From 0a7a7bc0d154e7a7eaf0f5682b9a3109f017e413 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Marko=20Budiseli=C4=87?= <marko.budiselic@memgraph.com>
Date: Sat, 13 Jan 2024 08:43:33 +0100
Subject: [PATCH] Add Tantivy ADR (#1633)

---
 ADRs/001_tantivy.md | 32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)
 create mode 100644 ADRs/001_tantivy.md

diff --git a/ADRs/001_tantivy.md b/ADRs/001_tantivy.md
new file mode 100644
index 000000000..ac887c861
--- /dev/null
+++ b/ADRs/001_tantivy.md
@@ -0,0 +1,32 @@
+# Tantivy ADR
+
+**Author**
+Marko Budiselic (github.com/gitbuda)
+
+**Status**
+APPROVED
+
+**Date**
+January 5, 2024
+
+**Problem**
+
+For some of Memgraph workloads, text search is a required feature. We don't
+want to build a new text search engine because that's not Memgraph's core
+value.
+
+**Criteria**
+
+- easy integration with our C++ codebase
+- ability to operate in-memory and on-disk
+- sufficient features (regex, full-text search, fuzzy search, aggregations over
+  text data)
+- production-ready
+
+**Decision**
+
+All known C++ libraries are not production-ready. Recent Rust libraries, in
+particular [Tantivy](https://github.com/quickwit-oss/tantivy), seem to provide
+much more features, it is production ready. The way how we'll integrate Tantivy
+into the current Memgraph codebase is via
+[cxx](https://github.com/dtolnay/cxx). **We select Tantivy.**