Add S3 Object Lambda ARN and S3 Outposts ARN (#686)

Co-authored-by: Martin Durant <martin.durant@alumni.utoronto.ca>
fsspec · Jan 13, 2023 · 11c4ff5 · 11c4ff5
1 parent 9bf99f7
commit 11c4ff5
Show file tree

Hide file tree

Showing 2 changed files with 27 additions and 15 deletions.
diff --git a/s3fs/core.py b/s3fs/core.py
@@ -377,20 +377,29 @@ def _find_bucket_key(self, s3_path):
         It will return the bucket and the key represented by the s3 path
         """
 
-        _S3_ACCESSPOINT_TO_BUCKET_KEY_REGEX = re.compile(
-            r"^(?P<bucket>arn:(aws).*:s3:[a-z\-0-9]*:[0-9]{12}:accesspoint[:/][^/]+)/?"
-            r"(?P<key>.*)$"
-        )
-        _S3_OUTPOST_TO_BUCKET_KEY_REGEX = re.compile(
-            r"^(?P<bucket>arn:(aws).*:s3-outposts:[a-z\-0-9]+:[0-9]{12}:outpost[/:]"
-            r"[a-zA-Z0-9\-]{1,63}[/:](bucket|accesspoint)[/:][a-zA-Z0-9\-]{1,63})[/:]?(?P<key>.*)$"
-        )
-        match = _S3_ACCESSPOINT_TO_BUCKET_KEY_REGEX.match(s3_path)
-        if match:
-            return match.group("bucket"), match.group("key")
-        match = _S3_OUTPOST_TO_BUCKET_KEY_REGEX.match(s3_path)
-        if match:
-            return match.group("bucket"), match.group("key")
+        bucket_format_list = [
+            re.compile(
+                r"^(?P<bucket>arn:(aws).*:s3:[a-z\-0-9]*:[0-9]{12}:accesspoint[:/][^/]+)/?"
+                r"(?P<key>.*)$"
+            ),
+            re.compile(
+                r"^(?P<bucket>arn:(aws).*:s3-outposts:[a-z\-0-9]+:[0-9]{12}:outpost[/:]"
+                r"[a-zA-Z0-9\-]{1,63}[/:](bucket|accesspoint)[/:][a-zA-Z0-9\-]{1,63})[/:]?(?P<key>.*)$"
+            ),
+            re.compile(
+                r"^(?P<bucket>arn:(aws).*:s3-outposts:[a-z\-0-9]+:[0-9]{12}:outpost[/:]"
+                r"[a-zA-Z0-9\-]{1,63}[/:]bucket[/:]"
+                r"[a-zA-Z0-9\-]{1,63})[/:]?(?P<key>.*)$"
+            ),
+            re.compile(
+                r"^(?P<bucket>arn:(aws).*:s3-object-lambda:[a-z\-0-9]+:[0-9]{12}:"
+                r"accesspoint[/:][a-zA-Z0-9\-]{1,63})[/:]?(?P<key>.*)$"
+            ),
+        ]
+        for bucket_format in bucket_format_list:
+            match = bucket_format.match(s3_path)
+            if match:
+                return match.group("bucket"), match.group("key")
         s3_components = s3_path.split("/", 1)
         bucket = s3_components[0]
         s3_key = ""
@@ -1544,7 +1553,9 @@ async def _chmod(self, path, acl, recursive=False, **kwargs):
         bucket, key, version_id = self.split_path(path)
         if recursive:
             allfiles = await self._find(path, withdirs=False)
-            await asyncio.gather(*[self._chmod(p, acl, recursive=False) for p in allfiles])
+            await asyncio.gather(
+                *[self._chmod(p, acl, recursive=False) for p in allfiles]
+            )
         elif key:
             if acl not in key_acls:
                 raise ValueError("ACL not in %s", key_acls)

diff --git a/s3fs/tests/test_s3fs.py b/s3fs/tests/test_s3fs.py
@@ -2433,6 +2433,7 @@ def test_split_path(s3):
         "arn:aws:s3:region:123456789012:accesspoint/my-access-point-name",
         "arn:aws:s3-outposts:region:123456789012:outpost/outpost-id/bucket/my-test-bucket",
         "arn:aws:s3-outposts:region:123456789012:outpost/outpost-id/accesspoint/my-accesspoint-name",
+        "arn:aws:s3-object-lambda:region:123456789012:accesspoint/my-lambda-object-name",
     ]
     test_key = "my/test/path"
     for test_bucket in buckets: