HSQLDB uses Java's Collation support. It probably depends on the Java version but I've noticed it ignores spaces with some high strength settings although the docs say it shouldn't.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
behave exactly the same way.. (case sensitivity is ignored regardless of strength)
But when using SQL_TEXT or SQL_TEXT_UCC it works as desired (whitespaces are taken into account, etc.), but accented letters won't be compared correctly. That is, in my example, "ÁABBCC" will be put at the end.
Is that expected? Wouldn't SQL_TEXT (UNICODE char) be equivalent to utf8_general_ci from MySQL?
SQL_TEXT or SQL_TEXT_UCC do not use Java collations. MySQL uses its own collation. We can define collations and add them to HSQLDB. You can develop one for this language and contribute it to the project.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Please, consider the list/values:
The above list is the result of the following command (using HSQLDB 2.3.2 on Windows 8.1 Pro):
I would expect the result to be:
With whitespaces not being ignored in the comparison because of strength 3 specified.
So, am I missing anything, or is that a bug?
Thank you.
HSQLDB uses Java's Collation support. It probably depends on the Java version but I've noticed it ignores spaces with some high strength settings although the docs say it shouldn't.
Thank you for your quick reply.
OK, right.
I can notice it affects any language specific collation, be it English, French, or Portuguese. Also, these parameters, e.g.:
"English 0", "English 1", "English 2", "English 3", "English 2 1", etc.
behave exactly the same way.. (case sensitivity is ignored regardless of strength)
But when using SQL_TEXT or SQL_TEXT_UCC it works as desired (whitespaces are taken into account, etc.), but accented letters won't be compared correctly. That is, in my example, "ÁABBCC" will be put at the end.
Is that expected? Wouldn't SQL_TEXT (UNICODE char) be equivalent to utf8_general_ci from MySQL?
Live example: http://sqlfiddle.com/#!9/5e4cd/1
Thank you.
SQL_TEXT or SQL_TEXT_UCC do not use Java collations. MySQL uses its own collation. We can define collations and add them to HSQLDB. You can develop one for this language and contribute it to the project.
OK. Thank you for all the clarification. And yes, I may consider the development.