bumps to version 1.15.0

dlt/destinations/impl/filesystem/sql_client.py

@@ -1,5 +1,5 @@
 from typing import Any, TYPE_CHECKING, Tuple, List
-import os
+import semver
 import duckdb
 
 from dlt.common import logger
@@ -159,9 +159,12 @@ def _escape_column_name(col_name: str) -> str:
             else:
                 compression = ""
 
-            from_statement = (
-                f"iceberg_scan('{last_metadata_file}'{compression}, skip_schema_inference=false)"
-            )
+            if semver.Version.parse(duckdb.__version__) > semver.Version.parse("1.3.0"):
+                scanner_options = "union_by_name=true"
+            else:
+                scanner_options = "skip_schema_inference=false"
+
+            from_statement = f"iceberg_scan('{last_metadata_file}'{compression}, {scanner_options})"
             # TODO: on duckdb > 1.2.1 register self.remote_client.fs_client as abfss fsspec filesystem
             #   this will enable iceberg but with lower performance
         else:

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "dlt"
-version = "1.14.1"
+version = "1.15.0"
 description = "dlt is an open-source python-first scalable data loading library that does not require any backend to run."
 authors = [{ name = "dltHub Inc.", email = "[email protected]" }]
 requires-python = ">=3.9.2, <3.15"
@@ -262,7 +262,8 @@ pipeline = [
     # this pandas and above has functional numpy dependency
     "pandas>=2.1.4",
     # keep last duckdb version that works with azure and iceberg correctly
-    "duckdb<1.2.1",
+    "duckdb<1.4",
+    "duckdb<1.2.1 ; os_name == 'nt'",
 ]
 airflow = ["apache-airflow>=2.8.0,<3 ; python_version < '3.12'"]
 ibis = [