fix(bedrock): fix default chunking strategy not deploying (#481)

awslabs · May 29, 2024 · 169bc69 · 169bc69
1 parent f749463
commit 169bc69
Show file tree

Hide file tree

Showing 4 changed files with 48 additions and 3 deletions.
diff --git a/apidocs/enums/bedrock.ChunkingStrategy.md b/apidocs/enums/bedrock.ChunkingStrategy.md
@@ -25,6 +25,8 @@ splitting them up such that each file corresponds to a chunk.
 • **DEFAULT** = ``"DEFAULT"``
 
 `FIXED_SIZE` with the default chunk size of 300 tokens and 20% overlap.
+If default is selected, chunk size and overlap set by the user will be
+ignored.
 
 ___
 

diff --git a/src/cdk-lib/bedrock/s3-data-source.ts b/src/cdk-lib/bedrock/s3-data-source.ts
@@ -37,6 +37,8 @@ export enum ChunkingStrategy {
   FIXED_SIZE = 'FIXED_SIZE',
   /**
    * `FIXED_SIZE` with the default chunk size of 300 tokens and 20% overlap.
+   * If default is selected, chunk size and overlap set by the user will be
+   * ignored.
    */
   DEFAULT = 'DEFAULT',
   /**
@@ -202,8 +204,16 @@ function vectorIngestionConfiguration(
       },
     };
 
-  } else {
-    return {};
+  } else { // DEFAULT
+    return {
+      chunkingConfiguration: {
+        chunkingStrategy: ChunkingStrategy.FIXED_SIZE,
+        fixedSizeChunkingConfiguration: {
+          maxTokens: CHUNKING_MAX_TOKENS,
+          overlapPercentage: CHUNKING_OVERLAP,
+        },
+      },
+    };
   }
 
 }
diff --git a/test/cdk-lib/bedrock/__snapshots__/agent.test.ts.snap b/test/cdk-lib/bedrock/__snapshots__/agent.test.ts.snap
diff --git a/test/cdk-lib/bedrock/s3-data-source.test.ts b/test/cdk-lib/bedrock/s3-data-source.test.ts
@@ -68,6 +68,31 @@ describe('S3 Data Source', () => {
     });
   });
 
+  test('Default chunking', () => {
+    new bedrock.S3DataSource(stack, 'TestDataSource', {
+      bucket,
+      knowledgeBase: kb,
+      dataSourceName: 'TestDataSource',
+      chunkingStrategy: bedrock.ChunkingStrategy.DEFAULT,
+      maxTokens: 1024,
+      overlapPercentage: 20,
+    });
+
+    Template.fromStack(stack).hasResourceProperties('AWS::Bedrock::DataSource', {
+
+      VectorIngestionConfiguration:
+       {
+         ChunkingConfiguration: {
+           ChunkingStrategy: 'FIXED_SIZE',
+           FixedSizeChunkingConfiguration: {
+             MaxTokens: 300,
+             OverlapPercentage: 20,
+           },
+         },
+       },
+    });
+  });
+
   test('No chunking', () => {
     new bedrock.S3DataSource(stack, 'TestDataSource', {
       bucket,