Pymantic

1.0.1 verified Mon Apr 27 auth: no python

Pymantic is a Semantic Web and RDF library for Python, providing parsers, serializers, and SPARQL client support. The latest version is 1.0.1 (released 2024) with slow release cadence. It supports RDF/XML, Turtle, JSON-LD, TriG, and N-Triples formats.

pip install pymantic

Common errors

error ImportError: cannot import name 'SPARQLClient' from 'pymantic' ↓

cause Old import path `from pymantic import SPARQLClient` no longer works after restructuring in v1.0.0.

fix

Use from pymantic.sparql import SPARQLClient.

error AttributeError: 'Graph' object has no attribute 'parse' ↓

cause `Graph.parse` method requires the graph to be created with a filename or explicitly parsed. In older versions, Graph had different API.

fix

Ensure you are using from pymantic.io import Graph and call graph.parse('file', format='...'). For in-memory, create with Graph() then parse.

Warnings

breaking Version 1.0.0 dropped Python 2.x support and restructured packaging. Import paths changed from old patterns like `from pymantic import sparql`. ↓

fix Use `from pymantic.sparql import SPARQLClient` etc. Update imports per new package structure.

deprecated The `lepl`-based parsers were replaced by `lark` in v0.2.0. Old code relying on `lepl` behavior may break. ↓

fix Ensure you use compatible syntax; lark may have stricter parsing. Update Turtle files if needed.

gotcha Parsing JSON-LD requires the `json-ld` optional dependency? Actually the library supports JSON-LD but may need `PyYAML` or extra care. Check documentation for exact parsing options (e.g., base URI). ↓

fix Install `PyYAML` if needed: `pip install PyYAML`. Use `graph.parse('file.jsonld', format='json-ld')`.

Imports

SPARQLClient

from pymantic.sparql import SPARQLClient

Graph
```
from pymantic.io import Graph
```
RDFXMLParser
```
from pymantic.io import RDFXMLParser
```
TurtleParser
```
from pymantic.io import TurtleParser
```
JSONLDParser
```
from pymantic.io import JSONLDParser
```
TriGSerializer
```
from pymantic.io import TriGSerializer
```

Quickstart

Basic usage: parse a Turtle file and query a SPARQL endpoint.

from pymantic.io import Graph
from pymantic.sparql import SPARQLClient

# Create a graph and parse a Turtle file
graph = Graph()
graph.parse('example.ttl', format='turtle')
print(f"Parsed {len(graph.triples)} triples")

# Query a SPARQL endpoint
client = SPARQLClient('http://dbpedia.org/sparql')
results = client.query('SELECT * WHERE {?s ?p ?o} LIMIT 10')
for result in results['results']['bindings']:
    print(result)