Initial commit with edited code

Author: chatrtham

src/fastapi_app/postgres_engine.py

@@ -17,11 +17,12 @@ async def create_postgres_engine(*, host, username, database, password, sslmode,
     else:
         logger.info("Authenticating to PostgreSQL using password...")
 
-    DATABASE_URI = f"postgresql+asyncpg://{username}:{password}@{host}/{database}"
+    DATABASE_URI = f"postgresql+asyncpg://{username}:{password}@{host}:5432/{database}"
     # Specify SSL mode if needed
     if sslmode:
         DATABASE_URI += f"?ssl={sslmode}"
 
+
     engine = create_async_engine(
         DATABASE_URI,
         echo=False,

src/fastapi_app/postgres_models.py

@@ -13,13 +13,47 @@ class Base(DeclarativeBase, MappedAsDataclass):
 
 
 class Item(Base):
-    __tablename__ = "items"
+    __tablename__ = "packages"
     id: Mapped[int] = mapped_column(primary_key=True, autoincrement=True)
-    type: Mapped[str] = mapped_column()
-    brand: Mapped[str] = mapped_column()
-    name: Mapped[str] = mapped_column()
-    description: Mapped[str] = mapped_column()
+    package_name: Mapped[str] = mapped_column()
+    package_picture: Mapped[str] = mapped_column()
+    url: Mapped[str] = mapped_column()
     price: Mapped[float] = mapped_column()
+    cash_discount: Mapped[float] = mapped_column()
+    installment_month: Mapped[str] = mapped_column()
+    installment_limit: Mapped[str] = mapped_column()
+    price_to_reserve_for_this_package: Mapped[str] = mapped_column()
+    shop_name: Mapped[str] = mapped_column()
+    category: Mapped[str] = mapped_column()
+    category_tags: Mapped[str] = mapped_column()
+    preview_1_10: Mapped[str] = mapped_column()
+    selling_point: Mapped[str] = mapped_column()
+    meta_keywords: Mapped[str] = mapped_column()
+    brand: Mapped[str] = mapped_column()
+    min_max_age: Mapped[str] = mapped_column()
+    locations_time_open_close_how_to_transport_parking_google_maps: Mapped[str] = mapped_column()
+    meta_description: Mapped[str] = mapped_column()
+    price_details: Mapped[str] = mapped_column()
+    package_details: Mapped[str] = mapped_column()
+    important_info: Mapped[str] = mapped_column()
+    payment_booking_info: Mapped[str] = mapped_column()
+    general_info: Mapped[str] = mapped_column()
+    early_signs_for_diagnosis: Mapped[str] = mapped_column()
+    how_to_diagnose: Mapped[str] = mapped_column()
+    hdcare_summary: Mapped[str] = mapped_column()
+    common_question: Mapped[str] = mapped_column()
+    know_this_disease: Mapped[str] = mapped_column()
+    courses_of_action: Mapped[str] = mapped_column()
+    signals_to_proceed_surgery: Mapped[str] = mapped_column()
+    get_to_know_this_surgery: Mapped[str] = mapped_column()
+    comparisons: Mapped[str] = mapped_column()
+    getting_ready: Mapped[str] = mapped_column()
+    recovery: Mapped[str] = mapped_column()
+    side_effects: Mapped[str] = mapped_column()
+    review_4_5_stars: Mapped[str] = mapped_column()
+    brand_option_in_thai_name: Mapped[str] = mapped_column()
+    brand_ranking_position: Mapped[str] = mapped_column()
+    faq: Mapped[str] = mapped_column()
     embedding: Mapped[Vector] = mapped_column(Vector(1536))  # ada-002
 
     def to_dict(self, include_embedding: bool = False):
@@ -31,10 +65,26 @@ 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} Price:{self.price} Brand:{self.brand} Type:{self.type}"
+            return f"""package_name: {self.package_name}
+package_picture: {self.package_picture}
+url: {self.url}
+price: {self.price}
+cash_discount: {self.cash_discount}
+installment_month: {self.installment_month}
+installment_limit: {self.installment_limit}
+price_to_reserve_for_this_package: {self.price_to_reserve_for_this_package}
+brand: {self.brand}
+price_details: {self.price_details}
+package_details: {self.package_details}
+important_info: {self.important_info}
+payment_booking_info: {self.payment_booking_info}
+general_info: {self.general_info}
+brand_ranking_position: {self.brand_ranking_position}
+faq: {self.faq}
+"""
 
     def to_str_for_embedding(self):
-        return f"Name: {self.name} Description: {self.description} Type: {self.type}"
+        return f"Name: {self.package_name} Description: {self.package_details} Type: {self.meta_keywords}"
 
 
 # Define HNSW index to support vector similarity search through the vector_cosine_ops access method (cosine distance).

src/fastapi_app/postgres_searcher.py

@@ -46,19 +46,19 @@ async def search(
 
         vector_query = f"""
             SELECT id, RANK () OVER (ORDER BY embedding <=> :embedding) AS rank
-                FROM items
+                FROM packages
                 {filter_clause_where}
                 ORDER BY embedding <=> :embedding
                 LIMIT 20
             """
 
         fulltext_query = f"""
-            SELECT id, RANK () OVER (ORDER BY ts_rank_cd(to_tsvector('english', description), query) DESC)
-                FROM items, plainto_tsquery('english', :query) query
-                WHERE to_tsvector('english', description) @@ query {filter_clause_and}
-                ORDER BY ts_rank_cd(to_tsvector('english', description), query) DESC
-                LIMIT 20
-            """
+            SELECT id, RANK () OVER (ORDER BY ts_rank_cd(to_tsvector('thai', package_details), query) DESC)
+            FROM packages, plainto_tsquery('thai', :query) query
+            WHERE to_tsvector('thai', package_details) @@ query {filter_clause_and}
+            ORDER BY ts_rank_cd(to_tsvector('thai', package_details), query) DESC
+            LIMIT 20
+        """
 
         hybrid_query = f"""
         WITH vector_search AS (

src/fastapi_app/prompts/answer.txt

@@ -1,9 +1,9 @@
-Assistant helps customers with questions about products.
-Respond as if you are a salesperson helping a customer in a store. Do NOT respond with tables.
-Answer ONLY with the product details listed in the products.
-If there isn't enough information below, say you don't know.
-Do not generate answers that don't use the sources below.
-Each product has an ID in brackets followed by colon and the product details.
-Always include the product ID for each product you use in the response.
-Use square brackets to reference the source, for example [52].
-Don't combine citations, list each product separately, for example [27][51].
+You are a woman name Jib (จิ๊บ), a sales consultant from HDmall which you need to convince the customers to buy some products from our store name HDmall, Please answer in casual and friendly manner. Don't recommend any product yet on the first conversation before the customer asking. You can use Emoji to make the conversation more friendly if you want.
+Represent youself as จิ๊บ and the customer as คุณลูกค้า when the conversation is in Thai language.
+Represent youself as Jib and the customer as you when the conversation is in English language.
+When the customer asks about any packages, please make sure to provide brand, price , URL and location every time.
+For EVERY URL that you provided, create a new line then wrap it with <a href=""> so the user can open it via UI
+If customer wants to talk with admin, please provide the URL link in the datasource.
+If the user is asking a question regarding location, proximity, or area, query relevant document from the source and ask where the user is at. please try to suggest services or answers closest to the location the user is asking as much as possible.
+Answer ONLY with the facts listed in the list of sources below. If there isn't enough information below, say sorry you don't know. Do not generate answers that don't use the sources below. If asking a clarifying question to the user would help, ask the question.
+Each source has a name followed by colon and the actual information, always include the source name for each fact you use in the response. Use square brackets to reference the source, for example [info1.txt]. Don't combine sources, list each source separately, for example [info1.txt][info2.pdf].

src/fastapi_app/prompts/query.txt

@@ -1,6 +1,6 @@
 Below is a history of the conversation so far, and a new question asked by the user that needs to be answered by searching database rows.
-You have access to an Azure PostgreSQL database with an items table that has columns for title, description, brand, price, and type.
+You have access to an Azure PostgreSQL database with an packages table that has columns for id, package_name, package_picture, url, price, cash_discount, installment_month, installment_limit, price_to_reserve_for_this_package, shop_name, category, category_tags, preview_1_10, selling_point, meta_keywords, brand, min_max_age, locations_time_open_close_how_to_transport_parking_google_maps, meta_description, price_details, package_details, important_info, payment_booking_info, general_info, early_signs_for_diagnosis, how_to_diagnose, hdcare_summary, common_question, know_this_disease, courses_of_action, signals_to_proceed_surgery, get_to_know_this_surgery, comparisons, getting_ready, recovery, side_effects, review_4_5_stars, brand_option_in_thai_name, brand_ranking_position and faq.
 Generate a search query based on the conversation and the new question.
-If the question is not in English, translate the question to English before generating the search query.
+If the question is not in Thai, translate the question to Thai before generating the search query.
 If you cannot generate a search query, return the original user question.
 DO NOT return anything besides the query.

src/fastapi_app/query_rewriter.py

@@ -34,20 +34,20 @@ def build_search_function() -> list[ChatCompletionToolParam]:
                                 },
                             },
                         },
-                        "brand_filter": {
-                            "type": "object",
-                            "description": "Filter search results based on brand of the product",
-                            "properties": {
-                                "comparison_operator": {
-                                    "type": "string",
-                                    "description": "Operator to compare the column value, either '==' or '!='",
-                                },
-                                "value": {
-                                    "type": "string",
-                                    "description": "Value to compare against, e.g. AirStrider",
-                                },
-                            },
-                        },
+                        # "brand_filter": {
+                        #     "type": "object",
+                        #     "description": "Filter search results based on brand of the product",
+                        #     "properties": {
+                        #         "comparison_operator": {
+                        #             "type": "string",
+                        #             "description": "Operator to compare the column value, either '==' or '!='",
+                        #         },
+                        #         "value": {
+                        #             "type": "string",
+                        #             "description": "Value to compare against, e.g. AirStrider",
+                        #         },
+                        #     },
+                        # },
                     },
                     "required": ["search_query"],
                 },
@@ -77,15 +77,15 @@ def extract_search_arguments(chat_completion: ChatCompletion):
                             "value": price_filter["value"],
                         }
                     )
-                if "brand_filter" in arg and arg["brand_filter"]:
-                    brand_filter = arg["brand_filter"]
-                    filters.append(
-                        {
-                            "column": "brand",
-                            "comparison_operator": brand_filter["comparison_operator"],
-                            "value": brand_filter["value"],
-                        }
-                    )
+                # if "brand_filter" in arg and arg["brand_filter"]:
+                #     brand_filter = arg["brand_filter"]
+                #     filters.append(
+                #         {
+                #             "column": "brand",
+                #             "comparison_operator": brand_filter["comparison_operator"],
+                #             "value": brand_filter["value"],
+                #         }
+                #     )
     elif query_text := response_message.content:
         search_query = query_text.strip()
     return search_query, filters

src/fastapi_app/rag_advanced.py

@@ -71,7 +71,7 @@ async def run(
         results = await self.searcher.search_and_embed(
             query_text,
             top=top,
-            enable_vector_search=vector_search,
+            # enable_vector_search=vector_search,
             enable_text_search=text_search,
             filters=filters,
         )