feat: Add ByteStream to_string method (#7009)

haystack/dataclasses/byte_stream.py

@@ -49,3 +49,13 @@ class ByteStream:
         :param meta: Additional metadata to be stored with the ByteStream.
         """
         return cls(data=text.encode(encoding), mime_type=mime_type, meta=meta or {})
+
+    def to_string(self, encoding: str = "utf-8") -> str:
+        """
+        Convert the ByteStream to a string, metadata will not be included.
+
+        :param encoding: The encoding used to convert the bytes to a string. Defaults to "utf-8".
+        :return: The string representation of the ByteStream.
+        :raises UnicodeDecodeError: If the ByteStream data cannot be decoded with the specified encoding.
+        """
+        return self.data.decode(encoding)

test/dataclasses/test_byte_stream.py

@@ -1,3 +1,5 @@
+import pytest
+
 from haystack.dataclasses import ByteStream
 
 
@@ -35,6 +37,30 @@ def test_from_string():
     assert b.meta == {"foo": "bar"}
 
 
+def test_to_string():
+    test_string = "Hello, world!"
+    b = ByteStream.from_string(test_string)
+    assert b.to_string() == test_string
+
+
+def test_to_from_string_encoding():
+    test_string = "Hello Baščaršija!"
+    with pytest.raises(UnicodeEncodeError):
+        ByteStream.from_string(test_string, encoding="ISO-8859-1")
+
+    bs = ByteStream.from_string(test_string)  # default encoding is utf-8
+
+    assert bs.to_string(encoding="ISO-8859-1") != test_string
+    assert bs.to_string(encoding="utf-8") == test_string
+
+
+def test_to_string_encoding_error():
+    # test that it raises ValueError if the encoding is not valid
+    b = ByteStream.from_string("Hello, world!")
+    with pytest.raises(UnicodeDecodeError):
+        b.to_string("utf-16")
+
+
 def test_to_file(tmp_path, request):
     test_str = "Hello, world!\n"
     test_path = tmp_path / request.node.name