REPORTED BUGS:

TRACKING AT: http://lindseylewis.spaces.live.com/default.aspx, Use the guestbook,
Or E-mail: support@lewissoftware.com


	reported: awittig: 08/02/2007,
	status: fixed. 10/15/2007.
	issue: The process indicator goes away after one of the databases 
		is done processing, even though the second is still in process.

	reported: llewis: 10/15/2007,
	status: fixed. 10/15/2007.
	issue: Error caused by using the ] and ) symbols in the find dialog.

	reported: llewis: 09/05/2007,
	status: fixed. 10/15/2007.
	issue: MAX column size for varchar which is reported as -1 in XML,
		XSLT now translates -1 into the MAX length for any char type of columns.

	Status: Fixed. 02/12/2009
	Issue: Compare of two XML files or XML vs SQL or SQL vs XML causes error with
		FileNames.  This was a problem in both commandline and GUI apps.

	Status: Fixed. 02/12/2009
	Issue: SQL Schema Generate from 2005 DB without functions fail, with Temp Table
		Error.

	Status: Fixed. 02/13/2009
	Issue: File Save As does not properly save the file of the current active document 
		after changing tabs.

Fixed prior to 09/05/2007.

	reported: llewis: 08/18/2007,
	status: fixed.
	issue: adding a server not already in the server tree, added it but
	    did not attempt to establish a connection or make the server node
	    accessable until exiting and re-entering the app.
	
	reported: awittig: 08/06/2007,
	status: fixed.
	issue: Not all data tables appear in table tree list in the 
		SQL Server Explorer.

	reported: awittig: 08/06/2007,
	status: fixed.
	issue: Selected data tables still selected but not shown after removing table tree 
		list in the SQL Server Explorer.  Fixed by deselecting tables on hide of 
		Table list. - this issue is two fold.  There is still an issue with the
		collapse of the table tree list losing the selected icons.

	reported: awittig: 08/06/2007,
	status: fixed.
	issue: after app processed a Compare and finished; Error: Unable to cast 
		object of type 'Lewis.SST.Controls.TableTreeNode' to 
		type 'Lewis.SST.Controls.DBTreeNode'.
		
	reported: awittig: 08/06/2007,
	status: fixed.
	issue: During close of app. System.NullReferenceException: Object reference 
		not set to an instance of an object at 
		Lewis.SST.Controls.Document.GetPersistString()
		
	reported: awittig: 08/06/2007,
	status: Needs to be re-tested since additional coding error checks have been added.
	issue: During opening of app. System.NullReferenceException for loading Settings or options.
		
	reported: llewis: 08/06/2007,
	status: fixed.
	issue: Run SQL on Server, dialog displayed user name and password as disabled fields,
		using current selected Server connection settings for, so lets not display any 
		user or pwd or security settings.
	
	reported: llewis: 08/06/2007,
	status: fixed.
	issue: Stack overflow error caused by recursive code at clearselected method in 
		SQLServerExplorer object when comparing two data tables.
	
	reported: llewis: 08/06/2007,
	status: fixed.
	issue: when "saving as", the document would report an error after saving the 
		file, about an illegal character (*) in the filename.
	
	reported: awittig: 08/02/2007, 
	status: fixed.
	issue: when clicking the report button and then cancel open file dialog 
		leaves process indicator spinning.

	reported: llewis: 08/02/2007, 
	status: fixed.
	issue: options dialog size is sized just to top window title bar on 
		fresh install of application.


NEW ENHANCEMENTS/CHANGES:

	reported: llewis: 08/17/2007,
	feature: enhanced html reports for Schema Diff, include color coding.
		
	reported: llewis: 08/07/2007,
	feature: ability to generate SST batch file commands for processing schema
		from the command line.
		
	reported: awittig: 08/07/2007,
	feature: ability to generate XML Data from data tables
		from the command line.  This also allows XML data queries from the 
		XML Node Explorer, so the XPath string /*/Table/[Field1='somevalue'] 
		will find the first node match in the XML data where there is a 
		Field1 with a inner text value of 'somevalue'.
		
	reported: awittig: 08/07/2007,
	feature: App will not populate server list, unless you refresh the servers 
		through the 3 possible ways: button/menu item.  This allows a user 
		to add only the SQL servers that he wants to see in the SQL explorer.
		


08/02/2007

Environment Notes:

	The application will a call to the local area network to discover any SQL servers when you select 
	either refresh servers button, using the System.Data.Sql.SqlDataSourceEnumerator class.
	
	This will make your firewall on your local PC ask you to allow this unless your firewall has already 
	established that port as open.



08/19/2007

Know Issues:
	Data export as XML:
	To optimize the resulting XML code, the WriteXml method drops column fields with null values. 
	Dropping the null column fields doesn't affect the usability of the DataSet object—you can 
	successfully rebuild the object from XML, and data-bound controls can easily manage null values. 
	This feature can become a problem, however, if you send the DataSet object's XML output to a 
	non-.NET platform. Other parsers, unaware that null values are omitted for brevity, might 
	fail to parse the document. If you want to represent null values in the XML output, replace 
	the null values (System.DBNull type) with other neutral values (for example, blank spaces).

    Find/Replace does not reset the start position when moving the caret via mouse or keyboard.

    Compare does not support views with indexes.

    Very large XML files can cause application to reload slowly when the settings/general/reload 
    files option is turned on.

    A large number of Views can cause initial XML serialization to take up to 10 minutes to 
    calculate all dependencies.
    
    There is no current data synchronization between tables.  By design, you cannot compare data 
    and expect the output to be the same if the tables being compared have differing schema's.

	Occasionally during heavy threaded activity, the application will get an non-recoverable MDA error
	that is thrown by the CLR as an internal CLR error, when using the mouse to select or change tabs in
	any open document window.
	
	The HTML report currently displays only the table columns that have altered or been added from the 
	XML Difference snapshot.  (use the SQL output to see all the proposed changes to the database)
	
	Installer (Current) does not copy all DLLs into the commandline versions of the tools. 
	Workaround: manually copy DLLs from GUI directory into commandline tool directories.  
	
	Installer (Current) does not make a shortcut for running the application on any Start menu, 
	or on the desktop.  
	Workaround:  locate application in the installed directory (typically Program Files\Sql Schema Tool)
	and run the application bu dbl-clicking the exe file, or make a shortcut using Windows File Explorer
	functionality.


Weaknesses:

	Reporting of Differences/Changes due to occur in the SQL script process.  The HTML diff report
	does not automatically display after the run of a Schema compare.  The HTML report button is
	the main button, but that may confuse people because of the dropdown menu for the standard report.
	
	Data Compare, can only process two tables, a source and a target, rather than the entire DB.
	
	Threading on single processor CPUs, still pegs the CPU usage, and makes things run pretty dang slow
	during a schema compare on Large Databases.  I'll be looking for ways to enhance this performance. A schema
	compare with a large number of views or large numbers of stored procedures, and with the text compare 
	option turned on is a really slow process and could take 15-20 minutes on a laptop (my laptop, debug mode).
	
	No Assembly is signed with a strong key yet.
	
	Installer is just a plain wrapper at this point and needs to be redone using WIX.
	
	Commandline apps need more testing, but they seem to use less CPU cycles than that of the GUI interface,
	even though the same methods are being called asynchronously both places.  Time of processing seems to be
	about the same, regardless.  Last large DB process time was around Lapsed time = 00:10:31.4375000.
	
	No help file is installed currently.
	
	
SQL Transformation Output from XML:

    SQL Transformation output replaces UDDTs with standard SQL types where possible, 
    to reduce issues for collation changes later on.


Data Compare:

    Currently only does one selected table from a source and one selected table from a destination.

    Schema of both tables need to be the same for a data compare to work properly. 
    Indexes can affect data order which the process is set to ignore, but could potentially 
	affect the outcome.
	
	There has not been extensive testing of the data compare yet - it has only seen limited use.


Logging:
	
	The application uses NLog to log errors, debug, warnings, and informational output, 
	from the command line tool as well as the GUI tool.  This functionality can be changed
	via the NLog.config file.
	

Settings Persistence:

	If for some reason the application fails to launch because of, or you are not satisfied with perviously
	persisted settings, delete the windows.config, dockPanel.config, and the options.config files and 
	the program will use the initial default settings for startup.


Commandline Tools:

	There are two command line applications, one that does DTS package de/serialization and one
	that does SQL schema de/serialization.


Design Considerations:

	This tool was started as a method for updating live databases with schema changes from the 
	ongoing development	efforts of multiple developers.  As such it started life as a commandline tool
	rather than using a GUI interface.
	
	The DTS Package Serializer tool was started as a method for delivery of DTS packages through some
	automated and scripted MSI or Installshield type of installer where server logins, and names could 
	be easily replaced in an XML file representation of the DTS package as opposed to using the DTS 
	Package designer that is built into SQL Server to make changes after some intall.
	
	The primary considerations of design, are to generate the SQL schema script so as to prevent any
	data loss on the target/destination SQL server, when the script is run. As such the scripting process
	was designed to not drop any existing tables, or views (the views maybe should be changed) to avoid 
	loosing any	data.  New tables will get the equivalent SQL for if exists then drop and create 
	with-in the SQL script.
	
	Secondly, the generated script should be able to run in an automated build. This requires the 
	script to be as error free as possible so as to avoid human intervention during its processing.
	
	The next design consideration, was to replace UDDT's with standard SQL datatypes during the output
	of any SQL schema script, so as to make database compares easier, and to help avoid potential 
	collation errors that can be caused by UDDTs during collation changes of a database.
	
	Another design consideration, was the all SQL scripts should be compatible with any SQL Server 2000
	standard tools, such as SQL Query Analyzer.  As such, the scripts are generated with full SQL Server 
	2000/2005 compatibility in mind.
	
	Data Compare was added after realizing that a developer will often need to compare two data tables
	records, to see that the necessary data records are in place.
	
	The application was not designed to be an editor of SQL or XML.  As such, there may be editing 
	weaknesses, but I'd prefer to concentrate on enhancing the application's strengths, unless an 
	editor is just to unusable.
	
	Future enhancements will include most or all of the items listed in the TODO List.txt file.
	

General Comments:
	
	The application is not designed to be an open souce tool, and as such with any errors that occur -  
	it would be nice to have a copy of the log file.  Just email the log (which appends), to me every 
	so often.  
	
	If the tool is too unfriendly or to hard to use no one will want it.  Please let me know about 
	UI design issues that are not neccessarily hard bugs, but maybe a usability issue.
	
	A few times I took the path of least resistance, just to get the application finished to this point.  
	I do plan on going back over the code, performing unit tests, and looking for ways to crank 
	up the performance.
	
	Any comments are appreciated.


My email address is lindsey.lewis@msn.com