Add new data from Rohit to the app

Author: pamelafox

.devcontainer/devcontainer.json

@@ -31,8 +31,7 @@
 				"ms-python.vscode-pylance",
 				"ms-python.vscode-python-envs",
 				"charliermarsh.ruff",
-				"mtxr.sqltools",
-				"mtxr.sqltools-driver-pg",
+				// TODO: Add PostgreSQL extension once its in marketplace
 				"esbenp.prettier-vscode",
 				"mechatroner.rainbow-csv",
 				"ms-vscode.vscode-node-azure-pack",

.vscode/settings.json

@@ -1,26 +1,20 @@
 {
-    "sqltools.connections": [
+    "pgsql.connections": [
         {
-            "previewLimit": 50,
+            "id": "92CC1089-BAD0-44A4-B071-A50A6EC12B67",
+            "groupId": "F3347CD6-9995-4EE9-8D98-D88DB010FA5B",
+            "authenticationType": "SqlLogin",
+            "connectTimeout": 15,
+            "applicationName": "vscode-pgsql",
+            "clientEncoding": "utf8",
+            "sslmode": "prefer",
             "server": "localhost",
-            "port": 5432,
-            "driver": "PostgreSQL",
-            "name": "local",
+            "user": "admin",
+            "password": "",
+            "savePassword": true,
             "database": "postgres",
-            "username": "admin",
-            "password": "postgres"
-        },
-        {
-            "name": "Azure database",
-            "driver": "PostgreSQL",
-            "server": "<HOSTNAME>.postgres.database.azure.com",
-            "port": 5432,
-            "database": "postgres",
-            "username": "<USERNAME>",
-            "askForPassword": true,
-            "pgOptions": {
-                "ssl": true
-            }
+            "profileName": "local-pg",
+            "expiresOn": 0
         }
     ],
     "python.testing.pytestArgs": [

convert_csv_json.py

@@ -3,7 +3,7 @@
 import json
 
 # Read CSV file - Using the correct dialect to handle quotes properly
-with open("data.csv", encoding="utf-8") as csv_file:
+with open("pittsburgh_restaurants.csv", encoding="utf-8") as csv_file:
     # Use the csv.reader with proper quoting parameters
     csv_reader = csv.reader(csv_file, quoting=csv.QUOTE_ALL, doublequote=True, escapechar="\\")
     header = next(csv_reader)  # Get the header row
@@ -42,10 +42,17 @@
             item[header[i]] = value
     # remove is_open column
     del item["is_open"]
+    del item["hours"]
+    del item["neighborhood"]
+    del item["tags"]
+    del item["vibe"]
+    del item["top_reviews"]
+    if item["price_level"] == "":
+        item["price_level"] = 2.0  # Assume 2.0 if empty
     json_data.append(item)
 
 # Write to JSON file
-with open("data.json", "w", encoding="utf-8") as f:
+with open("src/backend/fastapi_app/seed_data.json", "w", encoding="utf-8") as f:
     json.dump(json_data, f, indent=4, ensure_ascii=False)
 
 print(f"Successfully converted CSV data to JSON format with {len(json_data)} records")

pittsburgh_restaurants.csv

@@ -1,4 +1,4 @@
-id,name,neighborhood,cuisine,price_level,rating,review_count,is_open,hours,tags,description,menu summary,top_reviews,vibe
+id,name,neighborhood,cuisine,price_level,rating,review_count,is_open,hours,tags,description,menu_summary,top_reviews,vibe
 XDqSIoVRrG4JoF79HTT1Zw,Balvanera,Unknown,"Argentine, Wine Bars, Steakhouses",,4.5,132,True,"[{'is_overnight': False, 'start': '1700', 'end': '2200', 'day': 0}, {'is_overnight': False, 'start': '1700', 'end': '2200', 'day': 1}, {'is_overnight': False, 'start': '1700', 'end': '2200', 'day': 2}, {'is_overnight': False, 'start': '1700', 'end': '2200', 'day': 3}, {'is_overnight': False, 'start': '1700', 'end': '2200', 'day': 4}, {'is_overnight': False, 'start': '1100', 'end': '1500', 'day': 5}, {'is_overnight': False, 'start': '1700', 'end': '2200', 'day': 5}, {'is_overnight': False, 'start': '1100', 'end': '1500', 'day': 6}, {'is_overnight': False, 'start': '1700', 'end': '2200', 'day': 6}]","Argentine, Wine Bars, Steakhouses",Balvanera is an Argentine restaurant that also serves as a wine bar and steakhouse.,"The cuisine is Argentine, likely featuring grilled meats and steaks, along with a selection of wines.",No reviews available to summarize.,Cannot be determined without customer reviews and feedback.
 MdpoKByQGwhCb9ZWaPxUXg,The Forge,Unknown,"Lounges, Mediterranean, Cocktail Bars",,4.5,236,True,"[{'is_overnight': False, 'start': '1700', 'end': '2300', 'day': 0}, {'is_overnight': False, 'start': '1700', 'end': '2300', 'day': 1}, {'is_overnight': False, 'start': '1700', 'end': '2300', 'day': 2}, {'is_overnight': False, 'start': '1700', 'end': '2300', 'day': 3}, {'is_overnight': False, 'start': '1700', 'end': '2300', 'day': 4}, {'is_overnight': False, 'start': '1700', 'end': '2300', 'day': 5}]","Lounges, Mediterranean, Cocktail Bars",,,,
 g6_Mkq_86wniZ8hYZVjhIA,The Eagle - Pittsburgh,Unknown,"Southern, Soul Food, American",2.0,4.3,476,True,"[{'is_overnight': False, 'start': '1100', 'end': '2300', 'day': 0}, {'is_overnight': False, 'start': '1100', 'end': '2300', 'day': 1}, {'is_overnight': False, 'start': '1100', 'end': '2300', 'day': 2}, {'is_overnight': False, 'start': '1100', 'end': '2300', 'day': 3}, {'is_overnight': False, 'start': '1100', 'end': '0000', 'day': 4}, {'is_overnight': False, 'start': '1100', 'end': '0000', 'day': 5}, {'is_overnight': False, 'start': '1100', 'end': '2200', 'day': 6}]","Southern, Soul Food, American",The Eagle in Pittsburgh serves up Southern and soul food classics in an American diner setting.,"The menu features traditional Southern and soul food dishes as well as American comfort food favorites. Expect items like fried chicken, collard greens, mac and cheese, and other hearty, homestyle cooking.",No Yelp reviews found for The Eagle - Pittsburgh location.,Homey

src/backend/fastapi_app/api_models.py

@@ -42,22 +42,17 @@ class ChatRequest(BaseModel):
 
 
 class ItemPublic(BaseModel):
-    id: int
+    id: str
     name: str
-    location: str
     cuisine: str
     rating: int
     price_level: int
     review_count: int
-    hours: str
-    tags: list[str]
     description: str
     menu_summary: str
-    top_reviews: str
-    vibe: str
 
     def to_str_for_rag(self):
-        return f"Name:{self.name} Description:{self.description} Location:{self.location} Cuisine:{self.cuisine} Rating:{self.rating} Price Level:{self.price_level} Review Count:{self.review_count} Hours:{self.hours} Tags:{self.tags} Menu Summary:{self.menu_summary} Top Reviews:{self.top_reviews} Vibe:{self.vibe}"  # noqa: E501
+        return f"Name:{self.name} Description:{self.description} Cuisine:{self.cuisine} Rating:{self.rating} Price Level:{self.price_level} Review Count:{self.review_count} Menu Summary:{self.menu_summary}"  # noqa: E501
 
 
 class ItemWithDistance(ItemPublic):
@@ -75,7 +70,7 @@ class ThoughtStep(BaseModel):
 
 
 class RAGContext(BaseModel):
-    data_points: dict[int, ItemPublic]
+    data_points: dict[str, ItemPublic]
     thoughts: list[ThoughtStep]
     followup_questions: Optional[list[str]] = None
 

src/backend/fastapi_app/postgres_models.py

@@ -1,8 +1,7 @@
 from __future__ import annotations
 
 from pgvector.sqlalchemy import Vector
-from sqlalchemy import VARCHAR, Index
-from sqlalchemy.dialects import postgresql
+from sqlalchemy import Index
 from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column
 
 
@@ -13,19 +12,14 @@ class Base(DeclarativeBase):
 
 class Item(Base):
     __tablename__ = "items"
-    id: Mapped[int] = mapped_column(primary_key=True, autoincrement=True)
+    id: Mapped[str] = mapped_column(primary_key=True)
     name: Mapped[str] = mapped_column()
-    location: Mapped[str] = mapped_column()
     cuisine: Mapped[str] = mapped_column()
     rating: Mapped[int] = mapped_column()
     price_level: Mapped[int] = mapped_column()
     review_count: Mapped[int] = mapped_column()
-    hours: Mapped[str] = mapped_column()
-    tags: Mapped[list[str]] = mapped_column(postgresql.ARRAY(VARCHAR))  # Array of strings
     description: Mapped[str] = mapped_column()
     menu_summary: Mapped[str] = mapped_column()
-    top_reviews: Mapped[str] = mapped_column()
-    vibe: Mapped[str] = mapped_column()
 
     # Embeddings for different models:
     embedding_3l: Mapped[Vector] = mapped_column(Vector(1024), nullable=True)  # text-embedding-3-large
@@ -42,10 +36,10 @@ def to_dict(self, include_embedding: bool = False):
         return model_dict
 
     def to_str_for_rag(self):
-        return f"Name:{self.name} Description:{self.description} Location:{self.location} Cuisine:{self.cuisine} Rating:{self.rating} Price Level:{self.price_level} Review Count:{self.review_count} Hours:{self.hours} Tags:{self.tags} Menu Summary:{self.menu_summary} Top Reviews:{self.top_reviews} Vibe:{self.vibe}"  # noqa: E501
+        return f"Name:{self.name} Description:{self.description} Cuisine:{self.cuisine} Rating:{self.rating} Price Level:{self.price_level} Review Count:{self.review_count} Menu Summary:{self.menu_summary}"  # noqa: E501
 
     def to_str_for_embedding(self):
-        return f"Name: {self.name} Description: {self.description} Cuisine: {self.cuisine} Tags: {self.tags} Menu Summary: {self.menu_summary} Top Reviews: {self.top_reviews} Vibe: {self.vibe}"  # noqa: E501
+        return f"Name: {self.name} Description: {self.description} Cuisine: {self.cuisine} Menu Summary: {self.menu_summary}"  # noqa: E501
 
 
 """

src/backend/fastapi_app/rag_advanced.py

@@ -119,6 +119,8 @@ async def prepare_context(self) -> tuple[list[ItemPublic], list[ThoughtStep]]:
         most_recent_response = run_results.new_items[-1]
         if isinstance(most_recent_response, ToolCallOutputItem):
             search_results = most_recent_response.output
+            if not isinstance(search_results, SearchResults):
+                raise ValueError(f"Error retrieving search results: {search_results}")
         else:
             raise ValueError("Error retrieving search results, model did not call tool properly")