Source code for rdflib.query

from __future__ import annotations

import itertools
import types
import warnings
from io import BytesIO
from typing import (
    IO,
    TYPE_CHECKING,
    Any,
    BinaryIO,
    Dict,
    Iterator,
    List,
    Mapping,
    MutableSequence,
    Optional,
    Tuple,
    Union,
    cast,
    overload,
)
from urllib.parse import urlparse
from urllib.request import url2pathname

__all__ = [
    "Processor",
    "UpdateProcessor",
    "Result",
    "ResultRow",
    "ResultParser",
    "ResultSerializer",
    "ResultException",
    "EncodeOnlyUnicode",
]

if TYPE_CHECKING:
    from rdflib.graph import Graph, _TripleType
    from rdflib.plugins.sparql.sparql import Query, Update
    from rdflib.term import Identifier, Variable


[docs]class Processor(object): """ Query plugin interface. This module is useful for those wanting to write a query processor that can plugin to rdf. If you are wanting to execute a query you likely want to do so through the Graph class query method. """
[docs] def __init__(self, graph: "Graph"): pass
[docs] def query( self, strOrQuery: Union[str, "Query"], # noqa: N803 initBindings: Mapping["Variable", "Identifier"] = {}, # noqa: N803 initNs: Mapping[str, Any] = {}, # noqa: N803 DEBUG: bool = False, ) -> Mapping[str, Any]: pass
[docs]class UpdateProcessor(object): """ Update plugin interface. This module is useful for those wanting to write an update processor that can plugin to rdflib. If you are wanting to execute an update statement you likely want to do so through the Graph class update method. .. versionadded:: 4.0 """
[docs] def __init__(self, graph: "Graph"): pass
[docs] def update( self, strOrQuery: Union[str, "Update"], # noqa: N803 initBindings: Mapping["Variable", "Identifier"] = {}, # noqa: N803 initNs: Mapping[str, Any] = {}, ) -> None: pass
[docs]class ResultException(Exception): pass
[docs]class EncodeOnlyUnicode(object): """ This is a crappy work-around for http://bugs.python.org/issue11649 """
[docs] def __init__(self, stream: BinaryIO): self.__stream = stream
[docs] def write(self, arg): if isinstance(arg, str): self.__stream.write(arg.encode("utf-8")) else: self.__stream.write(arg)
[docs] def __getattr__(self, name: str) -> Any: return getattr(self.__stream, name)
[docs]class ResultRow(Tuple["Identifier", ...]): """ a single result row allows accessing bindings as attributes or with [] >>> from rdflib import URIRef, Variable >>> rr=ResultRow({ Variable('a'): URIRef('urn:cake') }, [Variable('a')]) >>> rr[0] rdflib.term.URIRef(u'urn:cake') >>> rr[1] Traceback (most recent call last): ... IndexError: tuple index out of range >>> rr.a rdflib.term.URIRef(u'urn:cake') >>> rr.b Traceback (most recent call last): ... AttributeError: b >>> rr['a'] rdflib.term.URIRef(u'urn:cake') >>> rr['b'] Traceback (most recent call last): ... KeyError: 'b' >>> rr[Variable('a')] rdflib.term.URIRef(u'urn:cake') .. versionadded:: 4.0 """ labels: Mapping[str, int]
[docs] def __new__( cls, values: Mapping["Variable", "Identifier"], labels: List["Variable"] ): # type error: Generator has incompatible item type "Optional[Any]"; expected "_T_co" instance = super(ResultRow, cls).__new__(cls, (values.get(v) for v in labels)) # type: ignore[misc] instance.labels = dict((str(x[1]), x[0]) for x in enumerate(labels)) return instance
[docs] def __getattr__(self, name: str) -> "Identifier": if name not in self.labels: raise AttributeError(name) return tuple.__getitem__(self, self.labels[name])
# type error: Signature of "__getitem__" incompatible with supertype "tuple" # type error: Signature of "__getitem__" incompatible with supertype "Sequence"
[docs] def __getitem__(self, name: Union[str, int, Any]) -> "Identifier": # type: ignore[override] try: # type error: Invalid index type "Union[str, int, Any]" for "tuple"; expected type "int" return tuple.__getitem__(self, name) # type: ignore[index] except TypeError: if name in self.labels: # type error: Invalid index type "Union[str, int, slice, Any]" for "Mapping[str, int]"; expected type "str" return tuple.__getitem__(self, self.labels[name]) # type: ignore[index] if str(name) in self.labels: # passing in variable object return tuple.__getitem__(self, self.labels[str(name)]) raise KeyError(name)
@overload def get(self, name: str, default: "Identifier") -> "Identifier": ... @overload def get( self, name: str, default: Optional["Identifier"] = ... ) -> Optional["Identifier"]: ...
[docs] def get( self, name: str, default: Optional["Identifier"] = None ) -> Optional["Identifier"]: try: return self[name] except KeyError: return default
[docs] def asdict(self) -> Dict[str, "Identifier"]: return dict((v, self[v]) for v in self.labels if self[v] is not None)
[docs]class Result(object): """ A common class for representing query result. There is a bit of magic here that makes this appear like different Python objects, depending on the type of result. If the type is "SELECT", iterating will yield lists of ResultRow objects If the type is "ASK", iterating will yield a single bool (or bool(result) will return the same bool) If the type is "CONSTRUCT" or "DESCRIBE" iterating will yield the triples. len(result) also works. """
[docs] def __init__(self, type_: str): if type_ not in ("CONSTRUCT", "DESCRIBE", "SELECT", "ASK"): raise ResultException("Unknown Result type: %s" % type_) self.type = type_ #: variables contained in the result. self.vars: Optional[List["Variable"]] = None self._bindings: MutableSequence[Mapping["Variable", "Identifier"]] = None # type: ignore[assignment] self._genbindings: Optional[Iterator[Mapping["Variable", "Identifier"]]] = None self.askAnswer: Optional[bool] = None self.graph: Optional["Graph"] = None
@property def bindings(self) -> MutableSequence[Mapping[Variable, Identifier]]: """ a list of variable bindings as dicts """ if self._genbindings: self._bindings += list(self._genbindings) self._genbindings = None return self._bindings @bindings.setter def bindings( self, b: Union[ MutableSequence[Mapping["Variable", "Identifier"]], Iterator[Mapping[Variable, Identifier]], ], ) -> None: if isinstance(b, (types.GeneratorType, itertools.islice)): self._genbindings = b self._bindings = [] else: # type error: Incompatible types in assignment (expression has type "Union[MutableSequence[Mapping[Variable, Identifier]], Iterator[Mapping[Variable, Identifier]]]", variable has type "MutableSequence[Mapping[Variable, Identifier]]") self._bindings = b # type: ignore[assignment]
[docs] @staticmethod def parse( source: Optional[IO] = None, format: Optional[str] = None, content_type: Optional[str] = None, **kwargs: Any, ) -> "Result": from rdflib import plugin if format: plugin_key = format elif content_type: plugin_key = content_type.split(";", 1)[0] else: plugin_key = "xml" parser = plugin.get(plugin_key, ResultParser)() # type error: Argument 1 to "parse" of "ResultParser" has incompatible type "Optional[IO[Any]]"; expected "IO[Any]" return parser.parse( source, content_type=content_type, **kwargs # type:ignore[arg-type] )
[docs] def serialize( self, destination: Optional[Union[str, IO]] = None, encoding: str = "utf-8", format: str = "xml", **args: Any, ) -> Optional[bytes]: """ Serialize the query result. The :code:`format` argument determines the Serializer class to use. - csv: :class:`~rdflib.plugins.sparql.results.csvresults.CSVResultSerializer` - json: :class:`~rdflib.plugins.sparql.results.jsonresults.JSONResultSerializer` - txt: :class:`~rdflib.plugins.sparql.results.txtresults.TXTResultSerializer` - xml: :class:`~rdflib.plugins.sparql.results.xmlresults.XMLResultSerializer` :param destination: Path of file output or BufferedIOBase object to write the output to. :param encoding: Encoding of output. :param format: One of ['csv', 'json', 'txt', xml'] :param args: :return: bytes """ if self.type in ("CONSTRUCT", "DESCRIBE"): # type error: Item "None" of "Optional[Graph]" has no attribute "serialize" # type error: Incompatible return value type (got "Union[bytes, str, Graph, Any]", expected "Optional[bytes]") return self.graph.serialize( # type: ignore[union-attr,return-value] destination, encoding=encoding, format=format, **args ) """stolen wholesale from graph.serialize""" from rdflib import plugin serializer = plugin.get(format, ResultSerializer)(self) if destination is None: streamb: BytesIO = BytesIO() stream2 = EncodeOnlyUnicode(streamb) # type error: Argument 1 to "serialize" of "ResultSerializer" has incompatible type "EncodeOnlyUnicode"; expected "IO[Any]" serializer.serialize(stream2, encoding=encoding, **args) # type: ignore[arg-type] return streamb.getvalue() if hasattr(destination, "write"): stream = cast(IO[bytes], destination) serializer.serialize(stream, encoding=encoding, **args) else: location = cast(str, destination) scheme, netloc, path, params, query, fragment = urlparse(location) if scheme == "file": if netloc != "": raise ValueError( f"the file URI {location!r} has an authority component which is not supported" ) os_path = url2pathname(path) else: os_path = location with open(os_path, "wb") as stream: serializer.serialize(stream, encoding=encoding, **args) return None
[docs] def __len__(self) -> int: if self.type == "ASK": return 1 elif self.type == "SELECT": return len(self.bindings) else: # type error: Argument 1 to "len" has incompatible type "Optional[Graph]"; expected "Sized" return len(self.graph) # type: ignore[arg-type]
[docs] def __bool__(self) -> bool: if self.type == "ASK": # type error: Incompatible return value type (got "Optional[bool]", expected "bool") return self.askAnswer # type: ignore[return-value] else: return len(self) > 0
[docs] def __iter__( self, ) -> Iterator[Union["_TripleType", bool, ResultRow]]: if self.type in ("CONSTRUCT", "DESCRIBE"): # type error: Item "None" of "Optional[Graph]" has no attribute "__iter__" (not iterable) for t in self.graph: # type: ignore[union-attr] yield t elif self.type == "ASK": # type error: Incompatible types in "yield" (actual type "Optional[bool]", expected type "Union[Tuple[Identifier, Identifier, Identifier], bool, ResultRow]") [misc] yield self.askAnswer # type: ignore[misc] elif self.type == "SELECT": # this iterates over ResultRows of variable bindings if self._genbindings: for b in self._genbindings: if b: # don't add a result row in case of empty binding {} self._bindings.append(b) # type error: Argument 2 to "ResultRow" has incompatible type "Optional[List[Variable]]"; expected "List[Variable]" yield ResultRow(b, self.vars) # type: ignore[arg-type] self._genbindings = None else: for b in self._bindings: if b: # don't add a result row in case of empty binding {} # type error: Argument 2 to "ResultRow" has incompatible type "Optional[List[Variable]]"; expected "List[Variable]" yield ResultRow(b, self.vars) # type: ignore[arg-type]
[docs] def __getattr__(self, name: str) -> Any: if self.type in ("CONSTRUCT", "DESCRIBE") and self.graph is not None: # type error: "Graph" has no attribute "__getattr__" return self.graph.__getattr__(self, name) # type: ignore[attr-defined] elif self.type == "SELECT" and name == "result": warnings.warn( "accessing the 'result' attribute is deprecated." " Iterate over the object instead.", DeprecationWarning, stacklevel=2, ) # copied from __iter__, above # type error: Item "None" of "Optional[List[Variable]]" has no attribute "__iter__" (not iterable) return [(tuple(b[v] for v in self.vars)) for b in self.bindings] # type: ignore[union-attr] else: raise AttributeError("'%s' object has no attribute '%s'" % (self, name))
[docs] def __eq__(self, other: Any) -> bool: try: if self.type != other.type: return False if self.type == "ASK": return self.askAnswer == other.askAnswer elif self.type == "SELECT": return self.vars == other.vars and self.bindings == other.bindings else: return self.graph == other.graph except: return False
[docs]class ResultParser(object):
[docs] def __init__(self): pass
[docs] def parse(self, source: IO, **kwargs: Any) -> Result: """return a Result object""" pass # abstract
[docs]class ResultSerializer(object):
[docs] def __init__(self, result: Result): self.result = result
[docs] def serialize(self, stream: IO, encoding: str = "utf-8", **kwargs: Any) -> None: """return a string properly serialized""" pass # abstract