Found duplicated primary key value Summer founders program, which violates the uniqueness constraint of the primary key column.[Bug]:

[andysingal]

Bug Description

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
[<ipython-input-11-9642f26cc04c>](https://localhost:8080/#) in <cell line: 6>()
      4 
      5 # NOTE: can take a while!
----> 6 index = KnowledgeGraphIndex.from_documents(
      7     documents,
      8     max_triplets_per_chunk=2,

8 frames
[/usr/local/lib/python3.10/dist-packages/kuzu/connection.py](https://localhost:8080/#) in execute(self, query, parameters)
    132         else:
    133             prepared_statement = self.prepare(query) if isinstance(query, str) else query
--> 134             _query_result = self._connection.execute(prepared_statement._prepared_statement, parameters)
    135         if not _query_result.isSuccess():
    136             raise RuntimeError(_query_result.getErrorMessage())

RuntimeError: Runtime exception: Found duplicated primary key value Summer founders program, which violates the uniqueness constraint of the primary key column.

Version

0.10.14

Steps to Reproduce

https://docs.llamaindex.ai/en/stable/examples/index_structs/knowledge_graph/KuzuGraphDemo/

only added

!mkdir -p 'data/paul_graham/'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay1.txt'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay2.txt'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay3.txt'

Relevant Logs/Tracbacks

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
[<ipython-input-11-9642f26cc04c>](https://localhost:8080/#) in <cell line: 6>()
      4 
      5 # NOTE: can take a while!
----> 6 index = KnowledgeGraphIndex.from_documents(
      7     documents,
      8     max_triplets_per_chunk=2,

8 frames
[/usr/local/lib/python3.10/dist-packages/kuzu/connection.py](https://localhost:8080/#) in execute(self, query, parameters)
    132         else:
    133             prepared_statement = self.prepare(query) if isinstance(query, str) else query
--> 134             _query_result = self._connection.execute(prepared_statement._prepared_statement, parameters)
    135         if not _query_result.isSuccess():
    136             raise RuntimeError(_query_result.getErrorMessage())

RuntimeError: Runtime exception: Found duplicated primary key value Summer founders program, which violates the uniqueness constraint of the primary key column.

[dosubot]

To resolve the RuntimeError related to a duplicated primary key value "Summer founders program" when creating a KnowledgeGraphIndex from documents using the llama_index package, you need to handle duplicates in the document store. You can use the upsert_triplet_and_node method from the KnowledgeGraphIndex class to handle duplicates when adding nodes and triplets.

Here is an example of how to use this method:

from llama_index.legacy.indices.knowledge_graph.base import KnowledgeGraphIndex
from llama_index.legacy.schema import TextNode
from llama_index.legacy.service_context import ServiceContext

# Create an instance of KnowledgeGraphIndex
index = KnowledgeGraphIndex([], service_context=ServiceContext())

# Define your triplets and nodes
triplets = [
    ("foo", "is", "bar"),
    ("hello", "is not", "world"),
    ("Jane", "is mother of", "Bob"),
]
nodes = [TextNode(text=str(tup)) for tup in triplets]

# Add triplets and nodes to the index
for triplet, node in zip(triplets, nodes):
    index.upsert_triplet_and_node(triplet, node)

This method ensures that duplicates are handled appropriately when adding triplets and nodes to the KnowledgeGraphIndex.

In the provided test cases, the upsert_triplet_and_node method is used to handle duplicates effectively. For example:

from unittest.mock import patch
from llama_index.legacy.indices.knowledge_graph.base import KnowledgeGraphIndex
from llama_index.legacy.schema import TextNode
from llama_index.legacy.service_context import ServiceContext

@patch.object(KnowledgeGraphIndex, "_extract_triplets", side_effect=mock_extract_triplets)
def test_build_kg_manual(_patch_extract_triplets, mock_service_context):
    index = KnowledgeGraphIndex([], service_context=mock_service_context)
    tuples = [
        ("foo", "is", "bar"),
        ("hello", "is not", "world"),
        ("Jane", "is mother of", "Bob"),
    ]
    nodes = [TextNode(text=str(tup)) for tup in tuples]
    for tup, node in zip(tuples, nodes):
        index.upsert_triplet_and_node(tup, node)

    # Verify the nodes and triplets
    docstore_nodes = index.docstore.get_nodes(list(index.index_struct.node_ids))
    table_chunks = {n.get_content() for n in docstore_nodes}
    assert len(table_chunks) == 3
    assert "('foo', 'is', 'bar')" in table_chunks
    assert "('hello', 'is not', 'world')" in table_chunks
    assert "('Jane', 'is mother of', 'Bob')" in table_chunks

    # Test inserting the same node twice
    node = TextNode(text=str(("foo", "is", "bar")), id_="test_node")
    index.upsert_triplet_and_node(tup, node)
    index.upsert_triplet_and_node(tup, node)

This test case demonstrates how the upsert_triplet_and_node method can be used to handle duplicates effectively, ensuring that the RuntimeError related to duplicated primary key values is resolved ^[1][2][3][4].

Details

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.