Add support for Oracle Cloud Infrastructure (OCI) (#4244)

* [feat]: Add support for Oracle Cloud Infrastructure (OCI) Object Storage filesystem Signed-off-by: Pedro Antonacio <[email protected]> * [test]: add test case for OCI protocol Signed-off-by: Pedro Antonacio <[email protected]> * update upcoming release notes with the proposed solution Signed-off-by: Pedro Antonacio <[email protected]> --------- Signed-off-by: Pedro Antonacio <[email protected]>
kedro-org · Oct 22, 2024 · 6836150 · 6836150
1 parent 9a0a779
commit 6836150
Show file tree

Hide file tree

Showing 3 changed files with 9 additions and 3 deletions.
diff --git a/RELEASE.md b/RELEASE.md
@@ -6,6 +6,8 @@
 **Note:** ``KedroDataCatalog`` is an experimental feature and is under active development. Therefore, it is possible we'll introduce breaking changes to this class, so be mindful of that if you decide to use it already. Let us know if you have any feedback about the ``KedroDataCatalog`` or ideas for new features.
 
 ## Bug fixes and other changes
+* Added I/O support for Oracle Cloud Infrastructure (OCI) Object Storage filesystem
+
 ## Breaking changes to the API
 ## Documentation changes
 * Updated CLI autocompletion docs with new Click syntax.

diff --git a/kedro/io/core.py b/kedro/io/core.py
@@ -51,6 +51,7 @@
     "gcs",
     "gdrive",
     "gs",
+    "oci",
     "oss",
     "s3",
     "s3a",
@@ -820,9 +821,11 @@ def _parse_filepath(filepath: str) -> dict[str, str]:
         host_with_port = parsed_path.netloc.rsplit("@", 1)[-1]
         host = host_with_port.rsplit(":", 1)[0]
         options["path"] = host + options["path"]
-        # Azure Data Lake Storage Gen2 URIs can store the container name in the
-        # 'username' field of a URL (@ syntax), so we need to add it to the path
-        if protocol == "abfss" and parsed_path.username:
+        # - Azure Data Lake Storage Gen2 URIs can store the container name in the
+        #   'username' field of a URL (@ syntax), so we need to add it to the path
+        # - Oracle Cloud Infrastructure (OCI) Object Storage filesystem (ocifs) also
+        #   uses the @ syntax for I/O operations: "oci://bucket@namespace/path_to_file"
+        if protocol in ["abfss", "oci"] and parsed_path.username:
             options["path"] = parsed_path.username + "@" + options["path"]
 
     return options

diff --git a/tests/io/test_core.py b/tests/io/test_core.py
@@ -263,6 +263,7 @@ def test_get_filepath_str(self):
                 "abfss://[email protected]/mypath",
                 ("abfss", "[email protected]/mypath"),
             ),
+            ("oci://bucket@namespace/file.txt", ("oci", "bucket@namespace/file.txt")),
             ("hdfs://namenode:8020/file.txt", ("hdfs", "/file.txt")),
             ("file:///tmp/file.txt", ("file", "/tmp/file.txt")),
             ("/tmp/file.txt", ("file", "/tmp/file.txt")),