Added RSS Feeds for Medium Articles and Individual Publications

.env.example

@@ -1,4 +1,4 @@
-SUPABASE_DB__TABLE_NAME=substack_articles
+SUPABASE_DB__TABLE_NAME=feed_articles
 SUPABASE_DB__HOST=your_supabase_db_host_here
 SUPABASE_DB__NAME=postgres
 SUPABASE_DB__USER=your_supabase_db_user_here
@@ -12,7 +12,7 @@ RSS__BATCH_SIZE=30
 # Qdrant configurationbatch
 QDRANT__API_KEY=your_qdrant_api_key_here
 QDRANT__URL=your_qdrant_url_here
-QDRANT__COLLECTION_NAME=substack_collection
+QDRANT__COLLECTION_NAME=feed_collection
 QDRANT__DENSE_MODEL_NAME=BAAI/bge-base-en-v1.5 # BAAI/bge-large-en-v1.5 (1024), BAAI/bge-base-en-v1.5 (HF, 768). BAAI/bge-base-en (Fastembed, 768)
 QDRANT__SPARSE_MODEL_NAME=Qdrant/bm25 # prithivida/Splade_PP_en_v1, Qdrant/bm25
 QDRANT__VECTOR_DIM=768 # 768, 1024
@@ -49,7 +49,7 @@ OPENROUTER__API_URL=https://openrouter.ai/api/v1
 
 # OPIK OBSERVABILITY
 OPIK__API_KEY=your_opik_api_key_here
-OPIK__PROJECT_NAME=substack-pipeline
+OPIK__PROJECT_NAME=feed-pipeline
 
 # FastAPI Endpoint
 BACKEND_URL=your_fastapi_backend_url_here

README.md

@@ -12,7 +12,7 @@ pinned: false
 
 # Articles Search Engine
 
-A compact, production-style RAG pipeline. It ingests Substack RSS articles, stores them in Postgres (Supabase), creates dense/sparse embeddings in Qdrant, and exposes search and answer endpoints via FastAPI with a simple Gradio UI.
+A compact, production-style RAG pipeline. It ingests Substack, Medium and top publications RSS articles, stores them in Postgres (Supabase), creates dense/sparse embeddings in Qdrant, and exposes search and answer endpoints via FastAPI with a simple Gradio UI.
 
 ## How it works (brief)
 - Ingest RSS → Supabase:
@@ -23,6 +23,7 @@ A compact, production-style RAG pipeline. It ingests Substack RSS articles, stor
 - Search + generate:
   - FastAPI (`src/api/main.py`) exposes search endpoints (keyword, semantic, hybrid) and assembles answers with citations.
   - LLM providers are pluggable with fallback (OpenRouter, OpenAI, Hugging Face).
+  - Opik is used for Evaluation
 - UI + deploy:
   - Gradio app for quick local search (`frontend/app.py`).
   - Containerization with Docker and optional deploy to Google Cloud Run.
@@ -31,7 +32,7 @@ A compact, production-style RAG pipeline. It ingests Substack RSS articles, stor
 - Python 3.12, FastAPI, Prefect, SQLAlchemy
 - Supabase (Postgres) for articles
 - Qdrant for vector search (dense + sparse/hybrid)
-- OpenRouter / OpenAI / Hugging Face for LLM completion
+- OpenRouter / OpenAI / Hugging Face for LLM completion, Opik for LLM Evaluation
 - Gradio UI, Docker, Google Cloud Run
 - Config via Pydantic Settings, `uv` or `pip` for deps
 

cloudbuild_fastapi.yaml

@@ -7,6 +7,6 @@ steps:
       export DOCKER_BUILDKIT=1
       docker build -t gcr.io/${PROJECT_ID}/${_SERVICE_NAME} -f Dockerfile .
 substitutions:
-  _SERVICE_NAME: "substack-pipeline-fastapi"
+  _SERVICE_NAME: "feed-pipeline-fastapi"
 images:
   - "gcr.io/${PROJECT_ID}/${_SERVICE_NAME}"

deploy_fastapi.sh

@@ -26,7 +26,7 @@ echo "✅ Environment variables loaded."
 # Configuration
 # -----------------------
 PROJECT_ID="personal-projects-477710"
-SERVICE_NAME="substack-pipeline-fastapi"
+SERVICE_NAME="feed-pipeline-fastapi"
 REGION="asia-south2" #europe-west1 "europe-west6"
 IMAGE_NAME="gcr.io/$PROJECT_ID/$SERVICE_NAME"
 

frontend/app.py

@@ -551,12 +551,12 @@ def update_model_choices(provider):
 # -----------------------
 # Gradio UI
 # -----------------------
-with gr.Blocks(title="Substack Articles LLM Engine", theme=gr.themes.Soft(), css=CUSTOM_CSS) as demo:
+with gr.Blocks(title="AI Search Engine for Articles", theme=gr.themes.Soft(), css=CUSTOM_CSS) as demo:
     # Header
     gr.HTML(
         "<div id='app-header'>"
-        "  <h1>📰 Substack Articles LLM Engine</h1>"
-        "  <p>Search Substack content or ask an AI across your feeds — fast and delightful.</p>"
+        "  <h1>📰 AI Search Engine for Articles</h1>"
+        "  <p>Search Substack, Medium and top publications content or ask an AI across your feeds — fast and delightful.</p>"
         "</div>"
     )
 

prefect-cloud.yaml

@@ -1,7 +1,7 @@
 pull:
 - prefect.deployments.steps.git_clone:
     id: clone-step
-    repository: https://github.com/Indraneel99/substack-newsletters-search-course
+    repository: https://github.com/Indraneel99/AISearchEngine
     credentials: "{{ prefect.blocks.github-credentials.my-gh-creds }}"
 
 - prefect.deployments.steps.run_shell_script:

prefect-local.yaml

@@ -1,7 +1,7 @@
 pull:
 - prefect.deployments.steps.git_clone:
     id: clone-step
-    repository: https://github.com/Indraneel99/substack-newsletters-search-course
+    repository: https://github.com/Indraneel99/AISearchEngine
     credentials: "{{ prefect.blocks.github-credentials.my-gh-creds }}"
 
 # This function ensures pip is installed in the environment (Only needed for Prefect Server)

pyproject.toml

@@ -1,7 +1,7 @@
 [project]
-name = "substack-newsletters-search-course"
+name = "AISearchEngine"
 version = "1.0.0"
-description = "A pipeline to retrieve Newsletters from Substack"
+description = "A pipeline to retrieve Newsletters from Substack, Medium and Top publications"
 readme = "README.md"
 authors = [
     {name = "Benito Martin"}

src/api/main.py

@@ -73,9 +73,9 @@ async def lifespan(app: FastAPI):
 # -----------------------
 
 app = FastAPI(
-    title="Substack RAG API",
+    title="Search Engine RAG API",
     version="1.0",
-    description="API for Substack Retrieval-Augmented Generation (RAG) system",
+    description="API for Articles Search Retrieval-Augmented Generation (RAG) system",
     lifespan=lifespan,
     # root_path=root_path,
 )

src/config.py

@@ -12,13 +12,13 @@ from src.models.article_models import FeedItem
 # Supabase database settings
 # -----------------------------
 class SupabaseDBSettings(BaseModel):
-    table_name: str = Field(default="substack_articles", description="Supabase table name")
+    table_name: str = Field(default="feed_articles", description="Supabase table name")
     host: str = Field(default="localhost", description="Database host")
     name: str = Field(default="postgres", description="Database name")
     user: str = Field(default="postgres", description="Database user")
     password: SecretStr = Field(default=SecretStr("password"), description="Database password")
     port: int = Field(default=6543, description="Database port")
-    test_database: str = Field(default="substack_test", description="Test database name")
+    test_database: str = Field(default="feed_test", description="Test database name")
 
 
 # -----------------------------
@@ -43,7 +43,7 @@ class QdrantSettings(BaseModel):
     api_key: str = Field(default="", description="Qdrant API key")
     timeout: int = Field(default=30, description="Qdrant client timeout")
     collection_name: str = Field(
-        default="substack_collection", description="Qdrant collection name"
+        default="feed_collection", description="Qdrant collection name"
     )
     dense_model_name: str = Field(default="BAAI/bge-base-en", description="Dense model name")
     sparse_model_name: str = Field(
@@ -126,7 +126,7 @@ class OpenRouterSettings(BaseModel):
 # -----------------------------
 class OpikObservabilitySettings(BaseModel):
     api_key: str = Field(default="", description="Opik Observability API key")
-    project_name: str = Field(default="substack-pipeline", description="Opik project name")
+    project_name: str = Field(default="feed-pipeline", description="Opik project name")
 
 
 # -----------------------------

src/configs/feeds_rss.yaml

@@ -89,3 +89,39 @@ feeds:
 - name: "slys.dev"
   author: "Anna & Jakub Slys"
   url: "https://iam.slys.dev/feed"
+- name: "Technology Review"
+  author: "Technology Review"
+  url: "https://www.technologyreview.com/feed"
+- name: "AI Trends"
+  author: "AI Trends"
+  url: "https://www.aitrends.com/feed"
+- name: "Machine Learning Mastery"
+  author: "Machine Learning Mastery"
+  url: "https://machinelearningmastery.com/feed"
+- name: "The Gradient"
+  author: "The Gradient"
+  url: "https://thegradient.pub/rss/"
+- name: "Towards Data Science"
+  author: "Towards Data Science"
+  url: "https://towardsdatascience.com/feed/"
+- name: "Microsoft AI"
+  author: "Microsoft AI"
+  url: "https://blogs.microsoft.com/ai/feed/"
+- name: "Marktechpost"
+  author: "Marktechpost"
+  url: "https://www.marktechpost.com/feed/"
+- name: "Daily AI"
+  author: "Daily AI"
+  url: "https://dailyai.com/feed/"
+- name: "TopBots"
+  author: "TopBots"
+  url: "https://www.topbots.com/feed/"
+- name: "Towards AI"
+  author: "Towards AI"
+  url: "https://pub.towardsai.net/feed/"
+- name: "The Pycoach"
+  author: "Frank Andrade"
+  url: "https://medium.com/feed/@frank-andrade"
+- name: "Anthony Alcaraz"
+  author: "Anthony Alcaraz"
+  url: "https://medium.com/feed/@alcarazanthony1"

src/infrastructure/qdrant/qdrant_vectorstore.py

@@ -26,7 +26,7 @@ from qdrant_client.models import Batch, Distance, SparseVector, models
 from sqlalchemy.orm import Session
 
 from src.config import settings
-from src.models.sql_models import SubstackArticle
+from src.models.sql_models import FeedArticle
 from src.models.vectorstore_models import ArticleChunkPayload
 from src.utils.logger_util import log_batch_status, setup_logging
 from src.utils.text_splitter import TextSplitter
@@ -522,7 +522,7 @@ class AsyncQdrantVectorStore:
 
     async def _article_batch_generator(
         self, session: Session, from_date: datetime | None = None
-    ) -> AsyncGenerator[list[SubstackArticle], None]:
+    ) -> AsyncGenerator[list[FeedArticle], None]:
         """Yield batches of articles from SQL database.
 
         Args:
@@ -530,7 +530,7 @@ class AsyncQdrantVectorStore:
             from_date (datetime, optional): Filter articles from this date.
 
         Yields:
-            list[SubstackArticle]: Batch of articles.
+            list[FeedArticle]: Batch of articles.
 
         Raises:
             Exception: If database query fails.
@@ -542,9 +542,9 @@ class AsyncQdrantVectorStore:
         try:
             offset = 0
             while True:
-                query = session.query(SubstackArticle).order_by(SubstackArticle.published_at)
+                query = session.query(FeedArticle).order_by(FeedArticle.published_at)
                 if from_date:
-                    query = query.filter(SubstackArticle.published_at >= from_date)
+                    query = query.filter(FeedArticle.published_at >= from_date)
                 articles = query.offset(offset).limit(self.article_batch_size).all()
                 if not articles:
                     break

src/infrastructure/supabase/create_db.py

@@ -2,17 +2,17 @@ from sqlalchemy import inspect
 from sqlalchemy.exc import SQLAlchemyError
 
 from src.infrastructure.supabase.init_session import init_engine
-from src.models.sql_models import Base, SubstackArticle
+from src.models.sql_models import Base, FeedArticle
 from src.utils.logger_util import setup_logging
 
 logger = setup_logging()
 
 
 def create_table() -> None:
-    """Create the SubstackArticle table in the Supabase Postgres database if it does not exist.
+    """Create the FeedArticle table in the Supabase Postgres database if it does not exist.
 
     This function initializes a SQLAlchemy engine, checks if the table defined by
-    `SubstackArticle.__tablename__` exists in the database, and creates it if necessary.
+    `FeedArticle.__tablename__` exists in the database, and creates it if necessary.
     The engine is properly disposed of after the operation to prevent resource leaks.
     Errors during table creation are logged and handled gracefully.
 
@@ -33,14 +33,14 @@ def create_table() -> None:
         # Create an inspector to check existing tables
         inspector = inspect(engine)
         existing_tables = inspector.get_table_names()
-        table_name = SubstackArticle.__tablename__
+        table_name = FeedArticle.__tablename__
 
         # Check if the table already exists
         if table_name in existing_tables:
             logger.info(f"Table '{table_name}' already exists. No action needed.")
         else:
             logger.info(f"Table '{table_name}' does not exist. Creating...")
-            # Create all tables defined in Base.metadata (includes SubstackArticle)
+            # Create all tables defined in Base.metadata (includes FeedArticle)
             Base.metadata.create_all(bind=engine)
             logger.info(f"Table '{table_name}' created successfully.")
     except SQLAlchemyError as e:

src/models/sql_models.py

@@ -12,7 +12,7 @@ class Base(DeclarativeBase):
     pass
 
 
-class SubstackArticle(Base):
+class FeedArticle(Base):
     __tablename__ = settings.supabase_db.table_name
 
     # Primary internal ID

src/pipelines/flows/rss_ingestion_flow.py

@@ -3,7 +3,7 @@ from prefect import flow, unmapped
 from src.config import settings
 from src.infrastructure.supabase.init_session import init_engine
 from src.models.article_models import FeedItem
-from src.models.sql_models import SubstackArticle
+from src.models.sql_models import FeedArticle
 from src.pipelines.tasks.fetch_rss import fetch_rss_entries
 from src.pipelines.tasks.ingest_rss import ingest_from_rss
 from src.utils.logger_util import setup_logging
@@ -16,7 +16,7 @@ from src.utils.logger_util import setup_logging
     retries=2,
     retry_delay_seconds=120,
 )
-def rss_ingest_flow(article_model: type[SubstackArticle] = SubstackArticle) -> None:
+def rss_ingest_flow(article_model: type[FeedArticle] = FeedArticle) -> None:
     """Fetch and ingest articles from configured RSS feeds concurrently.
 
     Each feed is fetched in parallel and ingested into the database
@@ -24,7 +24,7 @@ def rss_ingest_flow(article_model: type[SubstackArticle] = SubstackArticle) -> N
     after completion.
 
     Args:
-        article_model (type[SubstackArticle]): SQLAlchemy model for storing articles.
+        article_model (type[FeedArticle]): SQLAlchemy model for storing articles.
 
     Returns:
         None
@@ -115,4 +115,4 @@ def rss_ingest_flow(article_model: type[SubstackArticle] = SubstackArticle) -> N
 
 
 if __name__ == "__main__":
-    rss_ingest_flow(article_model=SubstackArticle)
+    rss_ingest_flow(article_model=FeedArticle)

src/pipelines/tasks/fetch_rss.py

@@ -8,13 +8,13 @@ from sqlalchemy.orm import Session
 
 from src.infrastructure.supabase.init_session import init_session
 from src.models.article_models import ArticleItem, FeedItem
-from src.models.sql_models import SubstackArticle
+from src.models.sql_models import FeedArticle
 from src.utils.logger_util import setup_logging
 
 
 @task(
     task_run_name="fetch_rss_entries-{feed.name}",
-    description="Fetch RSS entries from a Substack feed.",
+    description="Fetch RSS entries from a substack/Medium and top publications feed.",
     retries=2,
     retry_delay_seconds=120,
     cache_policy=NO_CACHE,
@@ -22,9 +22,9 @@ from src.utils.logger_util import setup_logging
 def fetch_rss_entries(
     feed: FeedItem,
     engine: Engine,
-    article_model: type[SubstackArticle] = SubstackArticle,
+    article_model: type[FeedArticle] = FeedArticle,
 ) -> list[ArticleItem]:
-    """Fetch all RSS items from a Substack feed and convert them to ArticleItem objects.
+    """Fetch all RSS items from a substack/Medium and top publications feed and convert them to ArticleItem objects.
 
     Each task uses its own SQLAlchemy session. Articles already stored in the database
     or with empty links/content are skipped. Errors during parsing individual items
@@ -33,8 +33,8 @@ def fetch_rss_entries(
     Args:
         feed (FeedItem): Metadata for the feed (name, author, URL).
         engine (Engine): SQLAlchemy engine for database connection.
-        article_model (type[SubstackArticle], optional): Model used to persist articles.
-            Defaults to SubstackArticle.
+        article_model (type[FeedArticle], optional): Model used to persist articles.
+            Defaults to FeedArticle.
 
     Returns:
         list[ArticleItem]: List of new ArticleItem objects ready for parsing/ingestion.

src/pipelines/tasks/ingest_rss.py

@@ -6,7 +6,7 @@ from sqlalchemy.orm import Session
 from src.config import settings
 from src.infrastructure.supabase.init_session import init_session
 from src.models.article_models import ArticleItem, FeedItem
-from src.models.sql_models import SubstackArticle
+from src.models.sql_models import FeedArticle
 from src.utils.logger_util import setup_logging
 
 
@@ -20,7 +20,7 @@ from src.utils.logger_util import setup_logging
 def ingest_from_rss(
     fetched_articles: list[ArticleItem],
     feed: FeedItem,
-    article_model: type[SubstackArticle],
+    article_model: type[FeedArticle],
     engine: Engine,
 ) -> None:
     """Ingest articles fetched from RSS (already Markdownified).
@@ -93,7 +93,7 @@ def ingest_from_rss(
 def _persist_batch(
     session: Session,
     batch: list[ArticleItem],
-    article_model: type[SubstackArticle],
+    article_model: type[FeedArticle],
 ) -> None:
     """Helper to bulk insert a batch of ArticleItems."""
     rows = [

tests/integration/test_db_connection.py

@@ -8,7 +8,7 @@ setup_logging()
 
 
 def test_connect_to_test_table(db_session: Connection) -> None:
-    """Test connectivity to the 'substack_test' table and fetch a single row.
+    """Test connectivity to the 'feed_test' table and fetch a single row.
 
     Args:
         db_session (Connection): SQLAlchemy Connection object.
@@ -18,12 +18,12 @@ def test_connect_to_test_table(db_session: Connection) -> None:
         Exception: If the table does not exist or query fails.
 
     """
-    logger.info("Testing connection to 'substack_test' table...")
+    logger.info("Testing connection to 'feed_test' table...")
 
     try:
-        result = db_session.execute(text("SELECT * FROM substack_test LIMIT 1")).fetchall()
+        result = db_session.execute(text("SELECT * FROM feed_test LIMIT 1")).fetchall()
         logger.info(f"Query result: {result}")
         assert isinstance(result, list), "Query result is not a list"
     except Exception as e:
-        logger.error(f"Failed to query 'substack_test' table: {e}")
+        logger.error(f"Failed to query 'feed_test' table: {e}")
         raise

tests/integration/test_rss_pipeline.py

@@ -4,7 +4,7 @@ from loguru import logger
 from sqlalchemy import text
 from sqlalchemy.engine import Engine
 from sqlalchemy.orm import Session
-from test_models.test_sql_models import SubstackTestArticle  # Test-specific table model
+from test_models.test_sql_models import FeedTestArticle  # Test-specific table model
 
 from src.models.article_models import FeedItem
 from src.pipelines.tasks.fetch_rss import fetch_rss_entries
@@ -29,12 +29,12 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
     """
 
     # Clear test table
-    logger.info("Clearing test table 'substack_test'")
-    db_session.execute(text("DELETE FROM substack_test"))
+    logger.info("Clearing test table 'feed_test'")
+    db_session.execute(text("DELETE FROM feed_test"))
     db_session.commit()
 
     # Verify table is empty
-    initial_count = db_session.query(SubstackTestArticle).count()
+    initial_count = db_session.query(FeedTestArticle).count()
     logger.info(f"Initial article count in test table: {initial_count}")
     assert initial_count == 0, "Test table was not cleared"
 
@@ -88,7 +88,7 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
     fetched_articles = fetch_rss_entries(
         test_feed,
         engine=db_engine,
-        article_model=SubstackTestArticle,
+        article_model=FeedTestArticle,
     )
     logger.info(f"Fetched {len(fetched_articles)} articles for feed '{test_feed.name}'")
 
@@ -99,14 +99,14 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
     ingest_from_rss(
         fetched_articles,
         feed=test_feed,
-        article_model=SubstackTestArticle,
+        article_model=FeedTestArticle,
         engine=db_engine,
     )
 
     # Verify DB insertion
     articles_in_db = (
-        db_session.query(SubstackTestArticle)
-        .order_by(SubstackTestArticle.published_at.desc())
+        db_session.query(FeedTestArticle)
+        .order_by(FeedTestArticle.published_at.desc())
         .all()
     )
     logger.info(f"Inserted article titles: {[a.title for a in articles_in_db]}")
@@ -122,7 +122,7 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
 
 ################################################################################
 # The code below calls out to live URLs and is not suitable for CI,
-# as Substack may block requests from CI environments.
+# as Substack/medium may block requests from CI environments.
 # It is left here for reference and can be run manually if desired.
 # Uncomment to enable live integration test
 
@@ -132,7 +132,7 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
 # from sqlalchemy import text
 # from sqlalchemy.engine import Engine
 # from sqlalchemy.orm import Session
-# from test_models.test_sql_models import SubstackTestArticle  # Test-specific table model
+# from test_models.test_sql_models import FeedTestArticle  # Test-specific table model
 
 # from src.models.article_models import FeedItem
 # from src.pipelines.tasks.batch_parse_ingest_articles import parse_and_ingest
@@ -153,12 +153,12 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
 
 #     """
 #     # Clear test table
-#     logger.info("Clearing test table 'substack_test'")
-#     db_session.execute(text("DELETE FROM substack_test"))
+#     logger.info("Clearing test table 'feed_test'")
+#     db_session.execute(text("DELETE FROM feed_test"))
 #     db_session.commit()
 
 #     # Verify table is empty
-#     initial_count = db_session.query(SubstackTestArticle).count()
+#     initial_count = db_session.query(FeedTestArticle).count()
 #     logger.info(f"Initial article count in test table: {initial_count}")
 #     assert initial_count == 0, "Test table was not cleared"
 
@@ -173,7 +173,7 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
 #     fetched_articles = fetch_rss_entries(
 #         test_feed,
 #         engine=db_engine,
-#         article_model=SubstackTestArticle,
+#         article_model=FeedTestArticle,
 #     )
 #     logger.info(f"Fetched {len(fetched_articles)} articles for feed '{test_feed.name}'")
 
@@ -185,14 +185,14 @@ def test_rss_pipeline_end_to_end_mocked(db_session: Session, db_engine: Engine)
 #     parse_and_ingest(
 #         fetched_articles,
 #         feed=test_feed,
-#         article_model=SubstackTestArticle,
+#         article_model=FeedTestArticle,
 #         engine=db_engine,
 #     )
 
 #     # Verify DB insertion
 #     articles_in_db = (
-#         db_session.query(SubstackTestArticle)
-#         .order_by(SubstackTestArticle.published_at.desc())
+#         db_session.query(FeedTestArticle)
+#         .order_by(FeedTestArticle.published_at.desc())
 #         .all()
 #     )
 #     logger.info(f"Inserted article titles: {[a.title for a in articles_in_db]}")

tests/test_models/test_sql_models.py

@@ -10,8 +10,8 @@ class Base(DeclarativeBase):
     pass
 
 
-class SubstackTestArticle(Base):
-    __tablename__ = "substack_test"
+class FeedTestArticle(Base):
+    __tablename__ = "feed_test"
 
     # Primary internal ID
     id: Mapped[int] = mapped_column(BigInteger, primary_key=True, index=True)

tests/unit/test_fetch_rss_entries.py

@@ -3,7 +3,7 @@ import responses
 from loguru import logger
 from sqlalchemy import text
 from sqlalchemy.orm import Session
-from test_models.test_sql_models import SubstackTestArticle
+from test_models.test_sql_models import FeedTestArticle
 
 from src.infrastructure.supabase.init_session import init_engine
 from src.models.article_models import ArticleItem, FeedItem
@@ -49,8 +49,8 @@ def test_fetch_rss_mocked_feed() -> None:
     try:
         # Clear the test table before running
         session = Session(bind=engine)
-        logger.info("Clearing test table 'substack_test' before test")
-        session.execute(text("DELETE FROM substack_test"))
+        logger.info("Clearing test table 'feed_test' before test")
+        session.execute(text("DELETE FROM feed_test"))
         session.commit()
         logger.info("Test table cleared")
 
@@ -58,7 +58,7 @@ def test_fetch_rss_mocked_feed() -> None:
         articles = fetch_rss_entries(
             feed=test_feed,
             engine=engine,
-            article_model=SubstackTestArticle,
+            article_model=FeedTestArticle,
         )
         logger.info(f"Fetched {len(articles)} articles from {test_feed.url}")
 

uv.lock

@@ -3746,7 +3746,7 @@ wheels = [
 ]
 
 [[package]]
-name = "substack-newsletters-search-course"
+name = "AISearchEngine"
 version = "1.0.0"
 source = { editable = "." }
 dependencies = [