sparql module

The sparql module can be invoked in several different ways. To quickly run a query use query(). Results are encapsulated in a _ResultsParser instance:

>>> result = sparql.query(endpoint, query)
>>> for row in result:
>>>    print row

Command-line use

sparql.py [-i] endpoint
    -i Interactive mode

If interactive mode is enabled, the program reads queries from the console and then executes them. Use a double line (two ‘enters’) to separate queries.

Otherwise, the query is read from standard input.

RDF wrapper classes

class sparql.RDFTerm

Super class containing methods to override. IRI, Literal and BlankNode all inherit from RDFTerm.

n3()

Return a Notation3 representation of this term.

class sparql.IRI(value)

An RDF resource.

class sparql.Literal(value, datatype=None, lang=None)

Literals. These can take a data type or a language code.

class sparql.BlankNode(value)

Blank node. Similar to IRI but lacks a stable identifier.

Query utilities

class sparql.Service(endpoint, qs_encoding='utf-8')

This is the main entry to the library. The user creates a Service, then sends a query to it. If we want to have persistent connections, then open them here.

class sparql._ResultsParser(fp)

Parse the XML result.

__iter__()

Synonim for fetchone().

fetchall()

Loop through the result to build up a list of all rows. Patterned after DB-API 2.0.

fetchone()

Fetches the next set of rows of a query result, returning a list. An empty list is returned when no more rows are available. If the query was an ASK request, then an empty list is returned as there are no rows available.

hasresult()

ASK queries are used to test if a query would have a result. If the query is an ASK query there won’t be an actual result, and fetchone() will return nothing. Instead, this method can be called to check the result from the ASK query.

If the query is a SELECT statement, then the return value of hasresult() is None, as the XML result format doesn’t tell you if there are any rows in the result until you have read the first one.

sparql.parse_n3_term(src)

Parse a Notation3 value into a RDFTerm object (IRI or Literal).

This parser understands IRIs and quoted strings; basic non-string types (integers, decimals, booleans, etc) are not supported yet.

sparql.unpack_row(row, convert=None, convert_type={})

Convert values in the given row from RDFTerm objects to plain Python values: IRI is converted to a unicode string containing the IRI value; BlankNode is converted to a unicode string with the BNode’s identifier, and Literal is converted based on its XSD datatype.

The library knows about common XSD types (STRING becomes unicode, INTEGER and LONG become int, DOUBLE and FLOAT become float, DECIMAL becomes Decimal, BOOLEAN becomes bool). If the python-dateutil library is found, then DATE, TIME and DATETIME are converted to date, time and datetime respectively. For other conversions, an extra argument convert may be passed. It should be a callable accepting two arguments: the serialized value as a unicode object, and the XSD datatype.

sparql.query(endpoint, query)

Convenient method to execute a query. Exactly equivalent to:

sparql.Service(endpoint).query(query)