Add support for preserving base64 encoded images (#1140)

* optional reserve base64 string in markdown _CustomMarkdownify and pptx
* add other converter para support
* fix linter
* Use *kwarg to pass keep_data_uri para.
* Add module cli vector tests
* Fixed formatting, and adjusted tests.

packages/markitdown/tests/test_module_vectors.py

@@ -6,9 +6,9 @@ import codecs
 
 
 if __name__ == "__main__":
-    from _test_vectors import GENERAL_TEST_VECTORS
+    from _test_vectors import GENERAL_TEST_VECTORS, DATA_URI_TEST_VECTORS
 else:
-    from ._test_vectors import GENERAL_TEST_VECTORS
+    from ._test_vectors import GENERAL_TEST_VECTORS, DATA_URI_TEST_VECTORS
 
 from markitdown import (
     MarkItDown,
@@ -124,10 +124,52 @@ def test_convert_url(test_vector):
         assert string not in result.markdown
 
 
+@pytest.mark.parametrize("test_vector", DATA_URI_TEST_VECTORS)
+def test_convert_with_data_uris(test_vector):
+    """Test API functionality when keep_data_uris is enabled"""
+    markitdown = MarkItDown()
+
+    # Test local file conversion
+    result = markitdown.convert(
+        os.path.join(TEST_FILES_DIR, test_vector.filename),
+        keep_data_uris=True,
+        url=test_vector.url,
+    )
+
+    for string in test_vector.must_include:
+        assert string in result.markdown
+    for string in test_vector.must_not_include:
+        assert string not in result.markdown
+
+
+@pytest.mark.parametrize("test_vector", DATA_URI_TEST_VECTORS)
+def test_convert_stream_with_data_uris(test_vector):
+    """Test the conversion of a stream with no stream info."""
+    markitdown = MarkItDown()
+
+    stream_info = StreamInfo(
+        extension=os.path.splitext(test_vector.filename)[1],
+        mimetype=test_vector.mimetype,
+        charset=test_vector.charset,
+    )
+
+    with open(os.path.join(TEST_FILES_DIR, test_vector.filename), "rb") as stream:
+        result = markitdown.convert(
+            stream, stream_info=stream_info, keep_data_uris=True, url=test_vector.url
+        )
+
+        for string in test_vector.must_include:
+            assert string in result.markdown
+        for string in test_vector.must_not_include:
+            assert string not in result.markdown
+
+
 if __name__ == "__main__":
     import sys
 
     """Runs this file's tests from the command line."""
+
+    # General tests
     for test_function in [
         test_guess_stream_info,
         test_convert_local,
@@ -141,4 +183,17 @@ if __name__ == "__main__":
             )
             test_function(test_vector)
             print("OK")
+
+    # Data URI tests
+    for test_function in [
+        test_convert_with_data_uris,
+        test_convert_stream_with_data_uris,
+    ]:
+        for test_vector in DATA_URI_TEST_VECTORS:
+            print(
+                f"Running {test_function.__name__} on {test_vector.filename}...", end=""
+            )
+            test_function(test_vector)
+            print("OK")
+
     print("All tests passed!")