Unicode problems

[Leon Kwan]

Hello all expert,

I would like to ask something about unicode datatype.
May be I should start from what I am doing now for the non-unicode data.
Now I have a table:

test
====
id int
keyword char(100)

In my existing SQL statement, one of them is used to search the records
which the keywords contains a input value, say "C++".
i.e. select * from test where upper(keyword) like upper('%sql%&#39
therefor all those records with keyword contains no matter "sql" or "SQL"
or "Sql" will be selected.
However if I want to apply chinese language in our database, I found that
this SQL statement will be no longer correct, because there exists some
chinese characters that the uppercase of them will becomes another word.
e.g. Â*»´ä, the upper case will becomes Â*»´Ä. This will search more unwanted
records as well.
so is there any workaround to due with this such that I can still maintain
the english matching feature?

Thanks in advance.

On 1/25/99 6:30:20 AM, Leon Kwan wrote:
Hello all expert,

I would like to ask something about unicode datatype.
May be I should start from what I am doing now for the non-unicode data.
Now I have a table:

test
====
id int
keyword char(100)

In my existing SQL statement, one of them is used to search the records
which the keywords contains a input value, say &#34;sql&#34;. <== should be &#34;sql&#34;
i.e. select * from test where upper(keyword) like upper(&#39;%sql%&#39
therefor all those records with keyword contains no matter &#34;sql&#34; or &#34;SQL&#34;
or &#34;Sql&#34; will be selected.
However if I want to apply chinese language in our database, I found that
this SQL statement will be no longer correct, because there exists some
chinese characters that the uppercase of them will becomes another word.
e.g. Â*»´ä, the upper case will becomes Â*»´Ä. This will search more unwanted
records as well.
so is there any workaround to due with this such that I can still maintain
the english matching feature?

Thanks in advance.