feat: implement Approximate Nearest Neighbor support for DDL (CREATE …

…TABLE, CREATE VECTOR INDEX) (#124)

* fix(testing+linting): add nox lint+format directives

This change introduces new nox directives:
* blacken: `nox -s blacken`
* format: `nox -s format` to apply formatting to files
* lint: `nox -s lint` to flag linting issues
* unit: to run unit tests locally

which are the basis to enable scalable development
and continuous testing as I prepare to bring in
Approximate Nearest Neighors (ANN) functionality into
this package.

Also while here, fixed a typo in the README.rst file
that didn't have the correct import path.

* feat: add Approximate Nearest Neighbor support to distance strategies

This change adds ANN distance strategies for GoogleSQL semantics.
While here started unit tests to effectively test out components
without having to have a running Cloud Spanner instance.

Updates #94

* Ensure vector fits within limits in sample

* Update ANN query names + test expectations

* Pass in strategy inferred from initialization

* Hook up get_documents_from_query_results

* Link up __search_by_ANN to similarity_search_by_vector

* Incorporate pre_filter and post_filter plus update tests

* Review addressing

* Simplified checking if using ANN

* Reduce the amount of changes

* More reductions

* More reductions to ease code review

* Fit with get_rows_by_similarity_search_ann

* Updates from nox

* Fix PostGreSQL

README.rst

@@ -253,3 +253,9 @@ Disclaimer
 
 This is not an officially supported Google product.
 
+
+Limitations
+-----------
+
+* Approximate Nearest Neighbors (ANN) strategies are only supported for the GoogleSQL dialect
+* ANN's `ALTER VECTOR INDEX` is not yet supported by [Google Cloud Spanner](https://cloud.google.com/spanner/docs/find-approximate-nearest-neighbors#limitations)

noxfile.py

@@ -19,7 +19,6 @@
 import os
 import pathlib
 import shutil
-from pathlib import Path
 
 import nox
 
@@ -33,6 +32,7 @@
     "docfx",
     "docs",
     "format",
+    "integration",
     "lint",
     "unit",
 ]

requirements.txt

@@ -1,4 +1,4 @@
-google-cloud-spanner==3.49.1
-langchain-core==0.3.9
+google-cloud-spanner==3.51.0
+langchain-core==0.3.15
 langchain-community==0.3.1
 pydantic==2.9.1

src/langchain_google_spanner/graph_qa.py

@@ -288,6 +288,7 @@ def _call(
         inputs: Dict[str, Any],
         run_manager: Optional[CallbackManagerForChainRun] = None,
     ) -> Dict[str, str]:
+
         intermediate_steps: List = []
 
         """Generate gql statement, uses it to look up in db and answer question."""

src/langchain_google_spanner/graph_retriever.py

@@ -225,7 +225,7 @@ def __clean_element(self, element: dict[str, Any], embedding_column: str) -> Non
             del element["properties"][embedding_column]
 
     def __get_distance_function(
-        self, distance_strategy=DistanceStrategy.EUCLIDEIAN
+        self, distance_strategy=DistanceStrategy.EUCLIDEAN
     ) -> str:
         """Gets the vector distance function."""
         if distance_strategy == DistanceStrategy.COSINE: