perl-unicode

Inserting to Oracle using Win32::ODBC

2005-05-10 12:32:06
I have hit a brick wall trying to insert UTF-8 characters into Oracle 9i from a Perl script and would appreciate any help anyone has to give.

I'm collecting user input from a web form (with UTF-8 encoding set in the content-type header), then trying to insert the data into an Oracle 9i table. What is happening is...

When the user input string = "ÃÄĀāĂăàáâ", Oracle inserts "<x-tad-smaller>ÀÁÂÃÄĀāĂăàáâ" and utf8::is_utf8($user_input) returns TRUE.

... however ...

When the user input string = "éêëìíîïñòóôõ", Oracle inserts the correct string and utf8::is_utf8($user_input) returns FALSE.

In both instances, the correct string prints to the screen, so I don't know why Oracle inserts something different when utf8::is_utf8($user_input) returns TRUE .

Code is below. Thanks in advance for the help.


use utf8;
use Win32::ODBC;

$ENV{NLS_LANG} = '.UTF8';
binmode STDOUT, ":utf8";

$user_input = pack "U0C*", unpack "C*", $user_input;
utf8::decode($user_input);

$db = new Win32::ODBC("DSN=DB;UID=username;PWD=password");
$db->Sql("UPDATE TEST_TABLE SET NAME = '" . $user_input. "' WHERE ID = 0;");

$db->Close();

print "Content-Type: text/html; charset=UTF-8\n\n";
print "<p>User entered: " . $user_input;


- Stuart Green</x-tad-smaller>
<Prev in Thread] Current Thread [Next in Thread>
  • Inserting to Oracle using Win32::ODBC, Stuart Green <=