madamsMy site contains some Japanese, and I can't get it totally restored when reimporting my database (pages with Japanese on them have gibberish interspersed with words that I know should be appearing eg. <http://raysend.com/mark/purofairu/>; note the name of the Page comes up correctly but the name of the site is 1/4th borked).
When I originally created the database, the encoding was latin_swedish_ci. and the MySQL charset is UTF-8 Unicode (utf8). In 'Reading Options > Encoding for pages and feeds' it's set to UTF-8. Despite the mismatch between the database collation and the page encoding, I never had a problem with Japanese text.
What charset should I select when importing the database tables, or for recreating the database. I wonder if there's some magic combination of 'MySQL connection collation', database Collation, and 'Character set of the file' to upload, or any bit-twiddling I can do to the backup.
Since the Japanese on my site is mainly used for category names and a Page here or there, it's not out of the question to fix by hand. What the, should I do to the database itself to make sure I don't have this problem again?
