diff --git a/docs/tutorials.md b/docs/tutorials.md
index 8388b98..2e7cc39 100644
--- a/docs/tutorials.md
+++ b/docs/tutorials.md
@@ -64,8 +64,12 @@ writer.add_document(tantivy.Document(
 ))
 # ... and committing
 writer.commit()
+writer.wait_merging_threads()
 ```
 
+Note that `wait_merging_threads()` must come at the end, because
+the `writer` object will not be usable after this call.
+
 ## Building and Executing Queries
 
 First you need to get a searcher for the index
diff --git a/src/snippet.rs b/src/snippet.rs
index bb19d82..dcd358f 100644
--- a/src/snippet.rs
+++ b/src/snippet.rs
@@ -2,10 +2,10 @@ use crate::to_pyerr;
 use pyo3::prelude::*;
 use tantivy as tv;
 
-/// Tantivy schema.
+/// Tantivy Snippet
 ///
-/// The schema is very strict. To build the schema the `SchemaBuilder` class is
-/// provided.
+/// Snippet contains a fragment of a document, and some highlighted
+/// parts inside it.
 #[pyclass(module = "tantivy.tantivy")]
 pub(crate) struct Snippet {
     pub(crate) inner: tv::Snippet,
diff --git a/tests/conftest.py b/tests/conftest.py
index af64cb3..313fdba 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -73,6 +73,7 @@ def create_index(dir=None):
         }"""
     )
     writer.commit()
+    writer.wait_merging_threads()
     index.reload()
     return index
 
@@ -117,6 +118,7 @@ def create_index_with_numeric_fields(dir=None):
     )
     writer.add_document(doc)
     writer.commit()
+    writer.wait_merging_threads()
     index.reload()
     return index
 
@@ -167,6 +169,7 @@ def create_spanish_index():
         }"""
     )
     writer.commit()
+    writer.wait_merging_threads()
     index.reload()
     return index