Be notified of new releases
Create your free GitHub account today to subscribe to this repository for new releases and build software alongside 28 million developers.Sign up
This release adds preliminary support for BIML-generated SSIS packages that load attributes in parallel. These drastically improve performance compared to inserting/merging into the latest views, which uses trigger logic to serially load attributes. In the current release only anchors and their attributes can be loaded through BIML-generated SSIS-packages. There is unfortunately very little documentation of the BIML generation. Loading of ties and better documentation will follow in a later release. In the meantime, feel free to ask us questions about BIML in our forums.
This release contains a few bug fixes for the Splitter method. It also fixes an issue where a specified on_fail_action in a workflow was overwritten in the resulting agent job.
Primarily fixes a number of bugs, but also adds new features. When jobs are recreated by the installation process, any schedules on the job will now remain. A simple application that displays a loading report has been added. A function that converts UTC to local time has been added to the .NET CLR assembly. A new and improved splitter (with respect to performance) has also been added. This release is fully backwards compatible and can be dropped in as a replacement for version 1.08.
Fixes a few annoyances, such as directories being changed when calling the bat-files. Added some preliminary support for SQL Server 2005. There is a video tutorial available on how to use the framework here: https://www.youtube.com/playlist?list=PLG6-3kKEOyYlWEaEFzhcARtjqHU6zn1cH
This release fixes a couple of bugs that prevented the Example from running properly on all versions of Windows. It also adds support for quoted terminators, which can be specified through
delimiter="","" for example.
This release separates the framework from the projects on which it operates, making it much easier to upgrade existing framework installations. Key and type checking can now be optionally controlled from the source XML description. BULK splitting has been improved and support for FIRSTROW has been added (in order to skip headers). A number of bug fixes have been included. Among them the important one which fixes the problem of upper case data types not being checked at all.
This release fixes the example files (vehicle collisions in Manhattan). GitHub were treating them as source code files and thereby felt at liberty to modify them, unfortunately making them unreadable. They are now marked as binary, which means GitHub is no longer able to touch them when they are checked in or out.
The TRY_CAST has been replaced by a CLR function that correctly detects truncation of types. It also removes the dependency on being on SQL Server 2012 or later. The framework should now work just as well in 2005/2008.
Some other features have also been added, such as pre- and post processing in the target stored procedures, specified rowlength instead of varchar(max) in the Raw table, and a rewritten Splitter in C# that has slightly better performance than the old one.
This release fixes two bugs. Spaces can now be used in the filename directory path and BULK INSERT will no longer time out after 30 seconds when loading large files.