apunkt/trustgraph
1
0
Fork 0
mirror of https://github.com/trustgraph-ai/trustgraph.git synced 2026-04-26 00:46:22 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Row embeddings APIs exposed (#646)

Browse source 
* Added row embeddings API and CLI support

* Updated protocol specs

* Row embeddings agent tool

* Add new agent tool to CLI
This commit is contained in:
cybermaggedon 2026-02-23 21:52:56 +00:00 • committed by GitHub
parent 1809c1f56d
commit 4bbc6d844f
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
25 changed files with 1090 additions and 29 deletions
Download patch file Download diff file Expand all files Collapse all files

60
trustgraph-base/trustgraph/api/async_flow.py
View file

@ -766,3 +766,63 @@ class AsyncFlowInstance:
request_data.update(kwargs)
return await self.request("rows", request_data)
async def row_embeddings_query(
self, text: str, schema_name: str, user: str = "trustgraph",
collection: str = "default", index_name: Optional[str] = None,
limit: int = 10, **kwargs: Any
):
"""
Query row embeddings for semantic search on structured data.
Performs semantic search over row index embeddings to find rows whose
indexed field values are most similar to the input text. Enables
fuzzy/semantic matching on structured data.
Args:
text: Query text for semantic search
schema_name: Schema name to search within
user: User identifier (default: "trustgraph")
collection: Collection identifier (default: "default")
index_name: Optional index name to filter search to specific index
limit: Maximum number of results to return (default: 10)
**kwargs: Additional service-specific parameters
Returns:
dict: Response containing matches with index_name, index_value,
text, and score
Example:
```python
async_flow = await api.async_flow()
flow = async_flow.id("default")
# Search for customers by name similarity
results = await flow.row_embeddings_query(
text="John Smith",
schema_name="customers",
user="trustgraph",
collection="sales",
limit=5
)
for match in results.get("matches", []):
print(f"{match['index_name']}: {match['index_value']} (score: {match['score']})")
```
"""
# First convert text to embeddings vectors
emb_result = await self.embeddings(text=text)
vectors = emb_result.get("vectors", [])
request_data = {
"vectors": vectors,
"schema_name": schema_name,
"user": user,
"collection": collection,
"limit": limit
}
if index_name:
request_data["index_name"] = index_name
request_data.update(kwargs)
return await self.request("row-embeddings", request_data)

23
trustgraph-base/trustgraph/api/async_socket_client.py
View file

@ -345,3 +345,26 @@ class AsyncSocketFlowInstance:
request.update(kwargs)
return await self.client._send_request("mcp-tool", self.flow_id, request)
async def row_embeddings_query(
self, text: str, schema_name: str, user: str = "trustgraph",
collection: str = "default", index_name: Optional[str] = None,
limit: int = 10, **kwargs
):
"""Query row embeddings for semantic search on structured data"""
# First convert text to embeddings vectors
emb_result = await self.embeddings(text=text)
vectors = emb_result.get("vectors", [])
request = {
"vectors": vectors,
"schema_name": schema_name,
"user": user,
"collection": collection,
"limit": limit
}
if index_name:
request["index_name"] = index_name
request.update(kwargs)
return await self.client._send_request("row-embeddings", self.flow_id, request)

75
trustgraph-base/trustgraph/api/flow.py
View file

@ -1297,3 +1297,78 @@ class FlowInstance:
return response["schema-matches"]
def row_embeddings_query(
self, text, schema_name, user="trustgraph", collection="default",
index_name=None, limit=10
):
"""
Query row data using semantic similarity on indexed fields.
Finds rows whose indexed field values are semantically similar to the
input text, using vector embeddings. This enables fuzzy/semantic matching
on structured data.
Args:
text: Query text for semantic search
schema_name: Schema name to search within
user: User/keyspace identifier (default: "trustgraph")
collection: Collection identifier (default: "default")
index_name: Optional index name to filter search to specific index
limit: Maximum number of results (default: 10)
Returns:
dict: Query results with matches containing index_name, index_value,
text, and score
Example:
```python
flow = api.flow().id("default")
# Search for customers by name similarity
results = flow.row_embeddings_query(
text="John Smith",
schema_name="customers",
user="trustgraph",
collection="sales",
limit=5
)
# Filter to specific index
results = flow.row_embeddings_query(
text="machine learning engineer",
schema_name="employees",
index_name="job_title",
limit=10
)
```
"""
# First convert text to embeddings vectors
emb_result = self.embeddings(text=text)
vectors = emb_result.get("vectors", [])
# Query row embeddings for semantic search
input = {
"vectors": vectors,
"schema_name": schema_name,
"user": user,
"collection": collection,
"limit": limit
}
if index_name:
input["index_name"] = index_name
response = self.request(
"service/row-embeddings",
input
)
# Check for system-level error
if "error" in response and response["error"]:
error_type = response["error"].get("type", "unknown")
error_message = response["error"].get("message", "Unknown error")
raise ProtocolException(f"{error_type}: {error_message}")
return response

70
trustgraph-base/trustgraph/api/socket_client.py
View file

@ -881,3 +881,73 @@ class SocketFlowInstance:
request.update(kwargs)
return self.client._send_request_sync("mcp-tool", self.flow_id, request, False)
def row_embeddings_query(
self,
text: str,
schema_name: str,
user: str = "trustgraph",
collection: str = "default",
index_name: Optional[str] = None,
limit: int = 10,
**kwargs: Any
) -> Dict[str, Any]:
"""
Query row data using semantic similarity on indexed fields.
Finds rows whose indexed field values are semantically similar to the
input text, using vector embeddings. This enables fuzzy/semantic matching
on structured data.
Args:
text: Query text for semantic search
schema_name: Schema name to search within
user: User/keyspace identifier (default: "trustgraph")
collection: Collection identifier (default: "default")
index_name: Optional index name to filter search to specific index
limit: Maximum number of results (default: 10)
**kwargs: Additional parameters passed to the service
Returns:
dict: Query results with matches containing index_name, index_value,
text, and score
Example:
```python
socket = api.socket()
flow = socket.flow("default")
# Search for customers by name similarity
results = flow.row_embeddings_query(
text="John Smith",
schema_name="customers",
user="trustgraph",
collection="sales",
limit=5
)
# Filter to specific index
results = flow.row_embeddings_query(
text="machine learning engineer",
schema_name="employees",
index_name="job_title",
limit=10
)
```
"""
# First convert text to embeddings vectors
emb_result = self.embeddings(text=text)
vectors = emb_result.get("vectors", [])
request = {
"vectors": vectors,
"schema_name": schema_name,
"user": user,
"collection": collection,
"limit": limit
}
if index_name:
request["index_name"] = index_name
request.update(kwargs)
return self.client._send_request_sync("row-embeddings", self.flow_id, request, False)