fix(server): exclude outdated doc id style in embedding count (#13269)

fix AI-392
fix AI-393

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->

## Summary by CodeRabbit

* **New Features**
* Improved filtering of outdated document ID styles in embedding status
reporting, ensuring more accurate counts of embedded documents.
* Stricter rate limiting applied to workspace embedding status queries
for enhanced system reliability.

* **Bug Fixes**
* Resolved issues with duplicate or outdated document IDs affecting
embedding status totals.

<!-- end of auto-generated comment: release notes by coderabbit.ai -->

packages/backend/server/src/__tests__/models/__snapshots__/copilot-workspace.spec.ts.md

@@ -89,3 +89,19 @@ Generated by [AVA](https://avajs.dev).
 > should not find docs to embed
 
     0
+
+## should filter outdated doc id style in embedding status
+
+> should include modern doc format
+
+    {
+      embedded: 0,
+      total: 1,
+    }
+
+> should count docs after filtering outdated
+
+    {
+      embedded: 1,
+      total: 1,
+    }

packages/backend/server/src/__tests__/models/copilot-workspace.spec.ts

@@ -306,3 +306,50 @@ test('should check embedding table', async t => {
   //   t.false(ret, 'should return false when embedding table is not available');
   // }
 });
+
+test('should filter outdated doc id style in embedding status', async t => {
+  const docId = randomUUID();
+  const outdatedDocId = `${workspace.id}:space:${docId}`;
+
+  await t.context.doc.upsert({
+    spaceId: workspace.id,
+    docId,
+    blob: Uint8Array.from([1, 2, 3]),
+    timestamp: Date.now(),
+    editorId: user.id,
+  });
+
+  await t.context.doc.upsert({
+    spaceId: workspace.id,
+    docId: outdatedDocId,
+    blob: Uint8Array.from([1, 2, 3]),
+    timestamp: Date.now(),
+    editorId: user.id,
+  });
+
+  {
+    const status = await t.context.copilotWorkspace.getEmbeddingStatus(
+      workspace.id
+    );
+    t.snapshot(status, 'should include modern doc format');
+  }
+
+  {
+    await t.context.copilotContext.insertWorkspaceEmbedding(
+      workspace.id,
+      docId,
+      [
+        {
+          index: 0,
+          content: 'content',
+          embedding: Array.from({ length: 1024 }, () => 1),
+        },
+      ]
+    );
+
+    const status = await t.context.copilotWorkspace.getEmbeddingStatus(
+      workspace.id
+    );
+    t.snapshot(status, 'should count docs after filtering outdated');
+  }
+});

packages/backend/server/src/models/copilot-workspace.ts

@@ -152,7 +152,7 @@ export class CopilotWorkspaceConfigModel extends BaseModel {
   }
 
   @Transactional()
-  async getWorkspaceEmbeddingStatus(workspaceId: string) {
+  async getEmbeddingStatus(workspaceId: string) {
     const ignoredDocIds = (await this.listIgnoredDocIds(workspaceId)).map(
       d => d.docId
     );
@@ -168,9 +168,13 @@ export class CopilotWorkspaceConfigModel extends BaseModel {
     };
 
     const [docTotal, docEmbedded, fileTotal, fileEmbedded] = await Promise.all([
-      this.db.snapshot.count({ where: snapshotCondition }),
-      this.db.snapshot.count({
+      this.db.snapshot.findMany({
+        where: snapshotCondition,
+        select: { id: true },
+      }),
+      this.db.snapshot.findMany({
         where: { ...snapshotCondition, embedding: { some: {} } },
+        select: { id: true },
       }),
       this.db.aiWorkspaceFiles.count({ where: { workspaceId } }),
       this.db.aiWorkspaceFiles.count({
@@ -178,9 +182,23 @@ export class CopilotWorkspaceConfigModel extends BaseModel {
       }),
     ]);
 
+    const docTotalIds = docTotal.map(d => d.id);
+    const docTotalSet = new Set(docTotalIds);
+    const outdatedDocPrefix = `${workspaceId}:space:`;
+    const duplicateOutdatedDocSet = new Set(
+      docTotalIds
+        .filter(id => id.startsWith(outdatedDocPrefix))
+        .filter(id => docTotalSet.has(id.slice(outdatedDocPrefix.length)))
+    );
+
     return {
-      total: docTotal + fileTotal,
-      embedded: docEmbedded + fileEmbedded,
+      total:
+        docTotalIds.filter(id => !duplicateOutdatedDocSet.has(id)).length +
+        fileTotal,
+      embedded:
+        docEmbedded
+          .map(d => d.id)
+          .filter(id => !duplicateOutdatedDocSet.has(id)).length + fileEmbedded,
     };
   }
 

packages/backend/server/src/plugins/copilot/context/resolver.ts

@@ -356,6 +356,7 @@ export class CopilotContextRootResolver {
     return false;
   }
 
+  @Throttle('strict')
   @Query(() => ContextWorkspaceEmbeddingStatus, {
     description: 'query workspace embedding status',
   })
@@ -372,9 +373,7 @@ export class CopilotContextRootResolver {
 
     if (this.context.canEmbedding) {
       const { total, embedded } =
-        await this.models.copilotWorkspace.getWorkspaceEmbeddingStatus(
-          workspaceId
-        );
+        await this.models.copilotWorkspace.getEmbeddingStatus(workspaceId);
       return { total, embedded };
     }
 

packages/backend/server/src/plugins/copilot/workspace/resolver.ts

@@ -103,6 +103,7 @@ export class CopilotWorkspaceEmbeddingConfigResolver {
 
     return ignoredDocs;
   }
+
   @Mutation(() => Number, {
     name: 'updateWorkspaceEmbeddingIgnoredDocs',
     complexity: 2,