From db5c57a21ac3ff77aa760f9d38914aa061d6b6b0 Mon Sep 17 00:00:00 2001
From: Ali Salimli <67149699+elisalimli@users.noreply.github.com>
Date: Sat, 2 Mar 2024 09:05:39 +0400
Subject: [PATCH] fix: Qdrant's delete method (#73)

* fix: qdrant delete method

* chore: update .env.example

* Fix formatting

---------

Co-authored-by: Ismail Pelaseyed <homanp@gmail.com>
---
 .env.example        |  3 +++
 .gitignore          |  1 +
 vectordbs/qdrant.py | 25 +++++++++++++++++++++++++
 3 files changed, 29 insertions(+)

diff --git a/.env.example b/.env.example
index ab88b0e1..31ff8a25 100644
--- a/.env.example
+++ b/.env.example
@@ -16,3 +16,6 @@ PINECONE_API_KEY=
 PINECONE_HOST=
 PINECONE_INDEX=
 
+# Unstructured API
+UNSTRUCTURED_IO_API_KEY=
+UNSTRUCTURED_IO_SERVER_URL=
diff --git a/.gitignore b/.gitignore
index 9e1098b5..c1337c83 100644
--- a/.gitignore
+++ b/.gitignore
@@ -2,3 +2,4 @@ venv
 .venv
 .env
 __pycache__/
+.DS_Store
\ No newline at end of file
diff --git a/vectordbs/qdrant.py b/vectordbs/qdrant.py
index 5e3853e9..7d43dc36 100644
--- a/vectordbs/qdrant.py
+++ b/vectordbs/qdrant.py
@@ -5,6 +5,7 @@
 from semantic_router.encoders import BaseEncoder
 from tqdm import tqdm
 
+from models.delete import DeleteResponse
 from models.document import BaseDocumentChunk
 from vectordbs.base import BaseVectorDatabase
 
@@ -89,6 +90,28 @@ async def query(self, input: str, top_k: int = MAX_QUERY_TOP_K) -> List:
         ]
 
     async def delete(self, file_url: str) -> None:
+        #         client.count(
+        #     collection_name="{collection_name}",
+        #     count_filter=models.Filter(
+        #         must=[
+        #             models.FieldCondition(key="color", match=models.MatchValue(value="red")),
+        #         ]
+        #     ),
+        #     exact=True,
+        # )
+
+        deleted_chunks = self.client.count(
+            collection_name=self.index_name,
+            count_filter=rest.Filter(
+                must=[
+                    rest.FieldCondition(
+                        key="file_url", match=rest.MatchValue(value=file_url)
+                    )
+                ]
+            ),
+            exact=True,
+        )
+
         self.client.delete(
             collection_name=self.index_name,
             points_selector=rest.FilterSelector(
@@ -101,3 +124,5 @@ async def delete(self, file_url: str) -> None:
                 )
             ),
         )
+
+        return DeleteResponse(num_of_deleted_chunks=deleted_chunks.count)