embed-client 1.0.1.1__tar.gz → 2.0.0.0__tar.gz

{embed_client-1.0.1.1 → embed_client-2.0.0.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: embed-client
-Version: 1.0.1.1
+Version: 2.0.0.0
 Summary: Async client for Embedding Service API
 Author: Your Name
 Requires-Dist: aiohttp

{embed_client-1.0.1.1 → embed_client-2.0.0.0}/embed_client/async_client.py

@@ -48,7 +48,7 @@ class EmbeddingServiceAsyncClient:
     
     Supports both old and new API formats:
     - Old format: {"result": {"success": true, "data": {"embeddings": [...]}}}
-    - New format: {"result": {"success": true, "data": [{"body": "text", "embedding": [...], "chunks": [...]}]}}
+    - New format: {"result": {"success": true, "data": {"embeddings": [...], "results": [{"body": "text", "embedding": [...], "tokens": [...], "bm25_tokens": [...]}]}}}
     
     Args:
         base_url (str): Base URL of the embedding service (e.g., "http://localhost").
@@ -172,13 +172,33 @@ class EmbeddingServiceAsyncClient:
             result: API response dictionary
             
         Returns:
-            List of dictionaries with 'body', 'embedding', and 'chunks' fields
+            List of dictionaries with 'body', 'embedding', 'tokens', and 'bm25_tokens' fields
             
         Raises:
             ValueError: If data cannot be extracted or is in old format
         """
         if "result" in result and isinstance(result["result"], dict):
             res = result["result"]
+            if "data" in res and isinstance(res["data"], dict) and "results" in res["data"]:
+                # New format: result.data.results[]
+                results = res["data"]["results"]
+                if isinstance(results, list):
+                    # Validate that all items have required fields
+                    for i, item in enumerate(results):
+                        if not isinstance(item, dict):
+                            raise ValueError(f"Item {i} is not a dictionary: {item}")
+                        if "body" not in item:
+                            raise ValueError(f"Item {i} missing 'body' field: {item}")
+                        if "embedding" not in item:
+                            raise ValueError(f"Item {i} missing 'embedding' field: {item}")
+                        if "tokens" not in item:
+                            raise ValueError(f"Item {i} missing 'tokens' field: {item}")
+                        if "bm25_tokens" not in item:
+                            raise ValueError(f"Item {i} missing 'bm25_tokens' field: {item}")
+                    
+                    return results
+            
+            # Legacy support for old format: result.data[]
             if "data" in res and isinstance(res["data"], list):
                 # Validate that all items have required fields
                 for i, item in enumerate(res["data"]):
@@ -188,8 +208,9 @@ class EmbeddingServiceAsyncClient:
                         raise ValueError(f"Item {i} missing 'body' field: {item}")
                     if "embedding" not in item:
                         raise ValueError(f"Item {i} missing 'embedding' field: {item}")
-                    if "chunks" not in item:
-                        raise ValueError(f"Item {i} missing 'chunks' field: {item}")
+                    # Old format had 'chunks' instead of 'tokens'
+                    if "chunks" not in item and "tokens" not in item:
+                        raise ValueError(f"Item {i} missing 'chunks' or 'tokens' field: {item}")
                 
                 return res["data"]
         
@@ -214,18 +235,60 @@ class EmbeddingServiceAsyncClient:
     def extract_chunks(self, result: Dict[str, Any]) -> List[List[str]]:
         """
         Extract text chunks from API response (new format only).
+        Note: This method now extracts 'tokens' instead of 'chunks' for compatibility.
         
         Args:
             result: API response dictionary
             
         Returns:
-            List of chunk lists for each text
+            List of token lists for each text
             
         Raises:
             ValueError: If chunks cannot be extracted or is in old format
         """
         data = self.extract_embedding_data(result)
-        return [item["chunks"] for item in data]
+        chunks = []
+        for item in data:
+            # New format uses 'tokens', old format used 'chunks'
+            if "tokens" in item:
+                chunks.append(item["tokens"])
+            elif "chunks" in item:
+                chunks.append(item["chunks"])
+            else:
+                raise ValueError(f"Item missing both 'tokens' and 'chunks' fields: {item}")
+        return chunks
+
+    def extract_tokens(self, result: Dict[str, Any]) -> List[List[str]]:
+        """
+        Extract tokens from API response (new format only).
+        
+        Args:
+            result: API response dictionary
+            
+        Returns:
+            List of token lists for each text
+            
+        Raises:
+            ValueError: If tokens cannot be extracted or is in old format
+        """
+        data = self.extract_embedding_data(result)
+        return [item["tokens"] for item in data]
+
+    def extract_bm25_tokens(self, result: Dict[str, Any]) -> List[List[str]]:
+        """
+        Extract BM25 tokens from API response (new format only).
+        
+        Args:
+            result: API response dictionary
+            
+        Returns:
+            List of BM25 token lists for each text
+            
+        Raises:
+            ValueError: If BM25 tokens cannot be extracted or is in old format
+        """
+        data = self.extract_embedding_data(result)
+        return [item["bm25_tokens"] for item in data]
 
     async def __aenter__(self):
         try:

{embed_client-1.0.1.1 → embed_client-2.0.0.0}/embed_client/example_async_usage_ru.py

@@ -88,9 +88,17 @@ async def main():
                     print("\nAdditional data from new format:")
                     for i, data in enumerate(embedding_data):
                         print(f"  Text: {data['body']!r}")
-                        print(f"  Chunks: {data['chunks']}")
-                except ValueError:
-                    print("(Old format detected - no additional data available)")
+                        print(f"  Tokens: {data['tokens']}")
+                        print(f"  BM25 tokens: {data['bm25_tokens']}")
+                        
+                    # Extract tokens and BM25 tokens separately
+                    tokens = client.extract_tokens(result)
+                    bm25_tokens = client.extract_bm25_tokens(result)
+                    print(f"\nExtracted tokens: {tokens}")
+                    print(f"Extracted BM25 tokens: {bm25_tokens}")
+                    
+                except ValueError as e:
+                    print(f"(Old format detected - no additional data available): {e}")
                     
             except EmbeddingServiceAPIError as e:
                 print("[API error]", e.error)

{embed_client-1.0.1.1 → embed_client-2.0.0.0}/embed_client.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: embed-client
-Version: 1.0.1.1
+Version: 2.0.0.0
 Summary: Async client for Embedding Service API
 Author: Your Name
 Requires-Dist: aiohttp

{embed_client-1.0.1.1 → embed_client-2.0.0.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "embed-client"
-version = "1.0.1.001"
+version = "2.0.0.0"
 description = "Async client for Embedding Service API"
 authors = [{name = "Your Name"}]
 dependencies = [

{embed_client-1.0.1.1 → embed_client-2.0.0.0}/tests/test_async_client.py

@@ -810,13 +810,49 @@ async def test_extract_chunks():
     result = {
         "result": {
             "data": [
-                {"body": "text1", "embedding": [0.1, 0.2], "chunks": ["chunk1", "chunk2"]},
-                {"body": "text2", "embedding": [0.3, 0.4], "chunks": ["chunk3"]}
+                {"body": "text1", "embedding": [0.1, 0.2], "tokens": ["text1"]},
+                {"body": "text2", "embedding": [0.3, 0.4], "tokens": ["text2"]}
             ]
         }
     }
     chunks = client.extract_chunks(result)
-    assert chunks == [["chunk1", "chunk2"], ["chunk3"]]
+    assert chunks == [["text1"], ["text2"]]
+
+@pytest.mark.asyncio
+async def test_extract_tokens():
+    """Test extracting tokens from new format responses."""
+    client = EmbeddingServiceAsyncClient(base_url=BASE_URL, port=PORT)
+    
+    result = {
+        "result": {
+            "data": {
+                "results": [
+                    {"body": "text1", "embedding": [0.1, 0.2], "tokens": ["text1"], "bm25_tokens": ["text1"]},
+                    {"body": "text2", "embedding": [0.3, 0.4], "tokens": ["text2"], "bm25_tokens": ["text2"]}
+                ]
+            }
+        }
+    }
+    tokens = client.extract_tokens(result)
+    assert tokens == [["text1"], ["text2"]]
+
+@pytest.mark.asyncio
+async def test_extract_bm25_tokens():
+    """Test extracting BM25 tokens from new format responses."""
+    client = EmbeddingServiceAsyncClient(base_url=BASE_URL, port=PORT)
+    
+    result = {
+        "result": {
+            "data": {
+                "results": [
+                    {"body": "text1", "embedding": [0.1, 0.2], "tokens": ["text1"], "bm25_tokens": ["text1"]},
+                    {"body": "text2", "embedding": [0.3, 0.4], "tokens": ["text2"], "bm25_tokens": ["text2"]}
+                ]
+            }
+        }
+    }
+    bm25_tokens = client.extract_bm25_tokens(result)
+    assert bm25_tokens == [["text1"], ["text2"]]
 
 @pytest.mark.asyncio
 async def test_close_without_open():