Re: Can't identify character set and collation used in string comparisons

[David Raimosson]

The values I reported for the SHOW VARIABLES LIKE 'char%' have already been obtained using the .NET connector, dumping them as strings from two columns the way you said.

And yes, it's true that the character_set_results variable doesn't have a value, or rather a NULL value. The reason for that is explained in the example here:

http://www.csse.uwa.edu.au/programming/mysql/connectors-apis.html
Section 22.2.5.14.1. Viewing MySQL Trace Information.

As seen in the example the character_set_results variable is explicitly set to NULL by the Connector/.NET, which is also done on my machine. My Connector/.NET also issues the following command:
"SET NAMES utf8", right before clearing the character_set_results variable.

(An explanation of why the connector clears this variable goes here in the final example:
http://dev.mysql.com/doc/refman/5.1/en/charset-connection.html)

SELECT HEX('ø'), 'ø' gives:
C3B8, ø

My current suspicion is that the utf8 strings being sent in gets a higher collation precedence when compared to the latin1 column. Could that be possible?

Best regards

David



Edited 1 time(s). Last edit at 07/23/2013 03:47AM by David Raimosson.