(arrow-rs) branch master updated: Decode URL paths (#5017) (#5018)

[tustvold]

This is an automated email from the ASF dual-hosted git repository.

tustvold pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow-rs.git


The following commit(s) were added to refs/heads/master by this push:
     new ec788e15a9 Decode URL paths (#5017) (#5018)
ec788e15a9 is described below

commit ec788e15a99835376430b27617b1bb766709e05c
Author: Raphael Taylor-Davies <1781103+tustv...@users.noreply.github.com>
AuthorDate: Wed Nov 1 14:33:47 2023 +0000

    Decode URL paths (#5017) (#5018)
---
 object_store/src/parse.rs | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/object_store/src/parse.rs b/object_store/src/parse.rs
index 51993e2455..0fbc33c935 100644
--- a/object_store/src/parse.rs
+++ b/object_store/src/parse.rs
@@ -98,8 +98,7 @@ impl ObjectStoreScheme {
             _ => return Err(Error::Unrecognised { url: url.clone() }),
         };
 
-        let path = Path::parse(path)?;
-        Ok((scheme, path))
+        Ok((scheme, Path::from_url_path(path)?))
     }
 }
 
@@ -240,6 +239,18 @@ mod tests {
             ),
             ("http://mydomain/path";, (ObjectStoreScheme::Http, "path")),
             ("https://mydomain/path";, (ObjectStoreScheme::Http, "path")),
+            (
+                "s3://bucket/foo%20bar",
+                (ObjectStoreScheme::AmazonS3, "foo bar"),
+            ),
+            (
+                "https://foo/bar%20baz";,
+                (ObjectStoreScheme::Http, "bar baz"),
+            ),
+            (
+                "file:///bar%252Efoo",
+                (ObjectStoreScheme::Local, "bar%2Efoo"),
+            ),
         ];
 
         for (s, (expected_scheme, expected_path)) in cases {
@@ -260,4 +271,12 @@ mod tests {
             assert!(ObjectStoreScheme::parse(&url).is_err());
         }
     }
+
+    #[test]
+    fn test_url_spaces() {
+        let url = Url::parse("file:///my file with spaces").unwrap();
+        assert_eq!(url.path(), "/my%20file%20with%20spaces");
+        let (_, path) = parse_url(&url).unwrap();
+        assert_eq!(path.as_ref(), "my file with spaces");
+    }
 }