# main.py

# A demonstration of fast keyword searching using the FlashText library.

#

# Before running, you will need to install flashtext:

# pip install flashtext

from flashtext import KeywordProcessor

print("--- Starting Fast Keyword Matching Demonstration with FlashText ---")

# --- Section 1: Define Keywords and Text ---

# We define a set of keywords we want to find and a body of text to search within.

# FlashText can handle hundreds of thousands of keywords efficiently.

keyword_dict = {

    "data science": ["Data Science", "data scientist"],

    "machine learning": ["Machine Learning", "ML"],

    "artificial intelligence": ["Artificial Intelligence", "AI", "A.I."],

    "new york": ["New York City", "NYC"]

}

text_to_search = """

Dr. Alice Smith, a brilliant data scientist from Google, is presenting her 

groundbreaking research on Artificial Intelligence in New York City today. 

Her work in ML has been pivotal.

"""

print("\n--- 1. Text to be Searched ---")

print(text_to_search)

print("\n--- Keywords to Find ---")

print(keyword_dict)

# --- Section 2: Initialize Keyword Processor and Add Keywords ---

# We create a KeywordProcessor and add our dictionary of keywords to it.

# The dictionary keys are the "clean" names, and the values are the different

# variations we want to match.

print("\n--- 2. Initializing FlashText Keyword Processor ---")

keyword_processor = KeywordProcessor()

keyword_processor.add_keywords_from_dict(keyword_dict)

print("Keyword processor is ready.")

# --- Section 3: Extract Keywords from Text ---

# The extract_keywords method finds all occurrences of the added keywords in the text.

print("\n--- 3. Extracting Keywords ---")

found_keywords = keyword_processor.extract_keywords(text_to_search)

print("\nFound keywords (clean names):")

print(found_keywords)

# To get more details, like the original word and its position:

print("\n--- 4. Extracting Keywords with Span Information ---")

found_keywords_with_span = keyword_processor.extract_keywords(text_to_search, span_info=True)

print("\nFound keywords with their start and end positions:")

for keyword, start, end in found_keywords_with_span:

    print(f"- Found '{keyword}' (Original text: '{text_to_search[start:end]}') at index {start}:{end}")

print("\n--- End of Demonstration ---")