[arrow-rs] branch master updated: Support bucket name with `.` when parsing GCS URL (#4991) (#4992)

[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 570c91eb06 Support bucket name with `.` when parsing GCS URL (#4991) 
(#4992)
570c91eb06 is described below

commit 570c91eb06d792aa7bd912e5eff9ca0d5848a1bc
Author: Raphael Taylor-Davies <1781103+tustv...@users.noreply.github.com>
AuthorDate: Thu Oct 26 09:16:48 2023 +0100

    Support bucket name with `.` when parsing GCS URL (#4991) (#4992)
    
    * Support bucket name with `.` when parsing GCS URL (#4991)
    
    * Update test
---
 object_store/src/gcp/builder.rs | 18 ++++++------------
 object_store/src/parse.rs       |  4 ++++
 2 files changed, 10 insertions(+), 12 deletions(-)

diff --git a/object_store/src/gcp/builder.rs b/object_store/src/gcp/builder.rs
index 2039d23783..5f718d63d9 100644
--- a/object_store/src/gcp/builder.rs
+++ b/object_store/src/gcp/builder.rs
@@ -337,13 +337,8 @@ impl GoogleCloudStorageBuilder {
         let parsed = Url::parse(url).context(UnableToParseUrlSnafu { url })?;
         let host = parsed.host_str().context(UrlNotRecognisedSnafu { url })?;
 
-        let validate = |s: &str| match s.contains('.') {
-            true => Err(UrlNotRecognisedSnafu { url }.build()),
-            false => Ok(s.to_string()),
-        };
-
         match parsed.scheme() {
-            "gs" => self.bucket_name = Some(validate(host)?),
+            "gs" => self.bucket_name = Some(host.to_string()),
             scheme => return Err(UnknownUrlSchemeSnafu { scheme 
}.build().into()),
         }
         Ok(())
@@ -630,13 +625,12 @@ mod tests {
     fn gcs_test_urls() {
         let mut builder = GoogleCloudStorageBuilder::new();
         builder.parse_url("gs://bucket/path").unwrap();
-        assert_eq!(builder.bucket_name, Some("bucket".to_string()));
+        assert_eq!(builder.bucket_name.as_deref(), Some("bucket"));
 
-        let err_cases = ["mailto://bucket/path";, "gs://bucket.mydomain/path"];
-        let mut builder = GoogleCloudStorageBuilder::new();
-        for case in err_cases {
-            builder.parse_url(case).unwrap_err();
-        }
+        builder.parse_url("gs://bucket.mydomain/path").unwrap();
+        assert_eq!(builder.bucket_name.as_deref(), Some("bucket.mydomain"));
+
+        builder.parse_url("mailto://bucket/path";).unwrap_err();
     }
 
     #[test]
diff --git a/object_store/src/parse.rs b/object_store/src/parse.rs
index 170726f452..51993e2455 100644
--- a/object_store/src/parse.rs
+++ b/object_store/src/parse.rs
@@ -234,6 +234,10 @@ mod tests {
                 "gs://bucket/path",
                 (ObjectStoreScheme::GoogleCloudStorage, "path"),
             ),
+            (
+                "gs://test.example.com/path",
+                (ObjectStoreScheme::GoogleCloudStorage, "path"),
+            ),
             ("http://mydomain/path";, (ObjectStoreScheme::Http, "path")),
             ("https://mydomain/path";, (ObjectStoreScheme::Http, "path")),
         ];