[Demo] All files needed for demo run (#1)

* all demo parts

* small fix to notebook

* notebook documentation

jewelry/data/ingest.py

@@ -0,0 +1,51 @@
+import sys
+import os
+from dotenv import load_dotenv
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.embeddings import OpenAIEmbeddings
+from langchain.vectorstores import Chroma
+from chromadb.config import Settings
+
+def load_environment():
+    load_dotenv()
+    if "OPENAI_API_KEY" not in os.environ:
+        print("Error: OPENAI_API_KEY is not set in the .env file or environment.")
+        sys.exit(1)
+
+def ingest_to_chroma(file_path):
+    with open(file_path, 'r') as file:
+        text = file.read()
+
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+    chunks = text_splitter.split_text(text)
+
+    embeddings = OpenAIEmbeddings()
+
+    # Initialize Chroma with persistence
+    chroma_settings = Settings(
+        chroma_db_impl='duckdb+parquet',
+        persist_directory='./chroma_db'  # This directory will store the persisted data
+    )
+
+    db = Chroma.from_texts(
+        texts=chunks,
+        embedding=embeddings,
+        client_settings=chroma_settings,
+        persist_directory='./chroma_db'
+    )
+
+    # Persistence is now handled automatically
+    print(f"Text from {file_path} has been successfully ingested into Chroma and persisted.")
+
+if __name__ == "__main__":
+    if len(sys.argv) != 2:
+        print("Usage: python script_name.py <path_to_text_file>")
+        sys.exit(1)
+
+    file_path = sys.argv[1]
+    if not os.path.exists(file_path):
+        print(f"Error: File {file_path} does not exist.")
+        sys.exit(1)
+
+    load_environment()
+    ingest_to_chroma(file_path)

jewelry/data/main.py

@@ -0,0 +1,60 @@
+import os
+from sqlalchemy import create_engine
+import pandas as pd
+from jewelry.data.sql_db import drop_tables, create_tables, get_engine, get_items, get_user_items_purchases_history
+from genai_factory.actions import ingest
+def init_sql_db(data_path: str = "data", mock_data_path: str = "./data/mock_data", reset: bool = True):
+    """
+    Initialize the SQL database and load the mock data if available.
+
+    :param data_path:      Data path.
+    :param mock_data_path: Mock data path.
+    :param reset:          Whether to reset the database.
+    """
+    # Create the base data path if it doesn't exist:
+    if not os.path.exists(data_path):
+        os.makedirs(data_path)
+
+    # Connect to the SQL database:
+    sql_connection_url = f"sqlite:///{data_path}/sql.db"
+    engine = get_engine(sql_connection_url=sql_connection_url)
+
+    # Drop the tables if reset is required:
+    if reset:
+        drop_tables(engine=engine)
+
+    # Create the tables:
+    create_tables(engine=engine)
+
+    # Check if needed to load mock data:
+    if not mock_data_path:
+        return
+
+    # Load the mock data:
+    products = pd.read_csv(os.path.join(mock_data_path, "products.csv"))
+    items = pd.read_csv(os.path.join(mock_data_path, "items.csv"))
+    users = pd.read_csv(os.path.join(mock_data_path, "users.csv"))
+    stocks = pd.read_csv(os.path.join(mock_data_path, "stocks.csv"))
+    purchases = pd.read_csv(os.path.join(mock_data_path, "purchases.csv"))
+    stock_to_purchase = pd.read_csv(os.path.join(mock_data_path, "stock_to_purchase.csv"))
+    reviews = pd.read_csv(os.path.join(mock_data_path, "reviews.csv"))
+
+    # Insert the mock data into tables:
+    products.to_sql(name="product", con=engine, if_exists="replace", index=False)
+    items.to_sql(name="item", con=engine, if_exists="replace", index=False)
+    users.to_sql(name="user", con=engine, if_exists="replace", index=False)
+    stocks.to_sql(name="stock", con=engine, if_exists="replace", index=False)
+    purchases.to_sql(name="purchase", con=engine, if_exists="replace", index=False)
+    stock_to_purchase.to_sql(name="stock_to_purchase", con=engine, if_exists="replace", index=False)
+    reviews.to_sql(name="review", con=engine, if_exists="replace", index=False)
+
+
+if __name__ == "__main__":
+    init_sql_db()
+    engine = get_engine(f"sqlite:///data/sql.db")
+    items = get_items(engine=engine, kinds=["rings", "bracelets"], stones=["no stones"], metals=["white gold"])
+    print(items)
+    items = get_user_items_purchases_history(engine=engine, user_id="6")
+    print(items)
+
+

jewelry/data/mock_data/documents/customer_policy.txt

@@ -0,0 +1,51 @@
+## Customer Policy for Iguazio Jewelry
+
+**General:**
+
+* We strive to provide excellent customer service and high-quality jewelry products.
+* We reserve the right to modify this policy at any time without prior notice.
+* By placing an order with us, you acknowledge and agree to be bound by the terms of this policy.
+
+**Orders:**
+
+* We accept orders placed on our website 24/7.
+* Orders are processed within 1 day after they are placed.
+* Order confirmation emails will be sent to the email address provided at checkout.
+* We reserve the right to cancel any order for any reason.
+
+**Payments:**
+
+* We accept visa and mastercard.
+* All payments are processed securely through a trusted payment gateway.
+* Your credit card information is never stored on our servers.
+
+**Shipping:**
+
+* We offer a variety of shipping options with varying delivery times and costs.
+* Shipping costs are calculated based on the weight, size, and destination of your order.
+* You will be able to choose your preferred shipping option at checkout.
+* We are not responsible for lost, stolen, or damaged packages once they have been shipped.
+
+**Returns and Exchanges:**
+
+* We accept returns and exchanges within 1 month of the purchase date.
+* Items must be returned in their original packaging, unworn, and in new condition.
+* We do not offer refunds for shipping costs on returned items.
+* Please contact us for a return authorization before sending any items back.
+
+**Warranties:**
+
+* All our jewelry comes with a 5-year warranty against manufacturing defects.
+* This warranty does not cover damage caused by normal wear and tear, misuse, or accidents.
+* Please contact us for warranty claims.
+
+**Privacy:**
+
+* We respect your privacy and are committed to protecting your personal information.
+* We will not share your personal information with any third party without your consent.
+* You can review our full privacy policy on our website.
+
+**Contact Us:**
+
+* If you have any questions or concerns, please contact us via email at [email protected] or phone at 123456789.
+* We will reply to your inquiries as soon as possible.

jewelry/data/mock_data/items.csv

@@ -0,0 +1,20 @@
+item_id,date_added,description,colors,metals,stones,stones_carat_total_weight,product_id,image
+1,2024-02-15,"A luxurious yellow gold ring adorned with sparkling diamonds, exuding timeless elegance and sophistication.","yellow, clear",yellow gold,diamonds,1.5,1,1.png
+2,2024-02-15,"A mesmerizing white gold ring featuring a stunning arrangement of diamonds, symbolizing everlasting beauty and opulence.","white, clear",white gold,diamonds,1.5,1,2.png
+3,2024-02-20,"A radiant pink gold bracelet, exuding regal charm and sophistication.",pink,pink gold,no stones,0.0,2,3.png
+4,2024-02-20,"An exquisite white gold bracelet, epitomizing luxury and grace.",white,white gold,no stones,0.0,2,4.png
+5,2024-02-25,"A majestic yellow gold necklace adorned with glistening diamonds, exuding royal elegance and grandeur.","yellow, clear",yellow gold,diamonds,3.0,3,5.png
+6,2024-02-25,"An opulent white gold necklace featuring a breathtaking arrangement of diamonds, capturing the essence of luxury and sophistication.","white, clear",white gold,diamonds,3.0,3,6.png
+7,2024-03-01,"Luxurious pink gold earrings adorned with sparkling diamonds, radiating opulent glamour and sophistication.","pink, clear",pink gold,diamonds,2.2,4,7.png
+8,2024-03-01,"Elegant white gold earrings featuring a dazzling array of diamonds, exuding timeless beauty and refinement.","white, clear",white gold,diamonds,2.2,4,8.png
+9,2024-02-15,"A timeless yellow gold ring adorned with sparkling diamonds, radiating elegance and grace.","yellow, clear",yellow gold,diamonds,1.0,5,9.png
+10,2024-02-15,"An exquisite pink gold ring featuring a delicate arrangement of diamonds, epitomizing sophistication and charm.","pink, clear",pink gold,diamonds,1.0,5,10.png
+11,2024-03-04,"An elegant white gold necklace adorned with a breathtaking butterfly pendant, symbolizing transformation and beauty.",white,white gold,no stones,0.0,6,11.png
+12,2024-03-04,"An elegant pink gold necklace adorned with a breathtaking butterfly pendant, symbolizing transformation and beauty.",pink,pink gold,no stones,0.0,6,12.png
+13,2024-03-04,"A mesmerizing white gold necklace featuring delicate leaf motifs and sparkling diamond accents, evoking the enchanting beauty of a mystical forest.",white,white gold,diamonds,0.7,7,13.png
+14,2024-03-04,"A mesmerizing yellow gold necklace featuring delicate leaf motifs and sparkling diamond accents, evoking the enchanting beauty of a mystical forest.",yellow,yellow gold,diamonds,0.7,7,14.png
+15,2024-02-15,"An enchanting white gold ring adorned with a heart shaped diamond, symbolizing eternal love and romance.","white, clear",white gold,diamonds,1.8,8,15.png
+16,2024-02-15,"An enchanting pink gold ring adorned with a heart shaped diamond, symbolizing eternal love and romance.","pink, clear",pink gold,diamonds,1.8,8,16.png
+17,2024-02-15,"An enchanting yellow gold ring adorned with a heart shaped diamond, symbolizing eternal love and romance.","yellow, clear",yellow gold,diamonds,1.8,8,17.png
+18,2024-02-25,"An exquisite chain pink gold necklace, capturing the essence of timeless elegance and sophistication.",pink,pink gold,no stones,0.0,9,18.png
+19,2024-02-25,"An exquisite chain yellow gold necklace, capturing the essence of timeless elegance and sophistication.",yellow,yellow gold,no stones,0.0,9,19.png

jewelry/data/mock_data/mocker.py

@@ -0,0 +1,153 @@
+import datetime
+import json
+import os
+import random
+import uuid
+from typing import List, Tuple
+
+import pandas as pd
+
+
+def _generate_stocks(
+    products: List[dict], items: List[dict], min_amount: int = 0, max_amount: int = 5
+) -> List[dict]:
+    price_map = {"rings": 400, "bracelets": 800, "necklaces": 1000, "earrings": 200}
+    size_map = {
+        "rings": [3, 5, 7, 9],
+        "bracelets": [6.3, 6.7, 7.1, 7.9, 9],
+        "necklaces": [16, 17.7, 19.4],
+        "earrings": [1],
+    }
+
+    stocks = []
+    for item in items:
+        kind = next(p["kind"] for p in products if item["product_id"] == p["product_id"])
+        price = random.randint(1, 10) * 50 + price_map[kind]
+        if item["stones"] == "diamonds":
+            price *= 3
+        if item["metals"] == "white gold":
+            price += 200
+        for size in size_map[kind]:
+            stocks.append(
+                {
+                    "stock_id": str(uuid.uuid4()).replace("-", ""),
+                    "item_id": item["item_id"],
+                    "size": size,
+                    "amount": random.randint(min_amount, max_amount),
+                    "price": price,
+                }
+            )
+
+    return stocks
+
+
+def _generate_purchases(
+    users: List[dict],
+    stocks: List[dict],
+    min_date: str,
+    max_date: str,
+    min_amount: int = 1,
+    max_amount: int = 4,
+) -> Tuple[List[dict], List[dict]]:
+    min_date = datetime.datetime.strptime(min_date, "%m/%d/%Y")
+    max_date = datetime.datetime.strptime(max_date, "%m/%d/%Y")
+
+    purchases = []
+    for _ in range(10):
+        user_id = random.choice(users)["user_id"]
+        stock_ids = [
+            stock["stock_id"]
+            for stock in random.sample(stocks, random.randint(min_amount, max_amount))
+        ]
+        purchase_id = str(uuid.uuid4()).replace("-", "")
+        date = min_date + datetime.timedelta(
+            days=random.randint(0, (max_date - min_date).days)
+        )
+        purchases.append(
+            {
+                "user_id": user_id,
+                "stocks": stock_ids,
+                "date": date.strftime("%m/%d/%Y"),
+                "purchase_id": purchase_id,
+            }
+        )
+
+    stock_to_purchase = []
+    for p in purchases:
+        for sid in p["stocks"]:
+            stock_to_purchase.append({"purchase_id": p["purchase_id"], "stock_id": sid})
+        p.pop("stocks")
+
+    return purchases, stock_to_purchase
+
+
+def _generate_reviews(
+    stocks: List[dict],
+    purchases: List[dict],
+    stock_to_purchase: List[dict],
+    review_chance: float = 0.5
+) -> List[dict]:
+    reviews = []
+    for stp in stock_to_purchase:
+        if random.random() < review_chance:
+            continue
+        purchase = next(p for p in purchases if p["purchase_id"] == stp["purchase_id"])
+        stock = next(s for s in stocks if s["stock_id"] == stp["stock_id"])
+        date = datetime.datetime.strptime(purchase["date"], "%m/%d/%Y")
+        date += datetime.timedelta(days=random.randint(1, 7))
+        rating = random.randint(1, 5)
+        review = {
+            "review_id": str(uuid.uuid4()).replace("-", ""),
+            "item_id": stock["item_id"],
+            "user_id": purchase["user_id"],
+            "date": date,
+            "text": "",  # TODO: Generate a review based on randomize score using ChatGPT
+            "rating": rating,
+            "is_recommend": rating > 3,
+        }
+        reviews.append(review)
+
+    return reviews
+
+
+def generate_mock_data(
+    sources_directory: str = "./sources", output_directory: str = "./"
+):
+    with open(os.path.join(sources_directory, "products.json")) as json_file:
+        products = json.load(json_file)
+    with open(os.path.join(sources_directory, "items.json")) as json_file:
+        items = json.load(json_file)
+    with open(os.path.join(sources_directory, "users.json")) as json_file:
+        users = json.load(json_file)
+
+    stocks = _generate_stocks(products=products, items=items)
+    purchases, stock_to_purchase = _generate_purchases(
+        users=users, stocks=stocks, min_date="05/03/2024", max_date="07/03/2024"
+    )
+    reviews = _generate_reviews(stocks=stocks, purchases=purchases, stock_to_purchase=stock_to_purchase)
+
+    products_df = pd.DataFrame(products)
+    items_df = pd.DataFrame(items)
+    items_df['date_added'] = pd.to_datetime(items_df['date_added'], format='%m/%d/%Y')
+    users_df = pd.DataFrame(users)
+    users_df['date_of_birth'] = pd.to_datetime(users_df['date_of_birth'], format='%m/%d/%Y')
+    stocks_df = pd.DataFrame(stocks)
+    purchases_df = pd.DataFrame(purchases)
+    purchases_df['date'] = pd.to_datetime(purchases_df['date'], format='%m/%d/%Y')
+    stock_to_purchase_df = pd.DataFrame(stock_to_purchase)
+    reviews_df = pd.DataFrame(reviews)
+    reviews_df['date'] = pd.to_datetime(reviews_df['date'], format='%m/%d/%Y')
+
+    products_df.to_csv(os.path.join(output_directory, "products.csv"), index=False)
+    items_df.to_csv(os.path.join(output_directory, "items.csv"), index=False)
+    users_df.to_csv(os.path.join(output_directory, "users.csv"), index=False)
+    stocks_df.to_csv(os.path.join(output_directory, "stocks.csv"), index=False)
+    purchases_df.to_csv(os.path.join(output_directory, "purchases.csv"), index=False)
+    stock_to_purchase_df.to_csv(
+        os.path.join(output_directory, "stock_to_purchase.csv"), index=False
+    )
+    reviews_df.to_csv(os.path.join(output_directory, "reviews.csv"), index=False)
+
+
+if __name__ == "__main__":
+    generate_mock_data()

jewelry/data/mock_data/products.csv

@@ -0,0 +1,10 @@
+product_id,name,kind,collections,gifts
+1,Eternal Splendor Ring,rings,Forever Diamonds,"wife, anniversary"
+2,Regal Enchantment Bracelet,bracelets,McKinsey Elegance,"wife, anniversary"
+3,Royal Splendor Necklace,necklaces,Forever Diamonds,"wife, anniversary"
+4,Opulent Glamour Earrings,earrings,Forever Diamonds,"wife, anniversary"
+5,Timeless Grace Ring,rings,Forever Diamonds,"wife, anniversary"
+6,Enchanted Forest Necklace,necklaces,Iguazio Nature Capture,"mom, birthday"
+7,Enchanted Leaf Necklace,earrings,"Iguazio Nature Capture, Forever Diamonds","best friend, mom, birthday"
+8,Enchanted Love Ring,rings,Forever Diamonds,"wife, anniversary, best friend"
+9,Majestic Chain Necklace,necklaces,Timeless,"wife, graduation, best friend, mom"

jewelry/data/mock_data/purchases.csv

@@ -0,0 +1,11 @@
+user_id,date,purchase_id
+5,2024-05-11,e12e1d1139e3481b8aa53a88cb95a156
+2,2024-05-20,1abc839fe844479f9154e5c683264436
+1,2024-05-29,773ca8faaf17471a82c0792e2ffeb661
+3,2024-07-01,02a941880e674f82ae0430c404c7f62b
+3,2024-05-09,72d033b842474d96aaaaf94153a6dc45
+3,2024-05-03,f469aa686ea24023b0241786068eab1f
+1,2024-05-04,8751e73e3f4b4b758931de66506f71a7
+3,2024-05-11,80d7415470294775aec85bffc41f852d
+4,2024-06-13,d4592d31cafd44cb8a64a44e64aff624
+3,2024-06-29,f315be28cd9f469ea713116faf49b123

jewelry/data/mock_data/reviews.csv

@@ -0,0 +1,16 @@
+review_id,item_id,user_id,date,text,rating,is_recommend
+75bcf4035378468d8251a58519b05e3e,14,5,2024-05-16,,5,True
+28ec7b810ad54bb88907344646107949,2,5,2024-05-18,,4,True
+14d7c597917a440c9bff6a8d510a4b9f,6,2,2024-05-24,,4,True
+d622a7d9a3884e2f91e05cfbca053c7a,10,2,2024-05-23,,3,False
+28bde0397b9a4336929f77b55a3a1bae,2,2,2024-05-25,,5,True
+66300279773e40709cc4348da4a814d6,10,1,2024-06-02,,1,False
+4c7a9c600eef46cfad6322ed2e14cfa2,6,3,2024-07-02,,1,False
+97b4a35f9cdf48479ae3361ecbcbada8,12,3,2024-05-06,,4,True
+d7cfcc1c2caa47749439cce3312fd32b,6,3,2024-05-10,,3,False
+cd25c018501040a18bf35891561b4e84,9,3,2024-05-07,,2,False
+b06e34a24c454af582780b017af37c4c,5,1,2024-05-11,,3,False
+a6195ed5ccbd4216b9bd34f79542d00a,17,3,2024-05-13,,4,True
+dfd00d8808f44eb081a760688ba0de0a,6,4,2024-06-15,,4,True
+fc437a4a1a0645568aca999c29fbb72b,1,3,2024-07-06,,3,False
+8da2df90c7c641a88c5b5a174c5386a7,6,3,2024-07-04,,5,True