This package has two main uses:
- To provide user-facing documentation of things like relations and parts of speech used by wordnets.
- To provide a Python API for querying this documentation, such as for retrieving the localized name or definition for specific relations.
You can use pip
to install gwadoc (use of a virtual environment is
strongly recommended; see here). The minimum Python
version is 3.5. As gwadoc is not on PyPI, you can
give the GitHub URL to pip
:
~$ pip install git+git://github.com/globalwordnet/gwadoc.git
... or clone the repository and install from your local copy:
~$ git clone https://github.com/globalwordnet/gwadoc.git
~$ cd gwadoc/
~/gwadoc$ pip install .
To mark gwadoc as a dependency for your project, just put the
git+git://github.com/globalwordnet/gwadoc.git
URL in your
requirements.txt
file or the appropriate place in your setup.py
file.
To install the requirements for building the documentation, use the
[build]
extra. For example, when installing from a local copy:
~/gwadoc$ pip install .[build]
To use gwadoc in your Python code Then you can use it like this:
>>> import gwadoc
>>> for relname in gwadoc.RELATIONS[:5]:
... print(relname, '\n ', gwadoc.relations[relname].df.en)
...
constitutive
Core semantic relations that define synsets
hyponym
a word that is more specific than a given word
hypernym
a word that is more general than a given word
instance_hyponym
an occurrence of something
instance_hypernym
the type of an instance
Where gwadoc.RELATIONS
is the inventory of relation names, and
gwadoc.relations
is the data structure containing documentation
about each relation. For each relation, there are several fields:
Field | Description | Comments |
---|---|---|
df |
short definition | Suitable for use in a mouseover |
ex |
short example | Suitable for use in a mouseover |
dfn |
long definition | |
exe |
more examples | |
fa |
formal attributes | |
name |
relation name | |
proj |
project-specific names | |
test |
linguistic tests | Probably language specific |
For the formal attributes and project names that are available, see
gwadoc.inventories
. The data structure
may be accessed (and defined) using dot-notation or index-notation:
>>> print(gwadoc.relations.hypernym.name)
Hypernym
>>> print(gwadoc.relations['hypernym']['name'])
Hypernym
String fields, such as df
, ex
, name
, test
, etc. use the
gwadoc.base.MultiString
class (see gwadoc.base
)
so that projects may provide these fields in their own language. The
language may be requested directly, or the default language may be
specified by calling the gwadoc.set_preferred_language()
function:
>>> gwadoc.relations.hypernym.name
<MultiString (Hypernym)>
>>> gwadoc.relations.hypernym.name.ja
'上位語'
>>> gwadoc.set_preferred_language('ja')
>>> gwadoc.relations.hypernym.name
<MultiString (上位語)>
After installing the requirements, you can build the HTML documentation:
# main documentation
~/gwadoc$ python docs/build.py html > docs/index.html
# summary
~/gwadoc$ python docs/build.py --object summary html > docs/summary.html
Note: if you installed the requirements in a virtual environment make sure you activate it before building the documentation.
On github, the documentation is automatically made when there is a commit to the master branch and pushed to the gh-pages branch.
Contributions are most welcome. Please see the contributing guide.
Global Wordnet Association Documentation by GWA Documentation Working Group is licensed under a Creative Commons Attribution 4.0 International License.