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-2.1.0-cp38-cp38-manylinux1_x86_64.whl (12.9 MB view details)

Uploaded CPython 3.8

pvapy-2.1.0-cp37-cp37m-manylinux1_x86_64.whl (12.8 MB view details)

Uploaded CPython 3.7m

pvapy-2.1.0-cp36-cp36m-manylinux1_x86_64.whl (12.8 MB view details)

Uploaded CPython 3.6m

pvapy-2.1.0-cp35-cp35m-manylinux1_x86_64.whl (12.8 MB view details)

Uploaded CPython 3.5m

pvapy-2.1.0-cp34-cp34m-manylinux1_x86_64.whl (12.8 MB view details)

Uploaded CPython 3.4m

pvapy-2.1.0-cp27-cp27mu-manylinux1_x86_64.whl (12.8 MB view details)

Uploaded CPython 2.7mu

pvapy-2.1.0-cp27-cp27m-manylinux1_x86_64.whl (12.8 MB view details)

Uploaded CPython 2.7m

File details

Details for the file pvapy-2.1.0-cp38-cp38-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-2.1.0-cp38-cp38-manylinux1_x86_64.whl
  • Upload date:
  • Size: 12.9 MB
  • Tags: CPython 3.8
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.7.6

File hashes

Hashes for pvapy-2.1.0-cp38-cp38-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 dccc7a274f33ced3aefde880223176618209e71cd847e9818c48fe7b0e1f8793
MD5 46ed783a080ed80f3839a9f067242e2d
BLAKE2b-256 9778a74e91d88fb15cb064b8ecab536e1163b234485cc882695fad80a5eebe31

See more details on using hashes here.

File details

Details for the file pvapy-2.1.0-cp37-cp37m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-2.1.0-cp37-cp37m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 12.8 MB
  • Tags: CPython 3.7m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.7.6

File hashes

Hashes for pvapy-2.1.0-cp37-cp37m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 bf831294868f55c16ae65f71d796d37da2a08fff69b9503fd1568ef428918d43
MD5 660ea75e13d22b42e65faa22e4e63936
BLAKE2b-256 33034a50916ead0b204b3da689bcc6167f1d583444da71c9c2ef0702c00ae902

See more details on using hashes here.

File details

Details for the file pvapy-2.1.0-cp36-cp36m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-2.1.0-cp36-cp36m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 12.8 MB
  • Tags: CPython 3.6m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.7.6

File hashes

Hashes for pvapy-2.1.0-cp36-cp36m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 98f452cca2b43897f1b0505d9ac899a804b0f1e1682658ea79a1d3a024c964cf
MD5 e5bf46fe019edc28cdc9ea4560e1d578
BLAKE2b-256 e0aba17af12582a860a289a3360de96a51d1d7a0feadc5620fb9eeb006175d31

See more details on using hashes here.

File details

Details for the file pvapy-2.1.0-cp35-cp35m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-2.1.0-cp35-cp35m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 12.8 MB
  • Tags: CPython 3.5m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.7.6

File hashes

Hashes for pvapy-2.1.0-cp35-cp35m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 c7f73e772b3be403d193298aae169a87c41fdff35754db3298ea0aa232ab8e80
MD5 97db1cd3cdf4510462b099c732e62554
BLAKE2b-256 58e745230c0940cb9fd38cdd6e40a73775b66aa7bf0adc2aae42a51c41348753

See more details on using hashes here.

File details

Details for the file pvapy-2.1.0-cp34-cp34m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-2.1.0-cp34-cp34m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 12.8 MB
  • Tags: CPython 3.4m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.7.6

File hashes

Hashes for pvapy-2.1.0-cp34-cp34m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 2a5db5d0d630d8a6e15c4749986224dacd3f2c8824bbc7c439fb6cd6397fbc46
MD5 28aaa02096db8947159c0cb10d210097
BLAKE2b-256 3ddb4853cc5eddec26c9794523ebbf9432db79f8f1995d14d23f951f83e989a3

See more details on using hashes here.

File details

Details for the file pvapy-2.1.0-cp27-cp27mu-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-2.1.0-cp27-cp27mu-manylinux1_x86_64.whl
  • Upload date:
  • Size: 12.8 MB
  • Tags: CPython 2.7mu
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.7.6

File hashes

Hashes for pvapy-2.1.0-cp27-cp27mu-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 318e34de38219d7cc171f61a5e67c348c999e8e7a9ae29968508d9c3b92bb999
MD5 49567a28db96129e70c7228e4984aa36
BLAKE2b-256 5cbf90ff6ef4a2e06e3a5c174daf167271b297e2c4448f8604ff4afa861bde29

See more details on using hashes here.

File details

Details for the file pvapy-2.1.0-cp27-cp27m-manylinux1_x86_64.whl.

File metadata

  • Download URL: pvapy-2.1.0-cp27-cp27m-manylinux1_x86_64.whl
  • Upload date:
  • Size: 12.8 MB
  • Tags: CPython 2.7m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.41.1 CPython/3.7.6

File hashes

Hashes for pvapy-2.1.0-cp27-cp27m-manylinux1_x86_64.whl
Algorithm Hash digest
SHA256 69c3c158cdce5aac4ebb4852dcb90482473c80b65e067bf3c3ff885d0ce8b84e
MD5 8a6b1ae556257485fb9d7e0d1cfbfde9
BLAKE2b-256 72abdf423b22d94a2d21dc9c7016c92246f79fde594087cdbeaa21e9e993127c

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