[Ritesh]

BETA 7 will be even better as we have implemented user defined CHARSET selection in HTTP Tunneling. I think we are now done with one byte data handling. v5.2 will have complete support for all kinds of languages.

[Ritesh]

Yes. I got the PM.

I will be working on it tomorrow. We have BETA 5 code with us. Lets hope that we are able to reproduce the problem. We will be then able to find the issue out.

BETA 6 seems to have solved the issue as another guy confirmed the bug fix too but the interesting thing is that we never changed any code regarding this issue between 5 and 6.

[Ritesh]

Bug fixed in BETA 7 development tree. The same bug effected ODBC Import wizard too so we fixed that too.

[Ritesh]

GREAT suggestion!

Why the f**k did we miss it all these years 😮

Scheduled for 5.2

[Ritesh]

Terminate button becoming invisible means that the background SJA process is over. If you execute SJA from the command line, the application should come out in between.

What happens when you run SJA from the command line?

[Ritesh]

Fixed in BETA 7 development tree.

[Ritesh]

Point 1 and 2 will be fixed today.

I can get point 3. I think you are getting them too as posted in another topic.

[Ritesh]

peterlaursen wrote on Apr 9 2006, 03:04 PM:

I added a column to a table like

Code:

alter table `test2`.`PLADDER` add column `newcolumn` varchar (50) NULL after `id`

.. no matter if I use the ALTER TABLE GUI or type it

and next write in editor

Code:

select * from pladder where newc

.. autocomplete does not find 'newcolumn'. Not even after a 'Rebuild tags' or a restart of the program.

(or similar if a Add a table ttttt and write select * from tt

After deleting all tag files it works as expected after a restart. So might be that the file was bad. It also has been rebuilt several hundred times!

I again request that the tag file name is human readable like connectionname_checksum.db and not only checksum.db . Then I would only have had to delete one file and not all about 40 files. I also request a embedded database maintenaince functionality (SQLite VACUUM command for instance) from inside SQLyog. And maybe even a 'Rebuild from scratch' option.

BTW is there any 'limit' to how many characters you will have to type for autocomplete to be functional?

typing tt does not find ttttt!

This looks like a bug. I will work on it today.

[Ritesh]

peterlaursen wrote on Apr 9 2006, 04:16 PM:

Thing got a little confused .. I'll write a summary of issues and requests here:

1) jrossitter reported a crash, that I could reproduce. Should be fixed.

2) jrossiter reported slow performance when deleting text from the editor. I cannot reproduce this, but Ritesh has confirmed the issue and plans to solve this in BETA7

3) jrossitter requests that autocomplete should only be effective 'at the end of a word', that is if there is a , a TAB, a LINEBREAK or NOTHING after cursor position. Reasonable in my opinion, but not high priority here.

4) I request that backticks ” `” (leading backticks at least) are stripped out from strings before looking up matches in the database. It would be elegant if a backtick was added at the end of the inserted string if it was stripped out at the beginning!

5) I request that TAG-file names are human readable. So that you have a chance to delete unused and corrupt files and only those.

6) I request the current 'rebuild TAGs' menuitem expanded to a menu 'group'

* rebuid TAGS (fast)

* rebuild TAGS from scratch (slow)

* REPAIR and VACUUM tag file (what ever SQLite allows for)

7) Finally there was a request (don't remember by whom) a few weeks ago that ALIAS'es should be supported. This is probably a tricky one, as an ALIAS only is valid inside a single statement. SQLyog will have to parse the complete statement for ALIASes when cursor is moved to another statement. Anything after an 'AS' followed by SPACE, TAB or LINEBREAK will have to be added to some temporary db (saved in memory only), and shall not be saved to disk when exiting! But this would 'kick'! For use with not-so-powerfull cumputers I think it should be an option!

1.) Fixed in BETA 7.

2.) Fixed in BETA 7.

3.) I am not sure if I understood the issue. If I show it at end then there is no meaning of TOOLTIP. The person has already written the word. One extra option we are adding in BETA 7, whereby a user can configure to show all TOOLTIP or autocomplete will only display a dropdown on pressing ., ( or ctrl + space and ctrl + enter. So you will not get tooltip for keywords if you dont require it.

4.) Currently being worked in the BETA 7 development tree.

5.) Added for v5.2.

6.) Hmm…Added in the TO-DO list but not sure which version we will implement this.

7.) Added in the TO-DO list but not sure which version we will implement this.

[Ritesh]

Great to know that. We fixed around 6 data crash issue between the last 3 BETA releases.

[Ritesh]

Yes, indeed this strange. Couple of more people had reported the same issue. Let us wait for couple of days so that other people can install and test out BETA 6.

The strange thing is that we are not able to reproduce it at our test servers so kind of difficult to find out 🙁

[Ritesh]

Thanks for the bug report. We will work on it today and fix the issues.

[Ritesh]

Thanks for the sample data. I will take a look into it today.

[Ritesh]

So did you get your upgrade key or you still want some help?

[Ritesh]

jrossiter wrote on Apr 9 2006, 01:47 AM:

Thanks about the menu – I swear I looked.

About the delay, it's not something that I have mentioned personally, but I have seen other people mention it.

It may be tied to not having tags built, but if I select a block of text and delete it, the UI pauses for 2-3 seconds before it responds while it tries to parse/autocomplete.

I see the same behavior if I am just keyboard-navigating the query window (arrow keys) when I move between lines or statements.

This is a known issue and we have actually improved upon it in BETA 7 development tree. BETA 7 which we expect to release on Wednesday will have this improvement.

[Author]

Posts