From d6f9a60a322998590bceb5d095c74f63e3077414 Mon Sep 17 00:00:00 2001
From: ManyTheFish <many@meilisearch.com>
Date: Wed, 3 Aug 2022 11:38:40 +0200
Subject: [PATCH] fix: Remove whitespace trimming during document id validation

fix #592
---
 milli/src/update/index_documents/enrich.rs |  2 -
 milli/src/update/index_documents/mod.rs    | 47 ++++++++++++++++++++++
 2 files changed, 47 insertions(+), 2 deletions(-)

diff --git a/milli/src/update/index_documents/enrich.rs b/milli/src/update/index_documents/enrich.rs
index 7c9a016d8..15fbe9319 100644
--- a/milli/src/update/index_documents/enrich.rs
+++ b/milli/src/update/index_documents/enrich.rs
@@ -294,9 +294,7 @@ pub fn fetch_matching_values_in_object(
     }
 }
 
-/// Returns a trimmed version of the document id or `None` if it is invalid.
 pub fn validate_document_id(document_id: &str) -> Option<&str> {
-    let document_id = document_id.trim();
     if !document_id.is_empty()
         && document_id.chars().all(|c| matches!(c, 'a'..='z' | 'A'..='Z' | '0'..='9' | '-' | '_'))
     {
diff --git a/milli/src/update/index_documents/mod.rs b/milli/src/update/index_documents/mod.rs
index c9890f93f..0f0eaca5a 100644
--- a/milli/src/update/index_documents/mod.rs
+++ b/milli/src/update/index_documents/mod.rs
@@ -2086,4 +2086,51 @@ mod tests {
         let (_builder, user_error) = builder.add_documents(doc4).unwrap();
         assert!(user_error.is_err());
     }
+
+    #[test]
+    fn primary_key_must_not_contain_whitespace() {
+        let tmp = tempfile::tempdir().unwrap();
+        let mut options = EnvOpenOptions::new();
+        options.map_size(4096 * 100);
+        let index = Index::new(options, tmp).unwrap();
+        let mut wtxn = index.write_txn().unwrap();
+        let indexer_config = IndexerConfig::default();
+        let builder = IndexDocuments::new(
+            &mut wtxn,
+            &index,
+            &indexer_config,
+            IndexDocumentsConfig::default(),
+            |_| (),
+        )
+        .unwrap();
+
+        let doc1 = documents! {[{
+            "id": " 1",
+            "title": "asdsad",
+        }]};
+
+        let doc2 = documents! {[{
+            "id": "\t2",
+            "title": "something",
+        }]};
+
+        let doc3 = documents! {[{
+            "id": "\r3",
+            "title": "something",
+        }]};
+
+        let doc4 = documents! {[{
+            "id": "\n4",
+            "title": "something",
+        }]};
+
+        let (builder, user_error) = builder.add_documents(doc1).unwrap();
+        assert!(user_error.is_err());
+        let (builder, user_error) = builder.add_documents(doc2).unwrap();
+        assert!(user_error.is_err());
+        let (builder, user_error) = builder.add_documents(doc3).unwrap();
+        assert!(user_error.is_err());
+        let (_builder, user_error) = builder.add_documents(doc4).unwrap();
+        assert!(user_error.is_err());
+    }
 }