Spaces:
Runtime error
Runtime error
File size: 10,577 Bytes
ed4d993 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 |
from __future__ import annotations
from typing import (
TYPE_CHECKING,
Dict,
List,
Optional,
)
if TYPE_CHECKING:
import rdflib
prefixes = {
"owl": """PREFIX owl: <http://www.w3.org/2002/07/owl#>\n""",
"rdf": """PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>\n""",
"rdfs": """PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>\n""",
"xsd": """PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>\n""",
}
cls_query_rdf = prefixes["rdfs"] + (
"""SELECT DISTINCT ?cls ?com\n"""
"""WHERE { \n"""
""" ?instance a ?cls . \n"""
""" OPTIONAL { ?cls rdfs:comment ?com } \n"""
"""}"""
)
cls_query_rdfs = prefixes["rdfs"] + (
"""SELECT DISTINCT ?cls ?com\n"""
"""WHERE { \n"""
""" ?instance a/rdfs:subClassOf* ?cls . \n"""
""" OPTIONAL { ?cls rdfs:comment ?com } \n"""
"""}"""
)
cls_query_owl = prefixes["rdfs"] + (
"""SELECT DISTINCT ?cls ?com\n"""
"""WHERE { \n"""
""" ?instance a/rdfs:subClassOf* ?cls . \n"""
""" FILTER (isIRI(?cls)) . \n"""
""" OPTIONAL { ?cls rdfs:comment ?com } \n"""
"""}"""
)
rel_query_rdf = prefixes["rdfs"] + (
"""SELECT DISTINCT ?rel ?com\n"""
"""WHERE { \n"""
""" ?subj ?rel ?obj . \n"""
""" OPTIONAL { ?rel rdfs:comment ?com } \n"""
"""}"""
)
rel_query_rdfs = (
prefixes["rdf"]
+ prefixes["rdfs"]
+ (
"""SELECT DISTINCT ?rel ?com\n"""
"""WHERE { \n"""
""" ?rel a/rdfs:subPropertyOf* rdf:Property . \n"""
""" OPTIONAL { ?rel rdfs:comment ?com } \n"""
"""}"""
)
)
op_query_owl = (
prefixes["rdfs"]
+ prefixes["owl"]
+ (
"""SELECT DISTINCT ?op ?com\n"""
"""WHERE { \n"""
""" ?op a/rdfs:subPropertyOf* owl:ObjectProperty . \n"""
""" OPTIONAL { ?op rdfs:comment ?com } \n"""
"""}"""
)
)
dp_query_owl = (
prefixes["rdfs"]
+ prefixes["owl"]
+ (
"""SELECT DISTINCT ?dp ?com\n"""
"""WHERE { \n"""
""" ?dp a/rdfs:subPropertyOf* owl:DatatypeProperty . \n"""
""" OPTIONAL { ?dp rdfs:comment ?com } \n"""
"""}"""
)
)
class RdfGraph:
"""RDFlib wrapper for graph operations.
Modes:
* local: Local file - can be queried and changed
* online: Online file - can only be queried, changes can be stored locally
* store: Triple store - can be queried and changed if update_endpoint available
Together with a source file, the serialization should be specified.
*Security note*: Make sure that the database connection uses credentials
that are narrowly-scoped to only include necessary permissions.
Failure to do so may result in data corruption or loss, since the calling
code may attempt commands that would result in deletion, mutation
of data if appropriately prompted or reading sensitive data if such
data is present in the database.
The best way to guard against such negative outcomes is to (as appropriate)
limit the permissions granted to the credentials used with this tool.
See https://python.langchain.com/docs/security for more information.
"""
def __init__(
self,
source_file: Optional[str] = None,
serialization: Optional[str] = "ttl",
query_endpoint: Optional[str] = None,
update_endpoint: Optional[str] = None,
standard: Optional[str] = "rdf",
local_copy: Optional[str] = None,
graph_kwargs: Optional[Dict] = None,
store_kwargs: Optional[Dict] = None,
) -> None:
"""
Set up the RDFlib graph
:param source_file: either a path for a local file or a URL
:param serialization: serialization of the input
:param query_endpoint: SPARQL endpoint for queries, read access
:param update_endpoint: SPARQL endpoint for UPDATE queries, write access
:param standard: RDF, RDFS, or OWL
:param local_copy: new local copy for storing changes
:param graph_kwargs: Additional rdflib.Graph specific kwargs
that will be used to initialize it,
if query_endpoint is provided.
:param store_kwargs: Additional sparqlstore.SPARQLStore specific kwargs
that will be used to initialize it,
if query_endpoint is provided.
"""
self.source_file = source_file
self.serialization = serialization
self.query_endpoint = query_endpoint
self.update_endpoint = update_endpoint
self.standard = standard
self.local_copy = local_copy
try:
import rdflib
from rdflib.plugins.stores import sparqlstore
except ImportError:
raise ImportError(
"Could not import rdflib python package. "
"Please install it with `pip install rdflib`."
)
if self.standard not in (supported_standards := ("rdf", "rdfs", "owl")):
raise ValueError(
f"Invalid standard. Supported standards are: {supported_standards}."
)
if (
not source_file
and not query_endpoint
or source_file
and (query_endpoint or update_endpoint)
):
raise ValueError(
"Could not unambiguously initialize the graph wrapper. "
"Specify either a file (local or online) via the source_file "
"or a triple store via the endpoints."
)
if source_file:
if source_file.startswith("http"):
self.mode = "online"
else:
self.mode = "local"
if self.local_copy is None:
self.local_copy = self.source_file
self.graph = rdflib.Graph()
self.graph.parse(source_file, format=self.serialization)
if query_endpoint:
store_kwargs = store_kwargs or {}
self.mode = "store"
if not update_endpoint:
self._store = sparqlstore.SPARQLStore(**store_kwargs)
self._store.open(query_endpoint)
else:
self._store = sparqlstore.SPARQLUpdateStore(**store_kwargs)
self._store.open((query_endpoint, update_endpoint))
graph_kwargs = graph_kwargs or {}
self.graph = rdflib.Graph(self._store, **graph_kwargs)
# Verify that the graph was loaded
if not len(self.graph):
raise AssertionError("The graph is empty.")
# Set schema
self.schema = ""
self.load_schema()
@property
def get_schema(self) -> str:
"""
Returns the schema of the graph database.
"""
return self.schema
def query(
self,
query: str,
) -> List[rdflib.query.ResultRow]:
"""
Query the graph.
"""
from rdflib.exceptions import ParserError
from rdflib.query import ResultRow
try:
res = self.graph.query(query)
except ParserError as e:
raise ValueError("Generated SPARQL statement is invalid\n" f"{e}")
return [r for r in res if isinstance(r, ResultRow)]
def update(
self,
query: str,
) -> None:
"""
Update the graph.
"""
from rdflib.exceptions import ParserError
try:
self.graph.update(query)
except ParserError as e:
raise ValueError("Generated SPARQL statement is invalid\n" f"{e}")
if self.local_copy:
self.graph.serialize(
destination=self.local_copy, format=self.local_copy.split(".")[-1]
)
else:
raise ValueError("No target file specified for saving the updated file.")
@staticmethod
def _get_local_name(iri: str) -> str:
if "#" in iri:
local_name = iri.split("#")[-1]
elif "/" in iri:
local_name = iri.split("/")[-1]
else:
raise ValueError(f"Unexpected IRI '{iri}', contains neither '#' nor '/'.")
return local_name
def _res_to_str(self, res: rdflib.query.ResultRow, var: str) -> str:
return (
"<"
+ str(res[var])
+ "> ("
+ self._get_local_name(res[var])
+ ", "
+ str(res["com"])
+ ")"
)
def load_schema(self) -> None:
"""
Load the graph schema information.
"""
def _rdf_s_schema(
classes: List[rdflib.query.ResultRow],
relationships: List[rdflib.query.ResultRow],
) -> str:
return (
f"In the following, each IRI is followed by the local name and "
f"optionally its description in parentheses. \n"
f"The RDF graph supports the following node types:\n"
f'{", ".join([self._res_to_str(r, "cls") for r in classes])}\n'
f"The RDF graph supports the following relationships:\n"
f'{", ".join([self._res_to_str(r, "rel") for r in relationships])}\n'
)
if self.standard == "rdf":
clss = self.query(cls_query_rdf)
rels = self.query(rel_query_rdf)
self.schema = _rdf_s_schema(clss, rels)
elif self.standard == "rdfs":
clss = self.query(cls_query_rdfs)
rels = self.query(rel_query_rdfs)
self.schema = _rdf_s_schema(clss, rels)
elif self.standard == "owl":
clss = self.query(cls_query_owl)
ops = self.query(op_query_owl)
dps = self.query(dp_query_owl)
self.schema = (
f"In the following, each IRI is followed by the local name and "
f"optionally its description in parentheses. \n"
f"The OWL graph supports the following node types:\n"
f'{", ".join([self._res_to_str(r, "cls") for r in clss])}\n'
f"The OWL graph supports the following object properties, "
f"i.e., relationships between objects:\n"
f'{", ".join([self._res_to_str(r, "op") for r in ops])}\n'
f"The OWL graph supports the following data properties, "
f"i.e., relationships between objects and literals:\n"
f'{", ".join([self._res_to_str(r, "dp") for r in dps])}\n'
)
else:
raise ValueError(f"Mode '{self.standard}' is currently not supported.")
|