Here is the situation:

1. User enters 'Română' ($cat) to search for this subcat in DMOZ database.
2. However, 'Română' is encoded as 'Română' in database (as it is by default in DMOZ dump).
3. Tried the following code, but fails:

my $str = "SELECT `catid` FROM `dstructure` WHERE UCASE(name) = UCASE(CONVERT( _utf8 '$cat' USING latin1 ))";
my $sth = $dbh->prepare($str);
$sth->execute();
$catid = $sth->fetchrow_array();

Ideas? Tried various encoding of input, but no luck. Try another encoding method of dstructure.name?

----
Cheers,

Dan
Principal

Virtual Solutions

Hi Dan,

Did you submit your text as utf-8? I think with that, the text will be the same as the one in mysql?

Cheers,

Dat

Scripts installation and plugin creation
Plugins

1. DMOZ cat structure file saved locally as UTF-8 using ASCII editor.
2. File uploaded (in ASCII) to server.
3. Call CGI-Perl script to import file to mySQL. Script also outputs each category. Română appears as Română.
4. After import, Română appears as Română in mySQL database.
* database and table use utf8_general_ci collation

----
Cheers,

Dan
Principal

Virtual Solutions

But was the connection setup to be in utf8 as well?

Adrian

Hmmm, that sounds like it. Like on of these?

$dbh = UTF8DBI->connect("DBI:mysql:$dbname:$dbhost", $dbuser, $dbpass);

or

$dbh->do('SET dstructure utf8');

----
Cheers,

Dan
Principal

Virtual Solutions

SET CHARACTER SET charset_name

Adrian

Brilliant, thanks! Database is as should be. Hate to go back to the well one more time, but when script queries the database for Top/World/Română, not found:


$catid undefined. But, if I enter that same SQL query via phpMyAdmin, it is found. Works for other cats (e.g., Top/World), so appears to be encoding related. Connection is setup as utf8.

----
Cheers,

Dan
Principal

Virtual Solutions