Merge pull request #8 from am9zZWY/intro-duckdb

Add duckdb database

engine/crawl.py

@@ -17,6 +17,7 @@
 ##### Language detection #####
 from eld import LanguageDetector
 ##### Database #####
+import duckdb
 from custom_db import *
 
 ##### Constants #####
@@ -45,9 +46,12 @@
 
 
 class Crawler(PipelineElement):
-    def __init__(self):
+    def __init__(self, dbcon: duckdb.DuckDBPyConnection):
         super().__init__("Crawler")
 
+        # Initialize the duckdb connection
+        self.cursor = dbcon.cursor()
+
         # Initialize the crawler state
         self.found_links = set()
         self.ignore_links = set()
@@ -65,6 +69,9 @@ def __init__(self):
         self.required_keywords = ["tübingen", "tuebingen", "tubingen", "t%C3%BCbingen"]
         self.user_agent = "Modern Search Engines University of Tuebingen Project Crawler (https://uni-tuebingen.de/de/262377)"
 
+    def __del__(self) -> None:
+        self.cursor.close()
+
     async def fetch(self, session, url):
         headers = {
             "User-Agent": self.user_agent,
@@ -191,10 +198,10 @@ def _load_state(self):
 
 # IMPORTANT: Please use main.py instead of this file
 if __name__ == "__main__":
-    crawler = Crawler()
+    con = duckdb.connect("crawlies.db")
+    con.install_extension("fts")
+    con.load_extension("fts")
+
+    crawler = Crawler(con)
     crawler.process()
-    # TODO - seperarw crawling and tokenizing
-    index_pages()
-    index_df = access_index()
-    index_df.to_csv("inverted_index.csv")
-    save_pages()
+    con.close()

engine/main.py

@@ -3,18 +3,27 @@
 """
 from concurrent.futures import ThreadPoolExecutor
 
+# Database
+import duckdb
+# Pipeline
 from crawl import Crawler
 from custom_db import index_pages, access_index, save_pages
 from custom_tokenizer import Tokenizer
+# Async
 import asyncio
 
 MAX_THREADS = 10
 
 if __name__ == "__main__":
+    con = duckdb.connect("crawlies.db")
     try:
+        # Database setup
+        con.install_extension("fts")
+        con.load_extension("fts")
+
         # Initialize the pipeline
         with ThreadPoolExecutor(max_workers=MAX_THREADS) as executor:
-            crawler = Crawler()
+            crawler = Crawler(con)
             crawler.max_size = 1000
             crawler.add_executor(executor)
 
@@ -29,9 +38,13 @@
     except (KeyboardInterrupt, SystemExit):
         print("Exiting...")
         crawler.save_state()
+        con.close()
         print("State saved")
 
     index_pages()
     index_df = access_index()
     index_df.to_csv("inverted_index.csv")
     save_pages()
+
+    # Close the connection
+    con.close()

engine/requirements.txt

@@ -2,6 +2,7 @@ beautifulsoup4==4.12.3
 eld==1.0.6
 Flask==3.0.3
 lxml==5.2.2
+duckdb==1.0.0
 nltk==3.8.1
 pandas==2.2.2
 requests==2.32.3