refactor: move metadata extracting code

sync_crawler/models/news.py

@@ -1,6 +1,5 @@
 from collections.abc import Sequence
 from datetime import datetime
-from typing import Any, ClassVar
 
 import pydantic
 
@@ -18,15 +17,6 @@ class News(pydantic.BaseModel):
 
     model_config = pydantic.ConfigDict(extra="forbid")
 
-    excluded_metadata_keys: ClassVar = ["modified_date"]
-
     @property
     def text(self) -> str:
         return " ".join(self.content)
-
-    @property
-    def metadata(self) -> dict[str, Any]:
-        return self.model_dump(
-            mode="json",  # use json mode to convert `modified_date` to string
-            include={"title", "category", "modified_date"},
-        )

sync_crawler/writer/chromadb_writer.py

@@ -41,15 +41,23 @@ def __init__(
             embed_model=HuggingFaceEmbedding(model_name=embedding_model),
         )
 
+    @staticmethod
+    def _get_metadata(news: News):
+        return news.model_dump(include={"title", "category"}) | {
+            "modified_date": news.modified_date.timestamp()
+        }
+
+    excluded_metadata_keys = ["modified_date"]
+
     @override
     def write(self, news_with_id):
         docs = [
             Document(
                 doc_id=str(id_),
                 text=ns.text,
-                extra_info=ns.metadata,
-                excluded_embed_metadata_keys=News.excluded_metadata_keys,
-                excluded_llm_metadata_keys=News.excluded_metadata_keys,
+                extra_info=self._get_metadata(ns),
+                excluded_embed_metadata_keys=self.excluded_metadata_keys,
+                excluded_llm_metadata_keys=self.excluded_metadata_keys,
             )
             for id_, ns in news_with_id
         ]