chore: Use a hardcoded constant for the maximum upper bounds instead of config.maxDocumentsPerQuery

Author: kmruiz

src/tools/mongodb/metadata/collectionSchema.ts

@@ -8,13 +8,21 @@ import { ONE_MB } from "../../../helpers/constants.js";
 import { collectCursorUntilMaxBytesLimit } from "../../../helpers/collectCursorUntilMaxBytes.js";
 import { isObjectEmpty } from "../../../helpers/isObjectEmpty.js";
 
+const MAXIMUM_SAMPLE_SIZE_HARD_LIMIT = 50_000;
+
 export class CollectionSchemaTool extends MongoDBToolBase {
     public name = "collection-schema";
     protected description = "Describe the schema for a collection";
     protected argsShape = {
         ...DbOperationArgs,
         sampleSize: z.number().optional().default(50).describe("Number of documents to sample for schema inference"),
-        responseBytesLimit: z.number().optional().default(ONE_MB).describe(`The maximum number of bytes to return in the response. This value is capped by the server’s configured maxBytesPerQuery and cannot be exceeded.`),
+        responseBytesLimit: z
+            .number()
+            .optional()
+            .default(ONE_MB)
+            .describe(
+                `The maximum number of bytes to return in the response. This value is capped by the server’s configured maxBytesPerQuery and cannot be exceeded.`
+            ),
     };
 
     public operationType: OperationType = "metadata";
@@ -24,7 +32,9 @@ export class CollectionSchemaTool extends MongoDBToolBase {
         { signal }: ToolExecutionContext
     ): Promise<CallToolResult> {
         const provider = await this.ensureConnected();
-        const cursor = provider.aggregate(database, collection, [{ $sample: { size: Math.min(sampleSize, this.config.maxDocumentsPerQuery) } }]);
+        const cursor = provider.aggregate(database, collection, [
+            { $sample: { size: Math.min(sampleSize, MAXIMUM_SAMPLE_SIZE_HARD_LIMIT) } },
+        ]);
         const { cappedBy, documents } = await collectCursorUntilMaxBytesLimit({
             cursor,
             configuredMaxBytesPerQuery: this.config.maxBytesPerQuery,
@@ -45,13 +55,13 @@ export class CollectionSchemaTool extends MongoDBToolBase {
         }
 
         const header = `Found ${fieldsCount} fields in the schema for "${database}.${collection}"`;
-        const cappedWarning = cappedBy !== undefined ? `\nThe schema was inferred from a subset of documents due to the response size limit. (${cappedBy})` : "";
+        const cappedWarning =
+            cappedBy !== undefined
+                ? `\nThe schema was inferred from a subset of documents due to the response size limit. (${cappedBy})`
+                : "";
 
         return {
-            content: formatUntrustedData(
-                `${header}${cappedWarning}`,
-                JSON.stringify(schema)
-            ),
+            content: formatUntrustedData(`${header}${cappedWarning}`, JSON.stringify(schema)),
         };
     }
 }

tests/integration/common/isObjectEmpty.test.ts

@@ -5,7 +5,7 @@ describe("isObjectEmpty", () => {
     it("returns true for null", () => {
         expect(isObjectEmpty(null)).toBe(true);
     });
-    
+
     it("returns true for undefined", () => {
         expect(isObjectEmpty(undefined)).toBe(true);
     });

tests/integration/tools/mongodb/metadata/collectionSchema.test.ts

@@ -15,26 +15,21 @@ import type { SimplifiedSchema } from "mongodb-schema";
 import { describe, expect, it } from "vitest";
 
 describeWithMongoDB("collectionSchema tool", (integration) => {
-    validateToolMetadata(
-        integration,
-        "collection-schema",
-        "Describe the schema for a collection",
-        [
-            ...databaseCollectionParameters,
-            {
-                name: "sampleSize",
-                type: "number",
-                description: "Number of documents to sample for schema inference",
-                required: false, 
-            },
-            {
-                name: "responseBytesLimit",
-                type: "number",
-                description: `The maximum number of bytes to return in the response. This value is capped by the server’s configured maxBytesPerQuery and cannot be exceeded.`,
-                required: false, 
-            }
-        ]
-    );
+    validateToolMetadata(integration, "collection-schema", "Describe the schema for a collection", [
+        ...databaseCollectionParameters,
+        {
+            name: "sampleSize",
+            type: "number",
+            description: "Number of documents to sample for schema inference",
+            required: false,
+        },
+        {
+            name: "responseBytesLimit",
+            type: "number",
+            description: `The maximum number of bytes to return in the response. This value is capped by the server’s configured maxBytesPerQuery and cannot be exceeded.`,
+            required: false,
+        },
+    ]);
 
     validateThrowsForInvalidArguments(integration, "collection-schema", databaseCollectionInvalidArgs);