Bug:MYSQL 4.0 -> 4.1 Sync on 2byte environment

[sdozono]

Thank you for this useful tool, SQLyog.

I'm using:

– SQLyog 4.1 Beta3(Latest.)

– MYSQL 4.0.22-standard + SQLyogTunnel.php (ISP)

– MYSQL 4.1.12 (Fedora Core 4. So, default MYSQL is 4.1) + SQLyogTunnel.php

Then, I set all the enviromental variables, and checked. And found one bug of SQLyog.

Test:

– MYSQL 4.0 -> 4.1

md5 Check => OK.

Copy Table to Different HOST/DB => OK. No problem.

HTTP tunneling/default character sets for 4.1 => OK.

– Database Synchronization from “4.1” to “4.0” => OK.

But,

***- MYSQL 4.0 -> 4.1 Database Synchronization => Failed.(SQLyog put in garbled characters and delete those by itself…)***

After spending some time, I found that:

SQLyog publishes wrong “Insert into” Query when synchronize from 4.0 to 4.1.

For example, if I have a data like,

“C6/FC CB/DC B8/EC” (Source MYSQL 4.0)

SQLyog changes for a query,

“C3-86/C3-BC C3-8B/C3-9C C2-B8/C3-AC” (for MYSQL 4.1)

I want SQLyog to publish “the same character codes for MYSQL 4.1 if two MYSQL's charactersets are same”.

So, I added this code into SQLyogTunnnel.php

(This is just filtering “0xC3” and “0xC2”, and restore original Codes.)

Then, it started synchronize properly.

Maybe, SQLyog is changing double bytes characters into garbled characters with UTF-8 convert function….

So far, I can synchronize 4.0 and 4.1 in this way, but this can work only as far as the source MYSQL doens't have garbage characters. (and it happens often.. 😉 )

So I hope your SQLyog team consider this….

What do you think?

=================================

function FilterStr($str)

{

$ary = array();

for($i=0; $i

---

Shunro Dozono

http://cba-japan.com/english

[sdozono]

And one more,

in the function ProcessQuery(),

I added..

> $tmp_query = “SET NAMES UJIS “;

> $tmp_res = mysql_query($tmp_query);

mysql_select_db ( str_replace ( '`', '', $db ), $mysql );

I'm using UJIS(EUC-JP) environment…

I hope this information is helpful for next version….

[Ritesh]

Thanks for your detailed reply. We plan to fix all the character-set related issues in v4.2.

As we had stated earlier, we would definitely require your help. 🙂

[Author]

Posts