init commit, basic tokenizing and indexing logic without any ranking of results on a simple wikipedia dump

1 files changed, 47 insertions, 0 deletions

diff --git a/main.go b/main.go
new file mode 100644
index 0000000..459da39
--- /dev/null
+++ b/main.go
@@ -0,0 +1,47 @@
+package main
+
+import (
+	"fmt"
+	"searchEngine/analyzer"
+	"searchEngine/index"
+	"searchEngine/ingest"
+	"searchEngine/search"
+	"sync"
+)
+
+func main() {
+	myIndex := index.New()
+	jobQueue := make(chan ingest.Page, 100)
+
+	var wg sync.WaitGroup
+
+	for range 4 {
+		wg.Go(func() {
+			for page := range jobQueue {
+				tokens := analyzer.ProcessText(page.Text)
+				myIndex.Add(page.Id, page.Title, tokens)
+			}
+		})
+	}
+
+	err := ingest.IngestWikiDump("C:/Users/bambi15/Documents/wikipedia/simplewiki-2026-07-01-p1p1273731/simplewiki-2026-07-01-p1p1273731.xml", jobQueue)
+	if err != nil {
+		panic(err)
+	}
+
+	close(jobQueue)
+	wg.Wait()
+	fmt.Println("Indexing complete! Running a test search...")
+
+	results := search.Search(myIndex, "computer science")
+
+	fmt.Printf("Found %d articles matching your query:\n", len(results))
+	for i, title := range results {
+		if i >= 10 {
+			fmt.Println("... and more.")
+			break
+		}
+		fmt.Printf("- %s\n", title)
+	}
+
+}