wizit-context-ingestor 0.3.0b1__tar.gz → 0.3.0b3__tar.gz

{wizit_context_ingestor-0.3.0b1 → wizit_context_ingestor-0.3.0b3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: wizit-context-ingestor
-Version: 0.3.0b1
+Version: 0.3.0b3
 Summary: Contextual Rag with Cloud Solutions
 Requires-Dist: anthropic[vertex]>=0.66.0
 Requires-Dist: boto3>=1.40.23

{wizit_context_ingestor-0.3.0b1 → wizit_context_ingestor-0.3.0b3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "wizit_context_ingestor"
-version = "0.3.0-beta-1"
+version = "0.3.0-beta-3"
 description = "Contextual Rag with Cloud Solutions"
 readme = "README.md"
 requires-python = ">=3.12"

{wizit_context_ingestor-0.3.0b1 → wizit_context_ingestor-0.3.0b3}/src/wizit_context_ingestor/application/transcription_service.py

@@ -146,7 +146,7 @@ class TranscriptionService:
         if result["transcription"]:
             document.page_text = result["transcription"]
         else:
-            raise ValueError("No transcription found")
+            raise ValueError(f"No transcription found: {result} ")
         return document
 
     def process_document(self, file_key: str) -> Tuple[List[ParsedDocPage], ParsedDoc]:

{wizit_context_ingestor-0.3.0b1 → wizit_context_ingestor-0.3.0b3}/src/wizit_context_ingestor/infra/rag/chroma_embeddings.py

@@ -1,10 +1,6 @@
-from typing_extensions import Sequence
-from test.test_typing import CoolEmployee
 from langchain_core.documents import Document
 from langchain_chroma import Chroma
-from typing import List
 import logging
-from uuid import uuid4
 from ...application.interfaces import EmbeddingsManager
 
 # load_dotenv()

{wizit_context_ingestor-0.3.0b1 → wizit_context_ingestor-0.3.0b3}/src/wizit_context_ingestor/infra/secrets/aws_secrets_manager.py

@@ -3,11 +3,10 @@ import logging
 
 logger = logging.getLogger(__name__)
 
-class AwsSecretsManager:
-
-    def __init__(self):
-        self.client = boto3_client('secretsmanager')
 
+class AwsSecretsManager:
+    def __init__(self, aws_region="us-east-1"):
+        self.client = boto3_client("secretsmanager", region_name=aws_region)
 
     def get_secret(self, secret_name):
         """

{wizit_context_ingestor-0.3.0b1 → wizit_context_ingestor-0.3.0b3}/src/wizit_context_ingestor/main.py

@@ -12,6 +12,7 @@ from .infra.secrets.aws_secrets_manager import AwsSecretsManager
 from .data.storage import storage_services, StorageServices
 from .data.kdb import kdb_services, KdbServices
 from .utils.file_utils import has_invalid_file_name_format
+from langsmith import Client, tracing_context
 
 
 class KdbManager:
@@ -69,6 +70,8 @@ class TranscriptionManager:
         gcp_project_id: str,
         gcp_project_location: str,
         gcp_secret_name: str,
+        langsmith_api_key: str,
+        langsmith_project_name: str,
         storage_service: storage_services,
         source_storage_route: str,
         target_storage_route: str,
@@ -94,6 +97,9 @@ class TranscriptionManager:
         self.max_transcription_retries = max_transcription_retries
         self.gcp_sa_dict = self._get_gcp_sa_dict(gcp_secret_name)
         self.vertex_model = self._get_vertex_model()
+        self.langsmith_api_key = langsmith_api_key
+        self.langsmith_project_name = langsmith_project_name
+        self.langsmith_client = Client(api_key=self.langsmith_api_key)
 
     def _get_gcp_sa_dict(self, gcp_secret_name: str):
         vertex_gcp_sa = self.aws_secrets_manager.get_secret(gcp_secret_name)
@@ -109,6 +115,18 @@ class TranscriptionManager:
         )
         return vertex_model
 
+    def tracing(func):
+        def gen_tracing_context(self, *args, **kwargs):
+            with tracing_context(
+                enabled=True,
+                project_name=self.langsmith_project_name,
+                client=self.langsmith_client,
+            ):
+                return func(self, *args, **kwargs)
+
+        return gen_tracing_context
+
+    @tracing
     def transcribe_document(self, file_key: str):
         """Transcribe a document from source storage to target storage.
         This method serves as a generic interface for transcribing documents from
@@ -171,6 +189,8 @@ class ChunksManager:
         gcp_project_id: str,
         gcp_project_location: str,
         gcp_secret_name: str,
+        langsmith_api_key: str,
+        langsmith_project_name: str,
         storage_service: storage_services,
         kdb_service: Literal["redis", "chroma"],
         kdb_params: Dict[Any, Any],
@@ -188,11 +208,13 @@ class ChunksManager:
         self.storage_service = storage_service
         self.kdb_params = kdb_params
         self.kdb_service = kdb_service
-        # self.redis_connection_string = redis_connection_string
         self.vertex_model = self._get_vertex_model()
         self.embeddings_model = self.vertex_model.load_embeddings_model(
             embeddings_model_id
         )
+        self.langsmith_api_key = langsmith_api_key
+        self.langsmith_project_name = langsmith_project_name
+        self.langsmith_client = Client(api_key=self.langsmith_api_key)
 
     def _get_gcp_sa_dict(self, gcp_secret_name: str):
         vertex_gcp_sa = self.aws_secrets_manager.get_secret(gcp_secret_name)
@@ -208,6 +230,18 @@ class ChunksManager:
         )
         return vertex_model
 
+    def tracing(func):
+        def gen_tacing_context(self, *args, **kwargs):
+            with tracing_context(
+                enabled=True,
+                project_name=self.langsmith_project_name,
+                client=self.langsmith_client,
+            ):
+                return func(self, *args, **kwargs)
+
+        return gen_tacing_context
+
+    @tracing
     def gen_context_chunks(
         self, file_key: str, source_storage_route: str, target_storage_route: str
     ):