SpecGenNG

Yet Another Version of SpecGen!

Documentation generator for RDF/OWL vocabularies/ontologies : RDF + template in, HTML out.

This is zazi0815's Version 6, ported to Python 3 (from 2.6).

Notes under About v6 are lifted from https://github.com/specgen/specgen, will mostly apply here too. Some notes about porting in here.

I'll update instructions here once I've actually used it properly and someone else has kicked its tyres (and time permits). The command line below is fairly self-explanatory.

Key bits are:

clone this repo
put your vocab somewhere convenient (like the onto dir)
cd specgenng
run something like :

python src/specgenng.py \
      --indir=onto/project/ \
      --ns=http://purl.org/stuff/project/ \
      --prefix=prj \
      --ontofile=project_2007-09-03.rdf \
      --templatedir=template \
      --outdir=spec/project/ \
      --outfile=project_2007-09-03.html

A template file is required, there's a blank one under template.

I wanted to update a vocab+docs and after spending some serious time hunting for a recent version of specgen or equivalent, gave up, bit the bullet, updated this v6. The heavy lifting was done by Python 2to3. I've named it SpecGenNG (v1.0.0 I guess) because it's a breaking change and there's probably a v7 already out there... There was no obvious license so I've slapped an MIT on it just in case, but to all intents and purposes it's public domain.

About v6 (2010-ish)

This is an experimental new codebase for specgen tools based on danbri's specgen5 version, which was heavily updated by Bo Ferri in summer 2010.

Features (incl. modifications + extensions):

multiple property and class types
muttiple restrictions modelling
rdfs:label, rdfs:comment
classes and properties from other namespaces
inverse properties (explicit and anonymous)
sub properties
union ranges and domains (appear only in the property descriptions, not on the class descriptions)
equivalent properties
simple individuals as optional feature

Dependencies

It depends utterly upon

rdflib
rdfextras (easy_install rdfextras)
pyparsing (easy_install pyparsing)
igraph (easy_install python-igraph)

(at least I had to install these packages additionally ;) )

If you're lucky, typing this is enough:

easy_install rdflib python-igraph

and if you have problems there, update easy_install etc with:

easy_install -U setuptools

Ubuntu, you can install the dependencies with pip after installing the relevant libraries

sudo apt-get install python-dev build-essential libxml2-dev libxslt python-igraph
sudo pip install -r requirements.txt

Purpose

Inputs: RDF, HTML and OWL description(s) of an RDF vocabulary
Output: an XHTML+RDFa specification designed for human users

Example

specgen6.py --indir=onto/olo/ --ns=http://purl.org/ontology/olo/core#  --prefix=olo --ontofile=orderedlistontology.owl --outdir=spec/olo/ --templatedir=onto/olo/ --outfile=orderedlistontology.html

the template of this example can also be found in the folder: onto/olo
the output of this example can also be found in the folder: spec/olo

See libvocab.py and specgen6.py for details.

Status

we load up and interpret the core RDFS/OWL
we populate Vocab, Term (Class, Property or Individual) instances
able to generate a XHTML/RDFa ontology specification with common concepts and properties from OWL, RDFS, RDF

Known Forks

WebID fork (note this link is only a reference to a specific revision of that fork; to ensure that you'll utilise the most recent one, go to summary and walk that path to the specgen directory again from the most recent revision ;) )

TODO

enable more OWL features, especially an automated construction of owl:Ontology (currently this must be done manually in the template)
enable more support for other namespaces (super classes and super properties from other namespaces already possible)
restructure the code !!!
write a cool parser for the "\n"'s and "\t"'s etc. in the parsed comments (e.g. "\n" to <br/> ...)

Known Issues

librdf doesn't seem to like abbreviations in FILTER clauses.

this worked:

q= 'SELECT ?x ?l ?c ?type WHERE { ?x rdfs:label ?l . ?x rdfs:comment ?c . ?x a ?type .  FILTER (?type = <http://www.w3.org/2002/07/owl#ObjectProperty\>)  } '

while this failed:

q= 'PREFIX owl:  SELECT ?x ?l ?c ?type WHERE { ?x rdfs:label ?l . ?x rdfs:comment ?c . ?x a ?type .  FILTER (?type = owl:ObjectProperty)  } '

(even when passing in bindings)

This forces us to be verbose, ie.

q= 'SELECT distinct ?x ?l ?c WHERE { ?x rdfs:label ?l . ?x rdfs:comment ?c . ?x a ?type . FILTER (?type = <http://www.w3.org/2002/07/owl#ObjectProperty> || ?type = <http://www.w3.org/2002/07/owl#DatatypeProperty> || ?type = <http://www.w3.org/1999/02/22-rdf-syntax-ns#Property> || ?type = <http://www.w3.org/2002/07/owl#FunctionalProperty> || ?type = <http://www.w3.org/2002/07/owl#InverseFunctionalProperty>) } '

TODO: work out how to do ".encode('UTF-8')" everywhere
Be more explicit and careful re defaulting to English, and more robust when multilingual labels are found.

PS

The old project repository location at SourceForge is now deprecated. All new developments will be pushed to this repository location here at GitHub.

danja / specgen-ng