
  ...     response.setString('hash', h)
  ...     return response
  >>> srv.registerService('hash', hash)
  >>> srv.listen()
  1. In terminal 1:
  >>> rpc = pvaccess.RpcClient('hash')
  >>> request = pvaccess.PvString('abcd')
  >>> print rpc.invoke(request)
  structure
      string hash 0a380e7375d8c3f68d1bbe068141d6ce
      string value

Project details


Download files

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

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

pvapy-1.5.0-cp37-cp37m-manylinux1_x86_64.whl (62.3 MB view details)

Uploaded CPython 3.7m

pvapy-1.5.0-cp36-cp36m-manylinux1_x86_64.whl (62.3 MB view details)

Uploaded CPython 3.6m

pvapy-1.5.0-cp27-cp27mu-manylinux1_x86_64.whl (62.3 MB view details)

Uploaded CPython 2.7mu

pvapy-1.5.0-cp27-cp27m-manylinux1_x86_64.whl (62.3 MB view details)

Uploaded CPython 2.7m

File details

Details for the file pvapy-1.5.0-cp37-cp37m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-1.5.0-cp37-cp37m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 62.3 MB
  • Tags: CPython 3.7m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.2

File hashes

Hashes for pvapy-1.5.0-cp37-cp37m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 dac2587f1fc48b08bd41fae994e44feffe70884dfb148303743875258b743666
MD5 8e0ac02da90ba6eb5d6cdae736440ab0
BLAKE2b-256 a3d926a1a069a02ef22dd9690231a6d7795ef2678540f1e6b311db26ae94ca69

See more details on using hashes here.

File details

Details for the file pvapy-1.5.0-cp36-cp36m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-1.5.0-cp36-cp36m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 62.3 MB
  • Tags: CPython 3.6m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.2

File hashes

Hashes for pvapy-1.5.0-cp36-cp36m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 a0597bd0278e578eae425daf01a92d38e05a945b03e8121bf0e4e930cd492d00
MD5 76198e60d077beccbaf3f2c792e4799a
BLAKE2b-256 3c8dbe0183a7f2177e81bee3c8b362aabbc236e3d1e8fd825da613856b5c28bc

See more details on using hashes here.

File details

Details for the file pvapy-1.5.0-cp27-cp27mu-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-1.5.0-cp27-cp27mu-manylinux1_x86_64.whl
  • Upload date:
  • Size: 62.3 MB
  • Tags: CPython 2.7mu
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.2

File hashes

Hashes for pvapy-1.5.0-cp27-cp27mu-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 c87467ed278cdae7ecaffc710c99e37be05683a0c5948e7802fbfb97ad0885c6
MD5 5d2f0feaa9c08f54e897429f8d19af51
BLAKE2b-256 bde20f9538a1aac48f9aee38218c7e4d40625579acee45dfbbfc3eb281a4f0cb

See more details on using hashes here.

File details

Details for the file pvapy-1.5.0-cp27-cp27m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-1.5.0-cp27-cp27m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 62.3 MB
  • Tags: CPython 2.7m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.2

File hashes

Hashes for pvapy-1.5.0-cp27-cp27m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 b0c29e50d28f071a19c22ec2545f8f1dfd0918a8965f599eb8e0c5b5f2394527
MD5 bb38be4af958773535cd1243eff77f62
BLAKE2b-256 02cc7dc95ad0c8d07ba83bda7bf81e7a912e274fdb1afb4d553bf0f4633d7bb3

See more details on using hashes here.

Supported by

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