From 3827c47515960aefff36bff443426fa80e708894 Mon Sep 17 00:00:00 2001
From: Stephen Hu <stephenhu@seismic.com>
Date: Thu, 8 May 2025 12:24:38 +0800
Subject: [PATCH] Feat: Add API to support get chunk by id (#7522)

### What problem does this PR solve?
https://github.com/infiniflow/ragflow/issues/7519
### Type of change
- [x] New Feature (non-breaking change which adds functionality)

---------

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
---
 api/apps/api_app.py | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

diff --git a/api/apps/api_app.py b/api/apps/api_app.py
index 5376ef0ed..27e1606de 100644
--- a/api/apps/api_app.py
+++ b/api/apps/api_app.py
@@ -345,7 +345,7 @@ def completion():
 
 @manager.route('/conversation/<conversation_id>', methods=['GET'])  # noqa: F821
 # @login_required
-def get(conversation_id):
+def get_conversation(conversation_id):
     token = request.headers.get('Authorization').split()[1]
     objs = APIToken.query(token=token)
     if not objs:
@@ -548,6 +548,31 @@ def list_chunks():
 
     return get_json_result(data=res)
 
+@manager.route('/get_chunk/<chunk_id>', methods=['GET'])  # noqa: F821
+# @login_required
+def get_chunk(chunk_id):
+    from rag.nlp import search
+    token = request.headers.get('Authorization').split()[1]
+    objs = APIToken.query(token=token)
+    if not objs:
+        return get_json_result(
+            data=False, message='Authentication error: API key is invalid!"', code=settings.RetCode.AUTHENTICATION_ERROR)
+    try:
+        tenant_id = objs[0].tenant_id
+        kb_ids = KnowledgebaseService.get_kb_ids(tenant_id)
+        chunk = settings.docStoreConn.get(chunk_id, search.index_name(tenant_id), kb_ids)
+        if chunk is None:
+            return server_error_response(Exception("Chunk not found"))
+        k = []
+        for n in chunk.keys():
+            if re.search(r"(_vec$|_sm_|_tks|_ltks)", n):
+                k.append(n)
+        for n in k:
+            del chunk[n]
+
+        return get_json_result(data=chunk)
+    except Exception as e:
+        return server_error_response(e)
 
 @manager.route('/list_kb_docs', methods=['POST'])  # noqa: F821
 # @login_required