And when will the first Windows version of ICE will be released please.

Amber,

Native UTF-8 support is currently planned for 2Q09.

Infobright Community Edition for 32-bit Windows is currently planned for the mid-February, 2009 release.

Thanks.

I have tried ICE 3.1 for Win32, Chinese text encoded with Windows(China locale) default code page can be loaded and exported correctlly, but the mysql client tool and Toad for MySQL freeware can’t display the loaded Chinese words.

How is your UTF-8 support going?

Hi !

Currently you can load any character set to VAR/CHAR columns, but ICE does not recognize character sets and treats them as sequence of 1-byte values. Therefore
- sorting/comparisons do not not handle collation,
- a 3-byte character is treated as 3 1-byte characters
- character set information is not passed consistently from engine to clients, so necessary character conversions may be omitted. Example is given in http://www.infobright.org/index.php/Forums/viewthread/551/ - results of MIN() aggregation

However, if you setup the environment correctly, selecting a UTF-8 column from a table should succeed - bytes sequences will be retrieved, they will be passed to mysql, which should take care for converting them into clients local character set. Example in http://www.infobright.org/index.php/Forums/viewthread/551/ is select distint [...]

Hi,

One important clarification on UTF-8 in the current version: to store it properly, columns must be declared with an appropriate width (in bytes). I.e. for safe storing of 20-characters UTF-8 values, one should use VARCHAR(80), as every character may be 4 bytes wide.

Regards,

David Lutz - 20 January 2009 12:17 PM

Amber,

Native UTF-8 support is currently planned for 2Q09.

Infobright Community Edition for 32-bit Windows is currently planned for the mid-February, 2009 release.

Hi, it’s June now, can UTF-8 support come out at the end of this month as planed?

Hi, it’s June now, can UTF-8 support come out at the end of this month as planed?

Hi Amber,

Regrettably, UTF-8 has been moved into Q3 in favour of enhanced SQL function support for complex expressions in the upcoming release in July. As soon as I have a firm new date for UTF-8, I will let you know.

Sorry again, and thanks for your patience!

Cheers,
Andrew

Andrew,

By any chance, is this enhanced support going to include optimized grouping by the available aggregates? I really hope so, that would be -excellent-. That has really been my biggest stumbling block so far, because I pretty much refuse to enable the MySQL fallback, and putting all of this stuff into subqueries takes a lot of extra work and often results in rather poor performance.

Hi jwilhelm !

If you mean things “... group by a+b”, or “select year(date1)-year(date2) ... group by 1”, then yes - the enhanced support will cover it.

As well as “SELECT YEAR(my_date), SUM(a*b) FROM t GROUP BY YEAR(my_date)”.

Regards,

Hi, it seems UTF-8 will be avaiable at least after 3.2, and 3.2 will come out in Sept. at least:)

I see IEE 3.3.1 supports UTF-8, what about ICE3.3.1?

It will be supported in both, probably in the next 3.x release of ICE.

Geoffrey

But the document of ICE 3.3.1 still says:

Supported Character Sets
Infobright storage supports all ANSI character sets. This means that Infobright can store and
retrieve data encoded in any 8-bit character set. Infobright does not support any other
character sets, such as the Unicode, UTF-8 or Latin1 (cp1252) character sets.

See http://www.infobright.org/Forums/viewthread/1383/