diff --git a/dictdatabase/io_bytes.py b/dictdatabase/io_bytes.py
index 7cc0edb..d5a87a9 100644
--- a/dictdatabase/io_bytes.py
+++ b/dictdatabase/io_bytes.py
@@ -4,7 +4,7 @@
 
 
 
-def read(db_name: str, start=None, end=None) -> bytes:
+def read(db_name: str, start: int = None, end: int = None) -> bytes:
 	"""
 		Read the content of a file as bytes. Reading works even when the config
 		changes, so a compressed ddb file can also be read if compression is
@@ -13,10 +13,11 @@ def read(db_name: str, start=None, end=None) -> bytes:
 		Note: Only specify either both start and end, or none of them.
 
 		Args:
-		- `db_name`: The name of the database to read from.
-		- `start`: The start index to read from.
-		- `end`: The end index to read up to (not included).
+		- `db_name`: The name of the database file to read from.
+		- `start`: The start byte index to read from.
+		- `end`: The end byte index to read up to (not included).
 	"""
+
 	json_path, json_exists, ddb_path, ddb_exists = utils.file_info(db_name)
 
 	if json_exists:
@@ -39,7 +40,7 @@ def read(db_name: str, start=None, end=None) -> bytes:
 
 
 
-def write(db_name: str, dump: bytes, start=None):
+def write(db_name: str, dump: bytes, start: int = None):
 	"""
 		Write the bytes to the file of the db_path. If the db was compressed but no
 		compression is enabled, remove the compressed file, and vice versa.
@@ -48,6 +49,8 @@ def write(db_name: str, dump: bytes, start=None):
 		- `db_name`: The name of the database to write to.
 		- `dump`: The bytes to write to the file, representing correct JSON when
 		decoded.
+		- `start`: The start byte index to write to. If None, the whole file is overwritten.
+		If the original content was longer, the rest truncated.
 	"""
 
 	json_path, json_exists, ddb_path, ddb_exists = utils.file_info(db_name)
diff --git a/tests/test_io_bytes.py b/tests/test_io_bytes.py
new file mode 100644
index 0000000..1e43841
--- /dev/null
+++ b/tests/test_io_bytes.py
@@ -0,0 +1,42 @@
+import pytest
+from dictdatabase import io_bytes
+
+
+
+def test_write_bytes(use_test_dir, name_of_test):
+    # Write shorter content at index
+    io_bytes.write(name_of_test, b"0123456789")
+    io_bytes.write(name_of_test, b"abc", 2)
+    assert io_bytes.read(name_of_test) == b"01abc"
+    # Overwrite with shorter content
+    io_bytes.write(name_of_test, b"xy")
+    assert io_bytes.read(name_of_test) == b"xy"
+    # Overwrite with longer content
+    io_bytes.write(name_of_test, b"0123456789")
+    io_bytes.write(name_of_test, b"abcdef", 8)
+    assert io_bytes.read(name_of_test) == b"01234567abcdef"
+    # Write at index out of range
+    io_bytes.write(name_of_test, b"01")
+    io_bytes.write(name_of_test, b"ab", 4)
+    assert io_bytes.read(name_of_test) == b'01\x00\x00ab'
+
+
+
+def test_read_bytes(use_test_dir, name_of_test, use_compression):
+    io_bytes.write(name_of_test, b"0123456789")
+    # In range
+    assert io_bytes.read(name_of_test, 2, 5) == b"234"
+    # Complete range
+    assert io_bytes.read(name_of_test, 0, 10) == b"0123456789"
+    assert io_bytes.read(name_of_test, 0, None) == b"0123456789"
+    assert io_bytes.read(name_of_test) == b"0123456789"
+    # End out of range
+    assert io_bytes.read(name_of_test, 9, 20) == b"9"
+    # Completely out of range
+    assert io_bytes.read(name_of_test, 25, 30) == b""
+    # Start negative
+    if use_compression:
+        assert io_bytes.read(name_of_test, -5, 3) == b""
+    else:
+        with pytest.raises(OSError):
+            io_bytes.read(name_of_test, -5, 3)