"a" and "A" are identical, according to utf8_unicode_ci and utf8_general_ci, but not
utf8_bin.
That is, I would not expect "[" to slip in between them.
About 80% of the way down this page, there is a list of some of the 'equivalences' in
selected collations:
http://mysql.rjweb.org/doc.php/charcoll
In particular:
utf8 : utf8_unicode_ci
A=a=ª=À=Á=Â=Ã=Ä=Å=à=á=â=ã=ä=å
ae az Æ=æ
says that lots of things collate equal to 'A'. All of them sort before ae > az >
the ligature Æ (upper or lower case)
> -----Original Message-----
> From: Wes James [mailto:comptekki@stripped]
> Sent: Wednesday, May 30, 2012 2:03 PM
> To: mysql general discussion
> Subject: sort order
>
> I have mysql 5.1.62-0ubuntu0.11.10.1 installed, but the sort order for
> utf8-unicode doesn't seem to be right. It is sorting some text like
> this (order by title):
>
> !
> ! *`-=[];',./~@#$%^&()_+{}|:"<>?\
> !"
> ![
> !@
> !a
> !A
> !t
> !test
> 'Coal age'
> [
> \--\
> 100 years
>
> when it should be (shouldn't it?):
>
> !
> ! *`-=[];',./~@#$%^&()_+{}|:"<>?\
> !"
> !@
> !A
> ![
> !a
> !t
> !test
> 'Coal age'
> 100 years
> [
> \--\
>
>
> ------------------------------
>
> Why is it doing this?
>
> Thanks,
>
> -wes