smw

smw copied to clipboard

Semantic MediaWiki

Sematic MediaWiki Python binding

This package depends on the mwclient 0.7 development version

https://github.com/mwclient/mwclient.git

To install (will install mwclient 0.7dev automatically)

pip install smw

Or install from github

pip install git+git://github.com/baojie/smw.git

Usage example

from smw.SemanticMediaWiki import SemanticMediaWiki

# change this to your wiki's config
config = {
    "host": "www.foo.com",
    "path": "/wiki/",
    "http_login": None,
    "http_pass": None,
    "wiki_login": None,
    "wiki_pass": None,
}

wiki = SemanticMediaWiki(
    host=config["host"],
    path=config["path"],
    http_login=config["http_login"],
    http_pass=config["http_pass"],
    wiki_login=config["wiki_login"],
    wiki_pass=config["wiki_pass"])

# or from a config file (json)
# default ~/.smwrc
# sample config  https://github.com/baojie/smw/blob/master/smwrc.sample
wiki = SemanticMediaWiki.from_config()

query = r"""
{{#ask:
    [[SMW_PYTHON_TEST::+]]
|?SMW_PYTHON_TEST
|format = json
}}
"""

res = wiki.get_data(query, format='json')
[query_result, query_path] = res
# process query results from SMW >= 1.8.0
items = query_result['results']
for page in items:
    print items[page]

page_name = 'Foo'

# get page metadata in RDF/XML
rdf = wiki.getRDF(page_name)

# get page metadata in JSON, simplified for human readability
json_data = wiki.getJSON(page_name)

# parse any wiki script
result = wiki.parse("'''Hello'''")

# get raw page
# this is useful when query is composed on Special:Ask
html = wiki.get('/wiki/Main_Page')
json_result = wiki.get('/wiki/index.php?title=Special:Ask&q=[[Category:People]]&p[format]=json')

# add semantic annotations
page = wiki.site.Pages['SnowWhite']
old_content = page.edit()
new_content = old_content + "[[Category:Princess]][[friend of::Grumpy]]"
page.save(new_content)