Skip to content
TripleLite

TripleLite

TripleLite is an in-memory RDF triple store for Python. It stores triples in nested dictionaries, with an optional reverse index configurable per predicate.

Install

Section titled “Install”
Terminal window
pip install triplelite

Need rdflib interop? Install the extra:

Terminal window
pip install triplelite[rdflib]

Quick start

Section titled “Quick start”

This example models the article “OpenCitations, an infrastructure organization for open scholarship” using predicates from the OpenCitations Data Model:

from triplelite import TripleLite, RDFTerm


DCTERMS = "http://purl.org/dc/terms/"
FABIO = "http://purl.org/spar/fabio/"
FRBR = "http://purl.org/vocab/frbr/core#"
PRISM = "http://prismstandard.org/namespaces/basic/2.0/"
XSD = "http://www.w3.org/2001/XMLSchema#"
RDF_TYPE = "http://www.w3.org/1999/02/22-rdf-syntax-ns#type"


OC = "https://w3id.org/oc/meta/"
article = f"{OC}br/062501777134"
journal = f"{OC}br/062501778099"


g = TripleLite(identifier=f"{OC}br/")


g.add_many([
    (article, RDF_TYPE, RDFTerm("uri", f"{FABIO}JournalArticle")),
    (article, f"{DCTERMS}title",
     RDFTerm("literal", "OpenCitations, An Infrastructure Organization For Open Scholarship")),
    (article, f"{PRISM}publicationDate",
     RDFTerm("literal", "2020-02", f"{XSD}gYearMonth")),
    (article, f"{PRISM}volume", RDFTerm("literal", "1")),
    (article, f"{PRISM}startingPage", RDFTerm("literal", "428")),
    (article, f"{PRISM}endingPage", RDFTerm("literal", "444")),
    (article, f"{FRBR}partOf", RDFTerm("uri", journal)),
])

Query by pattern (None is a wildcard):

for s, p, o in g.triples((article, None, None)):
    print(p.split("/")[-1], "->", o.value)


for obj in g.objects(article, f"{DCTERMS}title"):
    print(obj.value)


print(len(g))  # 7

Remove by pattern:

g.remove((article, f"{PRISM}startingPage", None))
g.remove((article, f"{PRISM}endingPage", None))
print(len(g))  # 5

Single add vs batch add

Section titled “Single add vs batch add”

add() inserts one triple. add_many() accepts any iterable and avoids per-call overhead by caching internal lookups across the batch. Use add_many() when loading data from a file, a SPARQL result set, or any source with more than a handful of triples:

g.add((article, f"{DCTERMS}title", RDFTerm("literal", "Some title")))


g.add_many(
    (str(s), str(p), RDFTerm("uri", str(o)))
    for s, p, o in some_external_source
)

Both methods deduplicate: adding a triple that already exists is a no-op.

What’s next

Section titled “What’s next”