Fix filesystem

pyiceberg/io/pyarrow.py

@@ -388,14 +388,29 @@ def __init__(self, properties: Properties = EMPTY_DICT):
 
     @staticmethod
     def parse_location(location: str) -> Tuple[str, str, str]:
-        """Return the path without the scheme."""
+        """Return (scheme, netloc, path) for the given location.
+
+        Uses environment variables DEFAULT_SCHEME and DEFAULT_NETLOC
+        if scheme/netloc are missing.
+        """
         uri = urlparse(location)
-        if not uri.scheme:
-            return "file", uri.netloc, os.path.abspath(location)
-        elif uri.scheme in ("hdfs", "viewfs"):
-            return uri.scheme, uri.netloc, uri.path
+
+        # Load defaults from environment
+        default_scheme = os.getenv("DEFAULT_SCHEME", "file")
+        default_netloc = os.getenv("DEFAULT_NETLOC", "")
+
+        # Apply logic
+        scheme = uri.scheme or default_scheme
+        netloc = uri.netloc or default_netloc
+
+        if scheme in ("hdfs", "viewfs"):
+            return scheme, netloc, uri.path
         else:
-            return uri.scheme, uri.netloc, f"{uri.netloc}{uri.path}"
+            # For non-HDFS URIs, include netloc in the path if present
+            path = uri.path if uri.scheme else os.path.abspath(location)
+            if netloc and not path.startswith(netloc):
+                path = f"{netloc}{path}"
+            return scheme, netloc, path
 
     def _initialize_fs(self, scheme: str, netloc: Optional[str] = None) -> FileSystem:
         """Initialize FileSystem for different scheme."""

tests/io/test_pyarrow.py

@@ -2638,3 +2638,41 @@ def test_retry_strategy_not_found() -> None:
     io = PyArrowFileIO(properties={S3_RETRY_STRATEGY_IMPL: "pyiceberg.DoesNotExist"})
     with pytest.warns(UserWarning, match="Could not initialize S3 retry strategy: pyiceberg.DoesNotExist"):
         io.new_input("s3://bucket/path/to/file")
+
+
+def test_parse_location_environment_defaults() -> None:
+    """Test that parse_location uses environment variables for defaults."""
+    import os
+
+    from pyiceberg.io.pyarrow import PyArrowFileIO
+
+    # Test with default environment (no env vars set)
+    scheme, netloc, path = PyArrowFileIO.parse_location("/foo/bar")
+    assert scheme == "file"
+    assert netloc == ""
+    assert path == "/foo/bar"
+
+    try:
+        # Test with environment variables set
+        os.environ["DEFAULT_SCHEME"] = "scheme"
+        os.environ["DEFAULT_NETLOC"] = "netloc:8000"
+
+        scheme, netloc, path = PyArrowFileIO.parse_location("/foo/bar")
+        assert scheme == "scheme"
+        assert netloc == "netloc:8000"
+        assert path == "netloc:8000/foo/bar"
+
+        # Set environment variables
+        os.environ["DEFAULT_SCHEME"] = "hdfs"
+        os.environ["DEFAULT_NETLOC"] = "netloc:8000"
+
+        scheme, netloc, path = PyArrowFileIO.parse_location("/foo/bar")
+        assert scheme == "hdfs"
+        assert netloc == "netloc:8000"
+        assert path == "/foo/bar"
+    finally:
+        # Clean up environment variables
+        if "DEFAULT_SCHEME" in os.environ:
+            del os.environ["DEFAULT_SCHEME"]
+        if "DEFAULT_NETLOC" in os.environ:
+            del os.environ["DEFAULT_NETLOC"]