UTFBIN is case sensitive on non-latin chars

[former_member329524]

Hello, all

I have a version 16 DB with Cyrillic text in it.

When I try to search for rows with specific text in there using sql command like

select * from table1 where field1 like '%техт%'

the search is case-sensitive, even when the DB itself is not.

If I am searching for text with latin characters, the case of the letters is ignored, as it should be.

Dose anyone know how to fix it?

Thank you

Arcady

[Former Member · Accepted solution]

Hi Acardy,

What is your dbinfo output? I wasn't able to reproduce the behaviour. For example,

SELECT* FROM table1 WHERE field1 = 'Техт'

or

SELECT* FROM table1 WHERE field1 = 'техт'

both returned the same expected row that has the text 'техт'.

Jinwoo

[former_member329524 · Accepted solution]

SQL Anywhere Information Utility Version 16.0.0.2234

Database  : C:\DB\Sybase16\Amarant.db

Log file  : C:\DB\Sybase16\Amarant.log

Log mirror: none

Page size : 4096

Encrypted : No

Strings padded with blanks for comparisons: No

CHAR collation sequence: UTF8BIN(CaseSensitivity=Ignore)

CHAR character set encoding: UTF-8

NCHAR collation sequence: UCA(CaseSensitivity=Ignore;AccentSensitivity=Ignore;PunctuationSensitivity=Primary)

NCHAR character set encoding: UTF-8

Database checksums enabled: Yes

Encrypted tables supported: No

[Former Member · Accepted solution]

Hi Acardy,

Are you querying against CHAR type columns? To meet your requirement, your CHAR collation sequence also needs to use UCA - it's using UTF8BIN which is by default.

As you probably know already, this requires your database to be rebuilt and your dbinit command need -z UCA to specify CHAR type also use UCA.

Regards,

Jinwoo

[former_member329524 · Accepted solution]

Problem soled,

Thank you, Jinwoo