castor-extractor 0.24.52__tar.gz → 0.24.54__tar.gz

{castor_extractor-0.24.52 → castor_extractor-0.24.54}/CHANGELOG.md

@@ -1,5 +1,13 @@
 # Changelog
 
+## 0.24.54 - 2025-09-18
+
+* SqlServer: fix typo in the extraction query of schemas
+
+## 0.24.53 - 2025-09-18
+
+* Fix CSV field size to support running on Windows
+
 ## 0.24.52 - 2025-09-18
 
 * SqlServer : improve extraction of users and technical owners

{castor_extractor-0.24.52 → castor_extractor-0.24.54}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: castor-extractor
-Version: 0.24.52
+Version: 0.24.54
 Summary: Extract your metadata assets.
 Home-page: https://www.castordoc.com/
 License: EULA
@@ -215,6 +215,14 @@ For any questions or bug report, contact us at [support@coalesce.io](mailto:supp
 
 # Changelog
 
+## 0.24.54 - 2025-09-18
+
+* SqlServer: fix typo in the extraction query of schemas
+
+## 0.24.53 - 2025-09-18
+
+* Fix CSV field size to support running on Windows
+
 ## 0.24.52 - 2025-09-18
 
 * SqlServer : improve extraction of users and technical owners

{castor_extractor-0.24.52 → castor_extractor-0.24.54}/castor_extractor/utils/formatter.py

@@ -1,6 +1,7 @@
 """convert files to csv"""
 
 import csv
+import ctypes
 import json
 import logging
 import re
@@ -21,6 +22,8 @@ CSV_OPTIONS: CsvOptions = {
     "quotechar": '"',
 }
 
+CSV_FIELD_SIZE_MB = 100
+
 ScalarValue = Union[int, float, None, str]
 
 logger = logging.getLogger(__name__)
@@ -132,6 +135,31 @@ class Formatter(ABC):
         pass
 
 
+def _set_csv_field_size_limit(target_limit_mb: int) -> None:
+    """
+    Safely set the maximum CSV field size limit across platforms.
+
+    This function wraps `csv.field_size_limit()` to avoid `OverflowError` on
+    Windows, where the maximum C long is only 32 bits (2^31 - 1). On Linux and
+    macOS, the C long is typically 64 bits, allowing much larger values.
+
+    The requested limit is specified in megabytes and converted to bytes.
+
+    It is then clamped to the maximum value supported by:
+      - the current platform's C long
+      - Python's `sys.maxsize`
+      - the requested target limit
+    """
+    target_limit_bytes = target_limit_mb * 1024**2
+
+    # max value of C long for the current platform
+    platform_c_long = (1 << (8 * ctypes.sizeof(ctypes.c_long) - 1)) - 1
+
+    limit_bytes = min(target_limit_bytes, sys.maxsize, platform_c_long)
+
+    csv.field_size_limit(limit_bytes)
+
+
 class CsvFormatter(Formatter):
     """
     Serialize/Deserialize CSV
@@ -141,9 +169,7 @@ class CsvFormatter(Formatter):
         return "csv"
 
     # increase the size limit (some fields are very large)
-    # Limit to 100 MB. Value must be smaller than the C long maximum value.
-    size_limit = min(sys.maxsize, 100 * 1024**3)
-    csv.field_size_limit(size_limit)
+    _set_csv_field_size_limit(target_limit_mb=CSV_FIELD_SIZE_MB)
 
     @staticmethod
     def serialize(buffer: IO[str], data: Iterable[dict]) -> bool:

{castor_extractor-0.24.52 → castor_extractor-0.24.54}/castor_extractor/warehouse/sqlserver/queries/schema.sql

@@ -17,7 +17,7 @@ SELECT
     schema_name = s.name,
     schema_id = CAST(d.database_id AS VARCHAR(10)) + '_' + CAST(s.schema_id AS VARCHAR(10)),
     schema_owner = u.name,
-    schema_owner_id = u.uid
+    schema_owner_id = u.name
 FROM [{database}].sys.schemas AS s
 INNER JOIN ids AS i
     ON s.name = i.table_schema

{castor_extractor-0.24.52 → castor_extractor-0.24.54}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools>=61.2"]
 
 [tool.poetry]
 name = "castor-extractor"
-version = "0.24.52"
+version = "0.24.54"
 description = "Extract your metadata assets."
 authors = ["Castor <support@castordoc.com>"]
 license = "EULA"