chore(server): improve ignored docs list & match (#12307)

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

- **Bug Fixes**
	- Improved the accuracy of document matching by excluding ignored documents from search results.
- **Chores**
	- Updated internal handling of ignored document lists for better consistency and reliability.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

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

@@ -24,6 +24,10 @@ Generated by [AVA](https://avajs.dev).
       'doc1',
     ]
 
+> should not change if ignored doc exists
+
+    0
+
 > should not add ignored doc again
 
     [
@@ -34,7 +38,7 @@ Generated by [AVA](https://avajs.dev).
 
 > should add new ignored doc
 
-    2
+    1
 
 > should add ignored doc
 

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

@@ -74,7 +74,7 @@ test('should manage copilot workspace ignored docs', async t => {
       workspace.id,
       [docId]
     );
-    t.is(count, 1, 'should not add ignored doc again');
+    t.snapshot(count, 'should not change if ignored doc exists');
 
     const ret = await t.context.copilotWorkspace.listIgnoredDocs(workspace.id);
     t.snapshot(cleanObject(ret), 'should not add ignored doc again');

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

@@ -1,7 +1,6 @@
 import { randomUUID } from 'node:crypto';
 
 import { Injectable } from '@nestjs/common';
-import { Transactional } from '@nestjs-cls/transactional';
 import { Prisma } from '@prisma/client';
 
 import { CopilotSessionNotFound } from '../base';
@@ -218,7 +217,6 @@ export class CopilotContextModel extends BaseModel {
     });
   }
 
-  @Transactional()
   async matchWorkspaceEmbedding(
     embedding: number[],
     workspaceId: string,
@@ -227,25 +225,23 @@ export class CopilotContextModel extends BaseModel {
     matchDocIds?: string[]
   ): Promise<DocChunkSimilarity[]> {
     const similarityChunks = await this.db.$queryRaw<Array<DocChunkSimilarity>>`
-      SELECT "doc_id" as "docId", "chunk", "content", "embedding" <=> ${embedding}::vector as "distance"
-      FROM "ai_workspace_embeddings"
-      WHERE "workspace_id" = ${workspaceId}
-      ${matchDocIds?.length ? Prisma.sql`AND "doc_id" IN (${Prisma.join(matchDocIds)})` : Prisma.empty}
+      SELECT
+        w."doc_id" as "docId",
+        w."chunk",
+        w."content",
+        w."embedding" <=> ${embedding}::vector as "distance"
+      FROM "ai_workspace_embeddings" w
+      LEFT JOIN "ai_workspace_ignored_docs" i
+        ON i."workspace_id" = w."workspace_id"
+          AND i."doc_id" = w."doc_id"
+          ${matchDocIds?.length ? Prisma.sql`AND w."doc_id" NOT IN (${Prisma.join(matchDocIds)})` : Prisma.empty}
+      WHERE
+        w."workspace_id" = ${workspaceId}
+        AND i."doc_id" IS NULL
       ORDER BY "distance" ASC
       LIMIT ${topK};
     `;
 
-    const matchedChunks = similarityChunks.filter(
-      c => Number(c.distance) <= threshold
-    );
-    const matchedDocIds = Array.from(new Set(matchedChunks.map(c => c.docId)));
-    if (!matchDocIds?.length && matchedDocIds.length) {
-      const ignoredDocs = await this.models.copilotWorkspace.checkIgnoredDocs(
-        workspaceId,
-        matchedDocIds
-      );
-      return matchedChunks.filter(c => !ignoredDocs.includes(c.docId));
-    }
-    return matchedChunks;
+    return similarityChunks.filter(c => Number(c.distance) <= threshold);
   }
 }

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

@@ -16,7 +16,6 @@ import type {
 
 @Injectable()
 export class CopilotWorkspaceConfigModel extends BaseModel {
-  @Transactional()
   private async listIgnoredDocIds(
     workspaceId: string,
     options?: PaginationInput
@@ -47,16 +46,15 @@ export class CopilotWorkspaceConfigModel extends BaseModel {
     );
     const added = add.filter(id => !ignored.has(id));
 
-    if (added.length) {
+    const { count: addedCount } =
       await this.db.aiWorkspaceIgnoredDocs.createMany({
         data: added.map(docId => ({
           workspaceId,
           docId,
         })),
       });
-    }
 
-    if (removed.size) {
+    const { count: removedCount } =
       await this.db.aiWorkspaceIgnoredDocs.deleteMany({
         where: {
           workspaceId,
@@ -65,11 +63,11 @@ export class CopilotWorkspaceConfigModel extends BaseModel {
           },
         },
       });
-    }
 
-    return added.length + ignored.size;
+    return addedCount + removedCount;
   }
 
+  @Transactional()
   async listIgnoredDocs(
     workspaceId: string,
     options?: PaginationInput
@@ -100,6 +98,7 @@ export class CopilotWorkspaceConfigModel extends BaseModel {
     });
   }
 
+  @Transactional()
   async countIgnoredDocs(workspaceId: string): Promise<number> {
     const count = await this.db.aiWorkspaceIgnoredDocs.count({
       where: {