 |
SoftTree Technologies
Technical Support Forums
|
|
Author |
Message |
SysOp
Site Admin
Joined: 26 Nov 2006 Posts: 7948
|
|
SQL Assistant version 6.0 public BETA available |
|
SQL Assistant build 5.0.96 is now available for download. Please give it a try, we are sure you won't be disappointed.
To participate in the beta program, visit http://www.softtreetech.com/sqlassist/isql_beta.htm
IMPORTANT: If you find issues with the beta version, please post description of each issue as a separate topic in this forum, or email it to support. Screenshots and other documents visually demonstrating the issue are much appreciated, they help us to fix issues faster and ensure they are corrected before the retail version is released.
Version 6.0 delivers exciting new functionality with several new major features and over 100 minor enhancements, including, but not limited to:
Here is a list of new features and enhancements in version 6.0
===========================================
1. Real time as-you-type automatic code performance analysis and suggestions.
2. Full-featured source control interface with 3-way database-files-repository code comparison and synchronization, checkin/checkout, change history, and so on…
3. Graphical code execution plans for all types of supported database servers.
4. Customizable database version specific queries and settings. Database version is recognized automatically and version specific settings are used as required.
5. On the fly reverse engineering of virtually all types of database objects, "Show DDL" is now available for all objects.
6. Customizable SQL Assistant menus, add/delete/reorder menu items to match your coding habits and increase your productivity.
7. All databases: Table statistical hints and information panes, including instant row count hints for tables of any size, table columns with unfolding user-defined data types, and more.
8. SQL Server and Sybase ASE/ASA databases: Intellisense is now offered for UPDATE...FROM and DELETE FROM ... FROM statements.
9. All databases: intellisense for Common Table Expressions (a.k.a CTE in SQL Server, and WITH clause in Oracle)
10. All databases: intellisense is now available for APPLY keywords
11. All databases: intellisense for UPDATE statements provides enhanced support for built-in database functions.
12. Test-data generator is now context sensitive. Right-click on a table name, choose Test-data Generator menu, and with one additional click you can start inserting realistic test-data into the selected table.
13. Test-data generator supports generating test values for Oracle's LONG data-type.
14. PostgreSQL: warnings and notices are reported separately from errors.
15. PostgreSQL: intellisense for SERIAL and IDENTITY column in INSERT/UPDATE statements.
16. All databases: intellisense for various SEQUENCE and sequence-like objects and functions, for example in Oracle, "<sequence_name>.nextval", in DB2: "NEXT VALUE FOR <sequence_name>", in PostgreSQL "naexval(<sequence_name>)" and pg_get_serial_sequence function, and so on
17. All databases: intellisense for JOIN suggestions now provides previews of table aliases for to be inserted code.
18. Sybase ASA: support for unsigned data types.
19. All databases: intellisense for INSERT...VALUES supports script variables and procedure parameters.
20. All databases: improved Intellisense for ALTER TABLE statement.
21. All databases: intellisense now supports coding database triggers.
22. Unit-test can now be scheduled directly from SQL Assistant interface.
23. Column name filtering is now available in code snippets using $OBJECT$ macros.
24. Refactoring templates now support standard set of SQL Assistant macros.
25. New pre-defined interactive code snippets.
26. Support for latest versions of MySQL Workbench.
27. Support for latest versions of Ultraedit.
28. Lots of other minor changes and enhancements.
Last edited by SysOp on Fri Sep 30, 2011 8:08 pm; edited 1 time in total |
|
Thu Jun 23, 2011 9:19 am |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
|
|
Excellent :)
Where shall we send error reports? Create a new topic for each one, post them here or send by email?
|
|
Thu Jun 23, 2011 9:59 am |
|
 |
SysOp
Site Admin
Joined: 26 Nov 2006 Posts: 7948
|
|
|
|
Please post new issues here.
Thank you.
|
|
Thu Jun 23, 2011 10:36 am |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
Code Formatting anomaly |
|
Having Code Formatting for Operator Spacing set to Space Before and After Operator breaks double-character operators (+=, -=, /=) in SQL Server
eg.
 |
 |
SELECT @i += @j
SELECT @i -= @j
SELECT @i *= @j
SELECT @i /= @j
|
will become
 |
 |
SELECT @i + = @j
SELECT @i - = @j
SELECT @i *= @j
SELECT @i / = @j
|
after the first application of formatting rules and
 |
 |
SELECT @i + = @j
SELECT @i - = @j
SELECT @i *= @j
SELECT @i / = @j
|
applying them the second time. No more changes occur in subsequent formatting.
This obviously results in a syntax error. Strangely *= remains unformatted (thus correct). While += and /= get two spaces, -= only gets one space in between, no matter how many times I apply the formatting.
I tried to create a += formatting rule but it did not help.
|
|
Thu Jun 23, 2011 7:18 pm |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
Small typo |
|
There is a typo in the dialog box asking 'Workspace is not versioned. Whould you like to checkout head revision?'
|
|
Fri Jun 24, 2011 3:29 am |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
Execution plan and CTE |
|
SA fails to retrieve the estimated execution plan when a CTE is involved in the query.
|
|
Fri Jun 24, 2011 4:43 am |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
Performance analysis |
|
Though clustered index scan are slightly better than full table scans, perhaps you could threat those the same way when analyzing performance. They are considered equally bad in certain situations (personal opinion :)
|
|
Fri Jun 24, 2011 5:27 am |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
UPDATE and ? |
|
Intellisense for UPDATE still inserts those wicked questionmarks.
|
|
Fri Jun 24, 2011 5:39 am |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
Source Control |
|
I created an SCS project with the following settings:
Workspace Directory: "T:\TEMP\TP_01"
Repository Server URL: "https://xp-repo/svn/Teszt/"
Project Path in Repository: "TP_01/trunk/SQL"
I was wondering if the contents of the repository on the given path will check out into ""T:\TEMP\TP_01\SQL" or plainly into "T:\TEMP\TP_01" but SA created the folder structure "T:\TEMP\TP_01\httpsxp-reposvnTeszt\TP_01\trunk\SQL" instead. Not exactly what I expected to happen. Is this the intended behavior?
Besides that, checking out the empty repository and updating the workspace with the contents of the database were flawless the first time. But after modifying one table in the database and updating the workspace (flawlessly once again) things have gone a bit awry. While trying to update the workspace skips all the tables saying that there aren't any modifications, comparing the database to the repository reports status for all the tables to be database only, though I just altered one of them. Also, trying to update the workspace from repository shows no files to be selected and trying to commit the changes made leaves me with the commit dialog in the screenshot:
Am I doing something wrong here?
Last edited by gemisigo on Fri Jun 24, 2011 11:41 am; edited 3 times in total |
|
Fri Jun 24, 2011 8:37 am |
|
 |
SysOp
Site Admin
Joined: 26 Nov 2006 Posts: 7948
|
|
|
|
Tank you very much for your feedback.
Would you please re-attach the image in a different format? The attached image seems to be broken.
Also if you can, please provide an example for the following
 |
 |
Though clustered index scan are slightly better than full table scans, perhaps you could threat those the same way when analyzing performance. They are considered equally bad in certain situations (personal opinion :) |
|
|
Fri Jun 24, 2011 9:33 am |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
|
|
Sure. Sorry for the screenshot, I don't know what happened, it seemed to be OK when I checked the preview. I hope it is visible now.
As for the example, I can only include execution plan as a screenshot. Example taken from a live database, tables cloned and indices created (or removed where necessary).
All three tables pairs are of the same data structure and data content. The first pair does not have any indices at all, the second pair has clustered indices on the columns used for joining and the third pair has a non-clustered index on join columns and includes the remaining columns that are SELECTed.
Surely, having a clustered index with the joined columns being ordered allows merge join instead of hash join and that makes quite a difference but having a well aimed non-clustered index probably beats these by miles. Then again, it's just my personal opinion :) Nevertheless, while in case 1 SA does warn about the existence of a better solution, in case 2 it does not.
With 3475 records in foldhely and 5144 records in kocsiall, though a small number of record, the difference is apparent.
Is there a way to attach files other than pictures?
|
|
Fri Jun 24, 2011 12:42 pm |
|
 |
SysOp
Site Admin
Joined: 26 Nov 2006 Posts: 7948
|
|
|
|
Would you please email files to support, with a reference to this topic?
And thanks again for your great feedback. Very helpful and allows us to fix issues quickly. I checked our internal issue tracking site and I see that two of the issues you reported today have been already auctioned. I hope they will be fixed in the next beta build (I assume that build would be available on Monday next week)
|
|
Fri Jun 24, 2011 1:18 pm |
|
 |
SQLTeacher
Joined: 27 Jun 2011 Posts: 1 Country: United States |
|
|
Mon Jun 27, 2011 1:45 pm |
|
 |
SysOp
Site Admin
Joined: 26 Nov 2006 Posts: 7948
|
|
|
|
Hi, both links in your message seem to be identical and working fine for me. I also tried the link on this page http://www.softtreetech.com/sqlassist/isql_beta.htm and that one is working fine too. If you have more than 1 browser on your system, can you try if the link works in your other browser?
|
|
Mon Jun 27, 2011 2:51 pm |
|
 |
gemisigo
Joined: 11 Mar 2010 Posts: 2165
|
|
Repository Browser |
|
When right-clicking a database node in the DB Connection tree and selecting Update Workspace from Database the resulting Progress of Work dialog says 'Local copy updated.' almost immediately but it does not do anything at all. Shouldn't it create a local copy of each and every object in the database like it does when trying the the same on the Tables/Views/Procedures/Functions nodes separately?
UPDATE: Strange, it did not work for the first database node the first three times , but it did after trying it on other ones. Cannot reproduce it since then. Might be machine specific.
UPDATE: also there seems to be some sort of caching present. I dropped and recreated the repository but after refreshing the repo tree still showed the nonexistent items with a blue questionmark overlay icon on them. The workspace was also cleared (though there may be some hidden files left).
I'm a bit confused about the icons. The one with the blue questionmark is supposed to be 'Compare Database to repository...' and the one with the red is 'Compare Database to Workspace...'. I updated the workspace from database and then altered a table in the database. Now comparing database to workspace indicates there is a change but comparing database to repository does not. Is it really comparing db to repo? Or workspace to repo? Also, update repo from db shows no changes to be committed.
|
|
Mon Jun 27, 2011 3:37 pm |
|
 |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|
|