This page describes how to deploy ProvPy and its prerequisites.
The instructions have been written with reference to the 64-bit Ubuntu 14.04.1 operating system.
Other operating systems, or versions of these, may differ in how packages are installed, the versions of these packages available from package managers etc. Consult the relevant documentation for your operating system and the products concerned.
Some prerequisites require you to have sudo access to install and configure software (or a local system administrator can do this for you):
$ sudo su -
dot is a program for drawing directed graphs. It is part of Graphviz open source graph visualization software.
Install:
$ sudo apt-get install graphviz
$ dot -V
dot - graphviz version 2.36.0 (20140111.2315)
cURL is a very useful command line tool and library for interacting over HTTP(S).
Install:
$ sudo apt-get install curl
$ curl --version
curl 7.35.0 (x86_64-pc-linux-gnu) libcurl/7.35.0 OpenSSL/1.0.1f zlib/1.2.8 libidn/1.28 librtmp/2.3
zlib is a compression library.
Install:
$ sudo apt-get install zlib1g-dev
libxslt is an XSLT library. It is part of the libxml2 XML parser and toolkit.
Install:
$ sudo apt-get install libxslt1-dev
If you have sudo access then you can install Python system-wide.
Alternatively, if you do not have sudo access, or want to install Python in your home directory, see the following section on pyenv.
Install Python 2:
$ sudo apt-get install python
$ python --version
Python 2.7.6
Alternatively, install Python 3:
$ sudo apt-get install python3
$ python3 --version
Python 3.4.0
pip is the Python package manager.
Install for Python 2:
$ sudo apt-get install python-pip
$ pip -V
pip 1.5.4 from /usr/lib/python2.7/dist-packages (python 2.7)
Alternatively, install for Python 3:
$ sudo apt-get install python3-pip
$ pip3 -V
pip 1.5.4 from /usr/lib/python3/dist-packages (python 3.4)
setuptools is a package that supports the download, build, install and upgrade of Python packages.
Install for Python 2:
$ sudo apt-get install python-setuptools
Alternatively, install for Python 3:
$ sudo apt-get install python3-setuptools
virtualenv is a tool that creates isolated Python environments with all the executables and packages that that specific environment needs. virtualenvwrapper extends virtualenv with wrappers to make creating and managing virtual environments easier.
Install for Python 2:
$ sudo pip install virtualenvwrapper
Alternatively, install for Python 3:
$ sudo pip3 install virtualenvwrapper
Create environments directory. For example:
$ mkdir ~/Envs
Edit ~/.bash_profile and add the lines:
export WORKON_HOME=~/Envs
source /usr/local/bin/virtualenvwrapper.sh
Set environment:
$ source ~/bash_profile
pyenv allows you to deploy multiple versions of Python within your own directory, and to switch between these versions.
pyenv dependencies has a number of dependencies, which need to be installed by someone with sudo access.
Install:
$ sudo apt-get install -y make build-essential libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev wget curl llvm
Run:
$ curl -L https://raw.githubusercontent.com/yyuu/pyenv-installer/master/bin/pyenv-installer | bash
Edit ~/.bash_profile and add the lines:
export PATH="$HOME/.pyenv/bin:$PATH"
eval "$(pyenv init -)"
eval "$(pyenv virtualenv-init -)"
Set environment:
$ source ~/.bash_profile
See Python versions available:
$ pyenv update
$ pyenv install -l
Install desired versions. For example:
$ pyenv install 2.7.6
Or:
$ pyenv install 3.4.0
Troubleshooting: 'WARNING: The Python ... extension was not compiled.'
If you get warnings like:
WARNING: The Python readline extension was not compiled. Missing the GNU readline lib?
WARNING: The Python bz2 extension was not compiled. Missing the bzip2 lib?
WARNING: The Python sqlite3 extension was not compiled. Missing the SQLite3 lib?
ERROR: The Python ssl extension was not compiled. Missing the OpenSSL lib?
ERROR: The Python ssl extension was not compiled. Missing the OpenSSL lib?
Then install the pyenv dependencies. See pyenv's common build problems
If using virtual environments, then create a virtual environment for ProvPy. For example:
$ mkvirtualenv prov
If using pyenv, then select the Python version into whose packages you want to install ProvPy. For example:
$ pyenv local 2.7.6
$ pip install prov
$ pip list | grep prov
prov (1.3.1)
$ pip show prov
---
Name: prov
Version: 1.3.1
Location: /usr/local/lib/python2.7/dist-packages
Requires: six, python-dateutil, lxml, networkx
Alternatively, if you have installed setuptools, you can do:
$ easy_install prov
prov-convert is a script that converts PROV documents. For usage information, run:
$ prov-convert --help
This only runs under Python 2.7.
For example, create a file, example.json:
{"prefix": {"default": "http://example.org/0/", "ex2": "http://example.org/2/", "ex1": "http://example.org/1/"}, "bundle": {"e001": {"prefix": {"default": "http://example.org/2/"}, "entity": {"e001": {}}}}, "entity": {"e001": {}}}
Run:
$ prov-convert -f provn example.json example.provn
$ cat example.provn
document
default <http://example.org/0/>
prefix ex2 <http://example.org/2/>
prefix ex1 <http://example.org/1/>
entity(e001)
bundle e001
default <http://example.org/2/>
entity(e001)
endBundle
endDocument
$ prov-convert -f xml example.json example.xml
$ cat example.xml
<?xml version='1.0' encoding='UTF-8'?>
<prov:document xmlns:prov="http://www.w3.org/ns/prov#" xmlns:ex2="http://example.org/2/" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:ex1="http://example.org/1/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://example.org/0/">
<prov:entity prov:id="e001"/>
<prov:bundleContent prov:id="e001">
<prov:entity prov:id="e001"/>
</prov:bundleContent>
</prov:document>
$ prov-convert -f pdf example.json example.pdf
$ prov-convert -f svg example.json example.svg
$ prov-convert -f dot example.json example.dot
If, when specifying .pdf, .svg, or .dot as an output format you see:
$ prov-convert -f pdf example.json example.pdf
prov-convert: No module named pydot
for help use --help
$ prov-convert -f svg example.json example.svg
prov-convert: No module named pydot
for help use --help(
If, when specifying .pdf, .svg, or .dot as an output format you see:
$ prov-convert -f png example.json example.png
prov-convert: GraphViz's executables not found
for help use --help
$ prov-convert -f dot example.json example.dot
...as above...
Then install Graphviz.
If, when using pyenv, you get:
$ prov-convert --help
prov-convert: command not found
then try:
.pyenv/versions/2.7.6/bin/prov-convert --help
Remember to replace 2.7.6 with your version of Python.
prov-compare is a script that compares PROV documents to see if they are semantically equivalent. For usage information, run:
$ prov-compare --help
This only runs under Python 2.7.
If, when using pyenv, you get:
$ prov-compare --help
prov-compare: command not found
then try:
.pyenv/versions/2.7.6/bin/prov-compare --help
Remember to replace 2.7.6 with your version of Python.
Create a file, simple-example.py:
import prov.model as prov
import datetime
document = prov.ProvDocument()
document.set_default_namespace('http://anotherexample.org/')
document.add_namespace('ex', 'http://example.org/')
e2 = document.entity('e2', (
(prov.PROV_TYPE, "File"),
('ex:path', "/shared/crime.txt"),
('ex:creator', "Alice"),
('ex:content', "There was a lot of crime in London last month"),
))
a1 = document.activity('a1', datetime.datetime.now(), None, {prov.PROV_TYPE: "edit"})
# References can be qnames or ProvRecord objects themselves
document.wasGeneratedBy(e2, a1, None, {'ex:fct': "save"})
document.wasAssociatedWith('a1', 'ag2', None, None, {prov.PROV_ROLE: "author"})
document.agent('ag2', {prov.PROV_TYPE: 'prov:Person', 'ex:name': "Bob"})
print(document.get_provn())
Run:
$ python simple-example.py
document
default <http://anotherexample.org/>
prefix ex <http://example.org/>
entity(e2, [prov:type="File", ex:content="There was a lot of crime in London last month", ex:creator="Alice", ex:path="/shared/crime.txt"])
activity(a1, 2015-05-07T07:45:28.698831, -, [prov:type="edit"])
wasGeneratedBy(e2, a1, -)
wasAssociatedWith(a1, ag2, -, [prov:role="author"])
agent(ag2, [prov:type="prov:Person", ex:name="Bob"])
endDocument
Create a file, bundle-example.py:
import prov.model as prov
document = prov.ProvDocument()
document.set_default_namespace('http://example.org/0/')
document.add_namespace('ex1', 'http://example.org/1/')
document.add_namespace('ex2', 'http://example.org/2/')
document.entity('e001')
bundle = document.bundle('e001')
bundle.set_default_namespace('http://example.org/2/')
bundle.entity('e001')
print (document.get_provn())
print (document.serialize())
Run:
$ python bundle-example.py
document
default <http://example.org/0/>
prefix ex2 <http://example.org/2/>
prefix ex1 <http://example.org/1/>
entity(e001)
bundle e001
default <http://example.org/2/>
entity(e001)
endBundle
endDocument
{"prefix": {"default": "http://example.org/0/", "ex2": "http://example.org/2/", "ex1": "http://example.org/1/"}, "bundle": {"e001": {"prefix": {"default": "http://example.org/2/"}, "entity": {"e001": {}}}}, "entity": {"e001": {}}}
If, when specifying pdf, svg, dot or another graphical format you see, when running your Python script:
ImportError: No module named pydot
Then, if using Python 2, install pydot:
$ pip install pydot
If using Python 3, install a Python 3-compatible version of pydot:
$ pip install https://bitbucket.org/prologic/pydot/get/ac76697320d6.zip
If you get this error when running code that uses ProvPy:
Couldn't import dot_parser, loading of dot files will not be possible.
then please ignore it. Dot files are not loaded by ProvPy components.