Forum Replies Created
-
AuthorPosts
-
peterlaursen
ParticipantCollation has nothing with the display to do. Only the character set has. Collation has with sorting to do primarily.
you should use latin1 or utf8 charset .. and a French collation collation if such exists if you want sorting done according to the rules of French language!
peterlaursen
ParticipantWhat it tried to say was that we do plan an 'overhaul' data synchronisation, including additon of more features. However we are fully 'booked' for the next about 2 months.
So only after those weeks we will plan for the future. And only after that we can tell if this will be included in 'next step' for us.
This is not a new request. I think a 'prefix-substition' mechanism will do for your too so that 'remote_mytables' can be synced with 'local_mytables'? A completely random mapping of soruce and target tables would easily become very messy!
The request for a formal XML-reference for SJA is not new not either. I agree we should have such – and it be be useful for us internally as well. We hope to be able to find a 'documentationist' who is also a 'code-reader' very soon.
peterlaursen
ParticipantExcept for this FAQ
http://webyog.com/faq/category27.html
.. and the examples that ship with SJA
we don't have.
You should consider SJA for Linux primarily as 'extended offer' for our SQLyog Enterprise users. You can build jobfiles from SQLyog GUI on Windows and copy the jobfile to a Linux machine!
peterlaursen
ParticipantThe short answer is NO!
The more extensive answer is that after (probably) 6-8 weeks we will have completed what we have started. Only after that we will plan for the next period to come!
Can you explain why this is important for you?
July 17, 2007 at 1:26 pm in reply to: Copy Database To Different Host/db Corrupts Stored Procedures #24532peterlaursen
ParticipantYou should ALWAYS tell the program version you are using (and research a little in the release notes of recent versions)!
The issue is not reproducable with version 6.05 beta 3. In the 6.05 tree we did another SP-related fix..
So please try 6.05 beta3!
If problems persist then please attach an EXACT case: what is is before and what it became after copy!
peterlaursen
Participantissue is confirmed!
We will try to 'push in' a fix for this next week.
peterlaursen
ParticipantIf you open more connections will it then crash if you close one of the connections?
peterlaursen
Participanthi …
did 6.x solve the issue or do you need more help on this?
peterlaursen
Participantok .. let us get back to the 'crash on exit' issue.
I have experienced that with previous versions with SuSE 10.0 too. Now I always update both SQLyog and Wine when a new version of either gets available and I cannot remember exactly what combination of program versions it was.
What is your Wine version? (console command 'wine –version' tells!) ?
peterlaursen
ParticipantYes, it is my 'impresssion' .. but I do not understand how you think your reply should help us to help you?
Still: are you able to connect to MySQL with any other PHP-script/application?
(Even if phpinfo() tells the extension is enabled it may not be functional.)
There also could be some issue related to some setting in Windows Server 2003. Is the webserver IIS or Apache (or something more 'exotic') ?
Actually we would like to see the complete phpinfo() response ourselves. Also if possible we would like to connect to the tunneller from inside a debugging environment. From such we can trace exactly what does not match the response that we expect.
I suggest that you create a ticket here
http://www.webyog.com/support/ttx.cgi
… so that your details can remain confidential.
… and do not forget to tell the EXACT SQLyog program versions that you tried! And if you did not try LATEST (6.04 or 6.05 beta) please try this too.
peterlaursen
Participant@krollaw
I think the error messages tell allt what there is to tell! PHP on your host is not configured to connect to MySQL. Only the SysAdmin/support can solve that! You will have to contact your ISP!
Now it is very unusual that PHP runs without MySQL support! So probably it is a silly mistake during PHP compilation or configuration!
Do you know about the PHP version?
If you create a simple PHP file with the single command
Code:phpinfo();.. you can connect to it with a browser and verify that no mysql or mysqli extension is loaded! If phpinfo() tells that the extension is loaded or if you know positively that some other PHP-script can connect to mysql, then we will have to deeper into details …
peterlaursen
Participantwe cannot use the crash dump from a Wine/Linux install unfortunately! The Visual Studio version we are using won't run on Wine!
I don't know Ubuntu, but with KDE on Fedora and SuSE you right-click in the menu area and add an entry 'wine SQLyog.exe'.
But I think Ubuntu desktop is based on GNOME?
peterlaursen
ParticipantIt was a silly idea then! Thanks anyway!!
At this moment I cannnot tell what we will do. It does not make sense to change anything as long as the issue is not fully understood.
If 5.32 works to your satisfaction then there also is no hurry, I think! But we do intend to find out!
peterlaursen
ParticipantIt is not that easy! I also have explained why 5.32 works. But it will not with multilingual data or languages for which no non-unicode charsset is availbale. It will not either if data uses one ANSI codepage and the computer uses a LOCALE setting not matching this.
The 'mysql' client you are using is running on Windows or Linux?
Also I think you never told about the server platform?
I have one (maybe silly) idea. Maybe the 'SET NAMES UTF8' statement simply comes too fast! Could you try connecting and then 'SET NAMES UTF8;' yourself from SQL editor before entering any data?
Also: do you have an option to try on another server?
peterlaursen
ParticipantI don't like either solutions!
The SQLyog concept is the opposite of what MA does. In SQLyog you enter what shall be stored and you see what is stored! That means that escaping, doubling '' with LIKE and charset conversion are all done in the backgorund. User shall not need to worry about that!
The SQL server client solution is better .. at least if it was a configuration option. But is is a terrible lot of code changes if we should implement that I think.
I favour implementing the (EMPTY) keyword in CREATE/ALTER table DEFAULT column of the grid (and a context menu).
But I have asked collegues to think … (and believe it or not .. I had success with that before! <_< )
-
AuthorPosts