ApexSQL Edit

This is a screenshot of the new mapping mgmt screen which will allow mapping operations to be performed using drag and drop of files and objects onto each other.

This example shows a database on the left and some Source Controlled items on the right. You can see how the form shows the linkings between the various database objects and their corresponding Source Controlled items (in this case a Source Gear Vault system being accessed through SCC).

Well, the SCC integration is almost complete and ready for testing by some users who have volunteered to test the SCC against a variety of systems.

My initial tests seem to show that it should work fairly well although with SCC being a somewhat limiting API, all of the features currently in Edit for SrcSafe will not be available with SCC. For example, the user will not be able to navigate the entire SC explorer as they can with SrcSafe but will be able to navigate at a project level. We are working to make it easy to switch between SCC projects and even have "saved" SCC projects so that the user can switch between them easily.

Some other differences will be in things like the history of files. Edit shows a custom history form where SCC shows the history form from the SCC system (if available and supported by a vendor's SCC implementation).

Once SCC is complete and out to testers we will begin on the object mapping mgmt screen and hope to have it complete sometime late next week. At that point we will begin wrapping up some items, adding any last minutes fixes and hopefully prepare for the release which should be the last week of March.

This has been the longest amount of time between releases, mainly due to the complexity of adding in the SCC integration. Future release should go back to 4-5 week range instead of the almost 7 weeks that this release has taken. In the future we may concentrate on getting out releases quicker with smaller number of items so possible 1-2 major enhancements and other smaller issues per release so that we can get changes out to the community quicker since this release has many items that would improve the current use by all users but is in a state where it cannot be released due to SCC and object mapping in a state of flux.

Well, SCC integration is coming along. I've been able to check out, check in, undo check out, and add items in the schema mapping. I still have a ways to go like mapping a database, handling the source control explorer, etc. Hope to have all of those done over the weekend so that I can start getting some SCC users to test the basic functionality early next week. The process is going a little slower than I had hoped but it is coming along and working very nicely (at least with SourceGear Vault which is the SCC system I'm testing against).

I also finished the Task form and item mods so that when you open the task form now it will immediately search all open windows for tasks and then will thread off looking at procs, functions, and trigger texts in open databases. These will appear in the task window as they are found. Double clicking on a task that is in a database script will now open the script in a new window if the object is not mapped or open the mapped object if it is. I am thinking about looking at mapped object scripts as well for tasks.

I also added the functionality although still need the UI to do custom keywords in the editor. I have added 5 server/database types

UserTableToken
UserViewToken
UserProcedureToken
UserFunctionToken
UserDataTypeToken

and one universal type:

UserCustomToken

That can be modified so that items in these parsed cases will format as defined by the user, so the user can modify formatting color for these items in the editor.

By default, when connecting to a database, the editor will load the basic types for the five main types, so that by default the editor will use tables/views/etc. from the current connection and highlight them. The user can customize these per server/database using a list of specified tokens or a sql script that returns a resultset.

The UserCustomToken can be customized by entering a list of tokens.

This feature will make it so that tables/procedures/etc in the current script are highlighted differently than standard tokens which should make scripts easier to read and items easier to find.

Well, I was finally able to hook a schema object to an SCC based Source Controlled item (in Source Gear Vault). I was able to successfully connect to the SCC system from Edit and also get latest and check out the file. I was not however able to check it back in since something is not wired correctly on that piece, but hey, who needs to check things in :)

Hopefully by early next week we should have the SCC functionality integrated and can begin full testing and possible involve others in the user community who use various SCC enabled systems since we are only testing using Source Gear Vault. We are currently working on opening pre-mapped objects so currently we have to add the mapping entry manually. For those that will be using the SCC object mapping, it will behave slightly different when specifying linkages between the database and SCC items since SCC does not allow the full tree navigation of all projects that Source Safe API does. This means that different "folders" or "projects" will maintain separate entries and may require a login when switching between projects. We are trying to minimize the logins required to the SCC system but some systems will require the SCC user to login more than others. This is not something related to Edit but related to SCC and the specific Source Control system.

We are looking forward to getting SCC out there for users to begin using and also the new object mapping interface to make the process of Database and Source Control integration easier and more powerful. Please continue to check the blog for additional information that we will be posting about the SCC integration.

For v2.16, I've modified the way Tasks work to make it more useful and provide more power in this feature. In addition to only having TODO tasks, you can now how TODO, BUG, and REMARK tasks. Additionally, you can create your own Task types and use them in the editor to markup your own specific tasks.

In addition to different types of tasks, you can now assign a user and status to a task and have that information displayed in the task view screen. So, the following task would identify itself for the user "Ashton" and having a status of "Open":

-- BUG{User:Ashton;Status:Open}: This item needs to be fixed.

Too much to type you say, well, you can now double click on the task item and a task edit form will display with the current task information and allow the user to edit it and have the task information change. Status and Task types can be selected from predefined lists. A status list is provided and can be modified to include other statuses.

All of this information becomes useful when you view the tasks using the Task viewer. On this form you can view all of the tasks for open files and then group by task type, user, status, etc. In addition, after the form opens, it will also look at any editor windows that have connections and look for task items in the text of system objects for these connections. These items will also appear in the task list and double clicking on these items will open the script from the database for the user to view/edit.

The improvements to tasks should make them more useful in managing tasks, bugs, etc in SQL scripts.

Well, I was able to implement custom tooltips when you hover over items in the editor. This should provide benefit to let users see info about a table/view/function/proc while in the editor. One of the nicer things I've found in testing is that, especially with table values functions, it will show not only the parameters, but the columns in the table that is returned.

Below is the tip from the Customers table, notice how it also incorporates MS_Description properties, if defined, for the table and columns:



Also, here is a function with a parameter and some columns:



I will probably add @parms for the v2.16 release. I may try to columns at a later date because to do columns properly, we'll have to do more parsing to try and determine ownership of the column.

If you have comments about what you see in this post, please post in the forums under BLOG:TOOLTIP so that others will know what the post is in reference to.

Also, in the shots, you might notice the new document tab list at top. This piece of the app was changed since the new control fixes the flicker when switching documents and provides the ability to re-arrange tabs as well as smart ctrl+tab tab switching.

Thanks