Release history

What's new in version 5.x

• You can find the latest changes here...

What's new in version 4.3.3

• Added: When connecting to a database you can choose to manual login instead of saving the username and password on disk
• Added: "Computed by" data type for Firebird databases
• Added: Missing border color property for view objects
• Fixed: Filename in the statusbar was not properly updated when model was saved under a new filename
• Fixed: View columns not synchronized properly in sub diagrams
• Fixed: Pasting objects would not clear the selected object (single instance)
• Fixed: Wrong sql scripts were generated with "compare with database". This occurred with routines defined as function or package
• Fixed: Removed some repaint/refresh problems in the docked object lists (domains, routines...) under Vista
•  Fixed: Foreign key parent mandatory flag was not saved when linked between two tables without columns
• Fixed: Syntax highlighting used the incorrect colors under certain situations
• Fixed: PostgreSQL all columns in tables on database side are always presented as "NULLABLE" using the ADO driver
• Fixed: SQL Server alter table scripts (syntax error)

What's new in version 4.3.0

• Added: Image preview window with zoom level control and page selection
  for saving a model as image
• Added: Todo List per model (CTRL-T)
• Added: Support for reserved words. Names are checked during model validation. Lists are located in the resource folder with the .rw file extension which can be edited with a normal text editor
• Added: Column comment export statement for SQL Server
• Added: New style property "Gradient end color" for tables and views
• Added: DDL preview tab in table editor
• Fixed: Style settings for an object on a sub page were applied to the main object
• Fixed: Ms Access script didn't allow double quotes <""> around identifiers. Changed to <[]>

What's new in version 4.2.1

• Added: Option to select a different default data type (Environment option)
• Added: Dialog to view and edit all model script templates
• Added: Templates are splitted in separate files and added to a configurable template folder
• Added: New PostgreSQL data types
• Change: Triggers are created after views and routines in DDL script
• Fixed: Some paint/refresh issues in the model designer
• Fixed: Added missing table name in the compare with database update script when a new column is added to a database table
• Fixed: Adding the same table multiple times in the Query Designer would set an invalid table alias
• Fixed: Help file (pdf) page reference index
• Fixed: Drop scripts for primary and foreign keys in MySQL
• Fixed: Name field of sequences and generators was restricted to 30 characters

What's new in version 4.2.0

• Added: Query designer tool to design SQL queries based on model tables
• Added: Relationships are reconnected after using the auto complete option in the "Add model objects" dialog
• Added: View settings to model options including a name template
• Added: Table image to indicate a primary key / foreign key column
• Fixed: SQL Server 2008 data type specification
• Fixed: Reverse engineering of SQL Server view and routine code over 4000 characters
• Fixed: Missing comma in generated ddl script using check and unique constraints
   

What's new in version 4.1.7

• Added: Option to display view columns. A view is now presented as a table but with rounded corners
• Added: New design object "Image" to add trademark pictures and/or company logo
• Added: Pressing the ALT key when selecting a groupbox will temporarily disable the auto select contents feature. This way it becomes easier to just resize the groupbox
• Added: Improved display control for comparing views and routines. Both panes are now included in one single control. This makes it easier to compare sql scripts
• Added: Auto complete option in "Add Model Objects" dialog. Auto complete will select all relationships for the tables within the active model page
• Added: Enhanced mousewheel support. While set to scrolling mode you can now use the "shift" key to scroll horizontal and use the "ctrl" key to zoom in or out
• Added: "Restore" button to the page properties editor in order to apply the default model page settings
• Added: Separate pdf help file for keyboard shortcuts (Quick reference)

What's new in version 4.1.0

• Added: Changed the compare with database feature and added generation of forward engineering sql scripts when comparing with an existing database
• Added: Table editor key shortcuts
  F2 - Jump between column and property editor
  ALT + U - Move column up
  ALT + D - Move column down
• Added: Additional script variables. Some databases, like SQL Server need access to the schema / table name part of the the full table name for generating comment DDL scripts. You can use one of the following variables to define the DDL statement:
  %TABLE_NAME% -> Full table name including schema name
  %TABLE_SCHEMANAME% -> Non quoted schema part of the table name definition
  %TABLE_OBJECTNAME% -> Non quoted table name part without schema prefix
• Added: Support for SQL Server clustered indexes
• Added: Support for Oracle packages (Routine list)
• Added: Parent mandatory flag can now be controlled from within the relationship
  editor. Context menu item removed
• Added: Option to auto select groupbox contents when selected (Environment option
• Added: Help file in PDF format in order to support searching and printing
• Added: Option to import columns from another table (table editor)
• Updated: Removed invalid relationships context menu item "Page control"
• Fixed: Using the "P" key in the comment editor would focus the page list. The "P" hotkey is replaced by CTRL + L
• Fixed: Single quote syntax highlight problem which occured when using multiple
  single quotes in one line (<text>''<text>)
• Fixed: Removing a sequence would sometimes result in an error situation
• Fixed: A relationship would drop the parent key after renaming a linked alternate key
• Fixed: SQL server identity script generation which didn't allow for "identity(0,1)"

What's new in version 4.0.2

• Added: You can alter the <database> ini file to let HappyFish generate the table constraints definition as part of the table script instead of using separate alter table statements. See the help documentation for more information
• Added: "Add to model" tool to import model objects into a submodel page through an object list
• Added: New page option to the designer context menu
• Added: Print option: Print page border (light shade of gray)
• Fixed: Improved reverse engineering of Oracle indexes
• Fixed: MySQL script generation to allow support for AutoIncrement fields
• Fixed: Primary key script generation without separate key definition. Added support for unnamed primary key constraints
• Fixed: The SQL term character of a connection is now properly applied
  within the Execute SQL console

What's new in version 4.0.1

• Added: Context menu for table column list
• Added: Column selection feature in index/key editor
• Added: Create Access Database option (Tool menu)
• Added: Schema prefix to the execute SQL table list
• Added: Show/Hide option for relationship caption text
• Added: Auto delete unused foreign key option (model option)
• Fixed: Reverse engineering of default values for Interbase and Firebird
• Fixed: Pressing the Execute SQL button in the table SQL dialog would keep the dialog open behind the Execute SQL window
• Fixed: Generating scripts for sequences and generators
• Fixed: Added missing default scripts for Oracle 11g database

What's new in version 4.0.0

Designer

• Added support for sub models (pages). You create new submodels by adding "page" objects to your design. Each page has it's own properties like font and background color
• Added support for creating object references. This means you can have multiple instances of the same object in different submodels
• The declaration of foreign key constraints is now more restricted in the area of column usage. You can select a parent key, primary or alternate key, which also enforces the used keys. You can still remap the foreign key columns or even let the system recreate them for you
• Added support for alternate keys for defining relationships. To accomplish this the
  unique constraints are moved to the index/key list in the table editor
• Added a separate tab for check constraints in the table editor
• Added description and comment fields to table check constraint editor
• Added: Index columns can now have a sort order even for primary keys and alternate keys
• Added cut option besides copy/paste
• Views are included in the design area and are therefore no longer supplied through a separate list
• You can have multiple search/validate/translate message windows open at the same time
• Domain/routine/sequence/generator windows are now embedded in the design window and are no longer floating around. You can access them through a tab control at the bottom of the screen
• Removed persistent groupbox property from the environment settings. It caused problems mainly during copy paste operations when switched on. Therefore we decided to remove it completely and added a select option which you can find through the context popup menu
• Removed the option to auto destroy unused foreign key columns. In most cases
  it became an unwanted operation when working with existing databases
• We've decided to drop the separate schema name field. To use schema names in your model you can add them by prefixing the objectname with the "." character
• User data settings are now stored per computer user profile
• User script templates are now stored per computer user profile
• The table 3D border paint style is replaced by a drop shadow feature (table option)
• Added support for multiple style settings using a new style manager
• Changed zoom option "full screen" to "zoom to fit"
• Changed: Foreign key constraint names are now checked for uniqness per table instead of per model
• The following table macros are available:
  
  %TABLE_NAME% : Schema and table name
  %ENTITY_NAME% : Table name without schema name and "."
  %ENTITY_SCHEMA% : Table schema name without table name and "."
   
•  Design grid and page preview colors are now automatically set depending on the used background color
• Added new search option: Whole words
• Option to minimize the application to the windows system tray area
• Option to turn sql syntax highlighting off
• Setting model option "Max. Visible Columns" to zero will allow for unlimited visible columns per table

Model properties

• Removed entity column alignment option
• Changed: Entity autosize width / height are combined in one single option "Autosize"

Execute SQL

• Statement terminator option moved to the database connection editor. Each database connection now has it's own terminator character
• Added a list with all database tables

Database related

Oracle

• Dropped support for Oracle 7
• Added support for Oracle 11g
• Added data types BINARY_FLOAT, BINARY_DOUBLE, ROWID, UROWID

Paradox

• Paradox "databases" are no longer supported in version 4

Ms Access

• Added support for Access 2007

Informix

• Added support for Informix 11

Firebird

• Added native reverse engineering script for Firebird 2

SQL Server

• Added support for calculated columns
• Added reverse engineering of column/table descriptions

PostgreSQL

• Added data type OID

What's new in version 3.3.7

• Fixed: Trigger comment editor was disabled
• Fixed: Table macro %ENTITY_NAME% did not work on trigger code
• Fixed: Domain column references not properly generated when enclosed in quotes

What's new in version 3.3.6

• Fixed: Foreign key field color wasn't properly restored when the option active highlight foreign keys was enabled
• Fixed: Changes during editing didn't always trigger the modified flag / undo buffer
• Fixed: Oracle "is not null" column constraint did not set the column NN flag during reverse engineering
• Fixed: Removed generator import message after reverse engineering for databases which don't support the use of generators

What's new in version 3.3.5

• Small, non critical update to fix a node problem in the object browser. The tree view nodes didn't always update/refresh properly using the generator and sequence nodes

What's new in version 3.3.4

• Added: Pressing <space> in a grid (data types) will cycle through all available data types
• Added: PostgreSQL 8.x support for CHAR data types without length specifier
• Added: Oracle 10g: Reverse engineering of table and table column comments
• Added: "Compare with database" will detect (not sync) dropped objects like
  columns, indexes, constraints etc
• Added: Storage parameter options for indexes
• Added: Included table storage parameters to model search option
• Fixed: Oracle 10g: Reverse engineering of NCHAR data length

What's new in version 3.3.2

• Added: Table style option: Foreign key text color
• Changed: The table prefix is also used with the foreign key name check
  when the reuse existing flag is enabled
• Fixed: Script editor: writing non-English letters such as å,ä or ö in combination with a random following letter would fail. Example: "åt" becomes "tt", ät becomes "tt" and öt becomes "tt"
• Fixed: Pressing the escape key during or after dragging objects would result in an unwanted undo operation
• Fixed: Database generation of foreign keys would add the table schema to the foreign key name, removed
• Fixed: Varchar(Max) data type for sql server 2005 would not export properly during script generation

What's new in version 3.3.1

• Added: Quote names (case sensitivity) made optional when generating a database script
• Added: A template list for table storage scripts
• Added: Model comment text added to default model options
• Changed: Model comments moved to the general tab
   

Compare with database

• Added: Additional compare options
• Added: Data type comparison for native script connections
• Added: Nullable flag comparison

SQL Server 2000

• Added: Reverse engineering of the routine type category: Function or Procedure
• Added: Reverse engineering of index sort order ASC / DESC
   

SQL Server 2005

• Added: Reverse engineering of the routine type category: Function or Procedure
• Added: Reverse engineering of index sort order ASC / DESC
• Fixed: Reverse engineering of XML types
• Fixed: Reverse engineering of Varchar(Max), NVarchar(Max), Varbinary(Max)
   

What's new in version 3.3.0

• Added: Support for MaxDB 7 with native reverse engineering script
• Added: Native reverse engineering script for Interbase 7.x
• Added: Native reverse engineering script for Informix 10.x
• Added: Comment field for indexes
• Added: Comment field for triggers
• Added: Routine create script has now separate sections for functions and procedures. Added type field for routines to control this behavior
• Added: Object names are now quoted in database scripts (when required). This includes fieldname macros in table scripts
• Added: Support for Interbase 7.x data type arrays
• Added: Cursor position indicator in Execute SQL window
• Added: 4 Memory slots in Execute SQL window to quick save / restore sql statements which are not yet executed
• Added: Clear style settings option for each object. Previous there was no easy way to go back to the default system style once a new style was saved
• Added: Table storage field to specify additional table create options
• Added: Entity display option: Entity Images. Used images are located in the subfolder "graphics". Each image is 16x16 pixels (maximum) in size and can be replaced with your own custom bitmap images
• Added: Model navigator, object browser and comment window can now be floating or docked
• Added: Additional zoom tool to quickly zoom into specific areas of your design
• Added: Generators to "Compare with database"
• Changed: Database resource script syntax in order to make it easier to read
• Changed: Database resource script per supported version
• Updated: Informix 10 data types
• Changed: Database resource templates per supported version
• Updated: Design object creation cursors
• Updated: Help documentation
• Fixed: Showing blob fields in the execute SQL window would sometimes fail with an exception
• Fixed: Syntax editor will now respond correctly to tab characters in text. It will convert tab characters to text with 4 spaces
• Fixed: Closing the table SQL tool window with ESC key would lockup the application

What's new in version 3.2.2
 

• General: internal file format upgrade (version number 5)
• General: Important update of the MySQL 5 implementation
• Added: Support for Ado data type mappings per database
• Added: Support for foreign key data type mappings per database
• Updated: Primary key divider is no longer shown when a table only contains primary key columns
• Changed: Index editor / Index sorting. Sorting is now defined per column
• Changed: Database translation table is dropped and replaced by a more flexible way of database translation using data type groups
• Fixed: Database connection settings became invalid while using long filenames or database names
• Fixed: Many to many relationship table name generation (schema prefix)
• Fixed: Ado reverse engineering for MySQL and PostgreSQL through ODBC
   

What's new in 3.2.1

• Added: Support for MySQL 5 Spatial data types
• Fixed: Pressing cancel in the table editor would remove the foreign key indicators
• Fixed: Vista support for the main toolbar. The toolbar sections did not paint properly under Vista while dragging

What's new in 3.2.0

Important change:

Relationships no longer support separate constraint names. The object name will be used as constraint name and the constraint name is no longer optional. The main reason is to make the design process of relationships more consistent with the way we define tables, indexes, views, routines, etc. The constraint name field is dropped in this version and you need, where needed, update the relationship names manually

Other changes

• Added: Native reverse engineering scripts for:
  - Oracle 10g
  - SQL Server 2000/2005
  - PostgresSQL 8.x
  - MySQL 5.0 (build 5.027)
  Other databases will follow
• Added: Clicking on a column in the Execute SQL grid will zoom in on the column data. This to view column data containing multiple lines
• Added: Execute SQL will also return result sets from functions
• Added: Right mouse button options Expand / Collapse in object browser
• Added: Table constraints and triggers to "compare with database" feature
• Added: Extended synchronization options in "Compare with database"
• Added: Environment option: Cancel edits with escape key
• Added: Triggers, table constraints, views, routines, sequences, generators to HTML Report
• Added: Schema prefix check in the table, view and routine editors for the "entity" name field
• Added: Optimized schema support in several areas
• Changed: Layout selection moved from reverse engineering window to the environment options to support the same layout settings in the compare with database part
• Changed: Execute SQL results are returned in the active window tab. Added CTRL-F9 to run the query in a new result window tab
• Changed: Increased the execute sql history buffer to 20 entries
• Changed: The primary key constraint is now displayed under the table constraints node in the object browser
• Changed: Color for selection markers in diagram window (black on white)
• Changed: Renamed "Procedures" to "Routines" to indicate that we also can have functions in a database
• Changed: Removed "Ignore existing" from reverse engineering. This is in the future the default setting
• Changed: Combined a HTML report layout in one single file with sections
• Changed: Relationships and tables can no longer have the same name (including schema name)
• Fixed: Validating a diagram with views and procedures would raise an exception
• Fixed: Comment linefeeds in HTML report

What's new 12-09-2006

• Added: Auto create foreign key option:
  Prefix column with parent table name + optional separator
• Added: Additional table macro %ENTITY_NAME%:
  In order to improve schema support there was a need for an additional table macro. The entity macro will point to the table name without schema notation. The original %TABLE_NAME% macro will show the physical table name including the schema prefix. The file format will be updated so all table element names like indexes, constraints and triggers will use the entity macro as new default
• Added: The original table name in the table editor is now splitted into two parts, an entity name and a schema name. Schema name + Entity name must be unique for each model where the schema name is optional
• Fixed: The relationship name template was created including the table schema prefix. This
  could result in an invalid constrained name

What's new 11-24-2006

General

• Added: CTRL-N Hotkey for a new model
• Added: CTRL-O Hotkey for a opening an existing model
• Added: CTRL-P Hotkey for printing
• Added: Table column alignment spacing is now calculated based on the selected font size (was previous fixed on 12 points) This may alter your current design layout in width
• Added: Printer name in print preview window
• Fixed: When a foreign key would contain invalid columns (out of sync) a copy paste operation in the model failed

Script generation

• Added: Two new sections for exporting comments to the database
  (TABLE_COMMENT, COLUMN_COMMENT). These sections are supported for
  Oracle, IBMDB2 and Informix
• Changed: Table parts are now by default separated with a comma character
  (Removed from the script definition file)
• Changed: For dbms which support domains with check constraints and default values, these are no longer defined per column on export but are defined only per domain (Applies to Firebird, Interbase, PostgresSQL)
• Fixed: SQL Server script (Percentage sign)
• Fixed: Firebird script (Double comma character)
   

What's new 11-12-2006

• Added: A domain data type in a foreign key definition can be set to the native data type on creation (optional)
• Added: Relationship roles can be selected through a dropdown list which contains already used roles in a model
• Added: Improved script generation speed for large models
• Added: The full data type is displayed in the table editor
• Added: Optional tab key support for the column grid (table editor)
• Added: Option to copy an existing column (table editor)
• Added: Missing integrity flags for SQL Server 2005
• Added: Validation now checks for duplicate object/constraint names
• Added: Additional notation style crows feet + roles
• Added: Several other small improvements
• Added: "Keys Only" attribute display value
• Changed: Removed print thick lines option (default)
• Changed: Rename objects allows an empty replace with value to enable removing parts of an object name
• Changed: The index display option now works without the column display option (optional)
• Updated: Database translation table
• Fixed: Quick lookup tool did not always expand properly
• Fixed: Error during HTML report generation
• Fixed: Table caption alignment option
• Fixed: Relationships did not connect properly when table auto size was disabled
• Fixed: Model comment was not exported properly
• Fixed: Empty primary key constraint was dropped on database generation
• Fixed: %Table_Name% macro for triggers was not translated on export
• Fixed: ADO would report unsupported fk integrity flags during reverse engineering
• Fixed: Print bug using thick lines option which caused lines to have different weights
• Fixed: Copying domains between different dbms platforms would not translate the data types
• Fixed: Domain name unique check failed
• Fixed: Print to PDF did not work properly (missing lines)
• Fixed: Added missing notation to the model options editor
   

What's new in version 3

Additions

• Configurable design width and height by dragging objects beyond the design window borders (right or bottom side)
• DBMS Schema name support as: <SCHEMA>.<OBJECTNAME>
• Improved relationship placement routines
• Table data type display style (char case)
• Syntax highlighting for SQL code editors (standard SQL set)
• Zoom factor selection tool in main toolbar
• Default model zoom factor to configuration
• Data types in object browser for domains and tables
• Triggers and table constraints to the object browser
• Quick lookup tool in the object browser window for the selected object
• Much more actions can be undone after being changed including model display settings
• Resizable design navigator
• Design navigator mouse wheel and panning support
• Font style settings for individual objects
• Printing of page numbers
• Print line width adjustment setting (thick lines)
• Setting for a default model folder
• Model panning tool
• Active foreign key table highlighting
• Configurable execute SQL terminator character
• Smart SQL Statement parser in execute SQL window and DDL viewer
• Model pre and post create scripts
• Table name is now displayed in the dialog caption
• Text search now remembers the last 15 search items through a dropdown box
• Additional macro's for foreign keys to the HTML Report
• Foreign key create and delete rule replaced by separate checkboxes in the configuration options window
• Optional display filter in the DBMS translation editor
• Parent mandatory flag for relationships is now properly set after reverse engineering. Previous the Parent Mandatory flag was default set to true
• Reverse Engineering schema support
• Compare with database schema support
• Relationship constraint name macro %RELATION_NAME%
• Improved reverse engineering and compare With database functionality
• Rename button for database connections
• Configuration window to customize foreign key data type mappings
• Custom macros for database scripting (.def files). Custom macros are inserted into the script after ddl generation
• Generation of domain DDL script is now optional.
• Rename objects feature
• Domain DDL export is now optional. If the DBMS supports domains the user can choose to export domains as native data types only
• Support for custom post script macros (Model Options)
• Support for Informix Dynamic Server
• Show differences only option in compare with database window

Changes

• %TAB_NAME% macro changed in %TABLE_NAME%
• Snap to grid now rounds down when below the midpoint and up when above the midpoint
• Modified flag per model tab instead of showing it in the general status bar
• Some relationship display style settings are removed
• Zoom tool buttons moved to main toolbar
• Foreign key columns are no longer cascade updated except for primary key changes and the optional cascade rename feature
• Refreshed the object browser node images to make it all easier to read
• Improved foreign key comparison in the Compare with database window
• Unique and check constraint names are no longer optional
• Primary Key constraint name replaced by an index definition (Primary Index) to prevent invalid creation of indexes. The conversion rule to version 3 is as follows:

• When opening a database connection the database connection description is shown in the gui instead of the ado provider name
• Removed Select Objects color from the configuration settings. The optimal color is now automatically determined by HF depending on the selected background color
• Execute SQL window: Result view column width restricted to a specific width
• Comments are no longer exported in a fixed comment format. This makes it possible to insert the comments to the DBMS with DDL statements. Comment codes are added to the comment macro result tag.
• Help documentation
   

IBM DB2

• Added support for version 9
• Added support for sequences
• Added support for distinct types

Oracle

• Added support for sequences

PostgreSQL
 

• Added support for sequences
   

Interbase
 

• Added support for generators
   

Firebird
 

• Added support for generators
   

MySQL
 

• Happy Fish now uses "PRIMARY_KEY" as the default PK Constraint name
  This alias is the same as "PRIMARY" used by MySQL. Please note that MySQL
  still doesn't support custom primary key constraint names

SQL Server
 

• Added: Identity definition is automatically disabled when used as source for a
  foreign key column
• Added: SQL_VARIANT datatype
• Added: Support for VarChar(max), NVarChar(max) and VarBinary(max)
  The "Max" property will override the length specifier
• Removed: Float datatype length specification

Happy Fish Version 2

General

• Fixed: Compare database feature failed when verifying tables defined in different schema's under SQL server 2005 (08-16-2006)
• Fixed: Primary key indicator misplaced in the Medium table display style
• Fixed: After upgrading from a previous version saving the model with "Save As" would fail
• Fixed: Copy Paste of tables with foreign key indexes failed. Indexes remained while there was no relationship defined. While in the table editor the index couldn't be removed
• Fixed: Updating a relationship failed when primary key columns of a parent table was changed. This only happened when the FK delete mode was set to No Action. This resulted in a invalid foreign key relationship
• Fixed: HTML report window position appeared centered on dual monitor systems
• Updated: Help documentation
• Changed: Increased column name edit field length in the table
  editor to improve support for longer column names
• Fixed: Reverse engineering of tables with large amount of data took
  a very long time to complete. The internal extraction routine for
  indexes has been modified to address this problem
• Fixed: Reverse engineering error were multiple foreign key definitions were added to one single relationship object between two tables
• Added: Reverse engineering now checks for duplicate column names. This in regard to reverse engineering Oracle databases. The ADO driver would report columns twice
• Changed: Table editor -> duplicate column names are now checked after pressing OK not after editing a column name. This to prevent the duplicate name popup message box
• Fixed: Support for brackets in the table editor (create scripts and column names)
• Added: Support for SQL Server 2005 data types VarChar(Max), NVarchar(Max) and Binary(Max)
• Fixed: Stack overflow error which occurred when defining a circular reference
  between three tables. To reproduce the error add three new tables and
  add relationships between them table1 -> Table2 -> Table3 -> Table1
• Fixed: SQL Server identity field set to disable when used as a foreign key
• Fixed: Domain default value not exported properly
• Fixed: Domain check constraint value not exported properly

Oracle

• Fixed Ado driver support for the Oracle provider for OLE DB
• Fixed SQL create script with the correct statement terminator character "/"

MySQL

• Auto Increment flag is no longer copied when creating a foreign key definition

SQL Server
 

• Identity seed and increment default to seed = 1, increment = 1

Additions

• Support for SQL server 2005
• Support for Oracle 10g databases
• Support for Firebird
• Inheritance of primary keys through existing relationships. Changes in the parents primary key now get inherited by the child tables except for column data types
• Changes in a parent table are now cascaded through all connected child tables
• Relationships can now generate foreign key indexes (optional)
• After reverse engineering the tables are no longer stacked on their default position. They are now properly positioned within the model boundaries (Horizontal or Vertical)
• Hint & warning messages are now each configurable through the environment settings
• Model messages can be saved to a text file
• Auto size table width/height options
• Max. visible table columns option to restrict table height for large tables
• Table column info can be displayed aligned (grid column style)
• Model properties title/author/copyright/version are now saved to the default model options
• Missing Oracle 9i data types Timestamp / Interval
• Added an primary key indicator

Additions

• User templates to the view and stored procedure editors
• Views and Procedures now have two additional options
  - Export script
  - Auto generate drop script
• Search option to the view/procedure list
• Included some missing ADO data types to reverse engineering
• The ADO data type mapping is now using the DBMS translation table to match the best available DBMS data type. Before each DBMS definition had it's own data type mapping list which made if more difficult to configure and maintain
• HTML Report is now also accessible through a shortcut button
• Undo/Copy/Paste support for domains
• Comment field for domains
• Support for [SHIFT] key scrolling option. Keep the shift key pressed while moving the mouse to the outer sides of the design window will scroll the model
• Layout property for notes
• Alignment property for table names
• Free resize for all design objects
• Resize object tool
• Visual design grid (optional)
• Snap to grid feature while placing/moving objects
• Groupboxes now also move their contents (optional)
• Creating objects trough hotkeys
  "T" Table
  "G" Groupbox
  "R" Relation
  "M" Many-to-Many relation
  "N" Notes
• The note object can now process macro's to insert model attributes into the text.
  The following macro's are available:
  %TITLE%
  %AUTHOR%
  %COPYRIGHT%
  %DBMS%
  %DATECREATED%
  %DATEMODIFIED%
  %VERSION%
• Export settings (database generation) are now saved for later use
• Print scale is now also visible in the diagram window
• Search text option to all text and code editors
• DDL generation for selected tables only
• Visual design object "Stamp" to display model attributes
• Text search for data type / domain attribute values
• Data type conversion tool
• Foreign key data type mapping. In some cases a native data type needs to be changed to a alternative data type when used within a foreign key definition. Examples are AutoIncrement en serial data types. You can add additional mappings by editing the file "fkmapping.ini" with an external text editor. However changing the entries is discouraged
• Model navigator. A birds overview of the current visible model used for easy navigation
• Reintroduced gradient table captions
• Non-identifying relationships are now painted with a dashed line style. You can uncheck this option in the model options editor to keep the old style
• Support for MySQL 5 including triggers, views and procedures
• Support for Paradox Tables
• Support for IBM DB2 Universal Database 8
• Unknown/invalid column data types in the table editor are displayed in different color
• Tool to compare a model with an existing database schema
• Tool to convert table column data types (Data type conversion)
• Table filter to the reverse engineering window
• Table filter to the generate database window
• Execute SQL window now supports multiple datasets to be opened
• Execute SQL window now supports a visible script history

Changes

• Moved Auto update NN flag for foreign keys to model options (Saved per model). When other users would load the model with a different setting the original foreign key state would change without notification
• ADO data type mapping, used for reverse engineering, is now accessible through the tools menu in the main window
• Views en procedures are now combined in one single window for easier access. Menu items combined to "Script objects"
• Main window toolbar is now ordered in a more logical way following the menu layout
• Database create/drop script, tables are no longer exported sorted by name.
  They are now generated according to the defined parent/child relationship order
• Database DDL create and drop scripts are now combined in one single file. This is a
  great benefit and time saver while creating and debugging existing databases
• Domain export to database DDL restricted to selected tables
• The model is no longer checked for errors before generating a database. Use the separate validate option instead. Before you had to correct all errors first before you could export a database schema. This introduced some problems while just debugging a new design
• Views & Procedures: replaced header/footer definitions with user templates
• Support for a non-modal domain list. Domains can be edited directly in the
  main window while working on the model
• Removed script preview controls from the view and procedure lists
• Alignment options are now accessible through a tool window
• Align to grid option now only applies to selected objects
• Groupbox and note editors: Enabled form resize
• Designer zoom factor is now modified in smaller steps
• Canceling a new model within the "save as" file dialog will keep the current window active. Before, the window was closed without further notification
• Domain properties are now editable within the table editor
• Standard HTML report layout
• Removing foreign key columns after deleting a relationship now also updates all other table settings like indexes, table constraints and scripts. Also table validation has been improved to check for missing column references
• Removed the lock objects feature. This option is replaced by locking the model by default. Only
  when the space bar is pressed objects are allowed to be moved. Optional through the environment settings window.
• Changed: Auto deleting of foreign keys is more restricted. If a foreign key column is used in any other part of a table definition (scripts, triggers, etc.) the foreign key is NOT deleted. This to prevent that a table definition becomes invalid when a relationship is deleted.
• Foreign Key "Create/Delete rule" moved from environment settings to model options
• Double clicking in the object browser no longer closes it's expanded nodes.
• Removed: Auto update NN flag for foreign keys option. The not null flag for foreign keys is now
  always managed by the relationships parent mandatory flag
• DBMS Translation table updated

Bug Fixes

• Tables where not always sized properly when a bold caption was used
• Relationship labels where not properly positioned when a large font was used
• Undo "edit" operation when an object name was changed resulted in an access violation
• HTML Report did not properly export the column description field
• Removing a column from a parent table used by a relationship would not update the relationship columns
• A relationship where the parent and child table are the same would use the primary column also as a foreign key column, referencing itself. This happened when the create rule was set to reuse existing
• Adding a new object to a translated model would not register the object in the object browser
• "Select All" command would not clear the active selection properly
• Default window position was desktop center. This caused wrong positioning on systems with dual monitors and extended desktops
• Oracle SQL script column output generation (Default/NN syntax)
• Stamp context menu "position" should read "position"
• Help was displayed twice when invoked from the main window
• Object browser would disappear and the splitter would get misplaced. Object browser is now checked for a minimum width
• Interbase column definition must export the default value before the NOT NULL flag. Script template updated
• Close All Windows would not add the model files to the recent file list
• Deleting a primary key column would not update a primary key index definition properly
• Print preview would raise an error message when there was no printer available
• SQL Server data types TEXT and NTEXT were created default with a length parameter. Removed.

Ms Access

• Missing numeric data type

SQL Server

• Added: identity export attribute tag definition. When used you must set the enable property to true first. The following flags are now properly supported:
  IDENTITY
  IDENTITY(n)
  IDENTITY(n,m)

Oracle

• Added: "CASCADE CONSTRAINTS" keywords to the drop table DDL command

MySQL 4

• Added: Character set and Collate data type options
• Added: Missing UNSIGNED flag for float data types
• Removed: Unsupported SET DEFAULT referential integrity flag
   

Version 09-20-2005

Additions

• Not null flag for foreign keys are automatically updated when the parent mandatory flag changes (relation editor). To enable this feature check the auto-update not null flag in the environment options screen
• New model option: Not Null indicator (-NN), default disabled
• New notation style "Referential Integrity"
  NO ACTION (-), RESTRICT (R), CASCADE (C), SET NULL (N), SET DEFAULT (D)
• Group boxes can now resize from all sides
• Attribute data types are now also saved to file with XML tag "full type" to enable reporting through a html page (no data type formatting required). This also applies to domain data types
• Full Page tool button to design window
• "Lock Objects" option which is used to disable model drawing (object moving) is now saved to the output file
• HTML Report generator
• Bold text font style for table captions
• Comments for views and procedures
• Tables can now reference themselves with a foreign key relationship
• Table script splitted in "Before Create" script and "After Create" script

Changes

• Table script splitted in "Before Create" script and "After Create" script
• View and procedure lists are no longer sorted by name to preserve/support code dependencies. instead a manual sort button is added. The way you define Views en Stored Procedures may change in upcoming releases
• In the object browser -> Model -> Tables -> <table name> -> Indices: The index name reads as IDX_%TAB_NAME% instead of the expanded table name, requested
• Object browser is now defined globally instead of within each design window
• Object browser settings are now automatically stored
• Overall performance while working with designs containing large number of objects

Bug Fixes

• Editing domains, views and procedures through the object browser would not set the modified flag
• Loading an empty model at startup (desktop) would raise an error message
• Print button not working in the Generate database > SQL output window
• Deleting a domain did not update the associated table columns
• Model modified date was not updated when saved to disk
• Changes made through the search text or validate dialog did not appear in the object browser
• Several problems in the reverse engineering module, importing from ODBC data sources now also returns foreign key information. The kind of meta-data extracted from the database still depends on the features an ole db provider supports. I could use some practical field testing reports concerning the current reverse engineering implementation. Please let me know about your experiences using it
• Setting a column as primary key would not turn on the Not Null flag when generating scripts
• Comments where not properly exported to database script
• Table scripts using column names between brackets, brackets where not removed when generating the database schema
• HTML reporting using filenames + paths longer then 255 characters would fail
• The options Show Indexes (in model options) was not saved in the project, on reopen the default choice was used
• When creating a new relationship, the name of the PK field of the master table was transferred in the child table with a "1" appended
• Close menu item should read "Close All"
• Index column names indent value not properly set within the table display routine. Names are now properly positioned.

PostgreSQL related

• Domain SQL support (PostgreSQL 8)
• Added "CASCADE" option to drop table syntax
• Removed drop primary key constraint definition
• Functions are now created before table triggers
• Drop stored procedure syntax (should read "DROP FUNCTION")