Re: DBD::Oracle not utf8 encoding PL/SQL out values

On 07/02/2012 19:42, Steve Baldwin wrote:
> Thanks a lot Martin.  Created
> https://rt.cpan.org/Public/Bug/Display.html?id=74753.
>
> Cheers,
>
> Steve

Could you post on the rt  if you apply the patch and it works for you.

Thanks

Martin

> On Tue, Feb 7, 2012 at 8:28 PM, Martin J. Evans
> <[email protected]>wrote:
>
>> On 07/02/12 04:56, Steve Baldwin wrote:
>>
>>> #!/usr/bin/perl -w
>>> use strict;
>>> use warnings;
>>> use DBI;
>>> use Encode;
>>>
>>> sub main {
>>>      my $dbh = DBI->connect(
>>>          'dbi:Oracle:',
>>>          'usr/pwd@instance',
>>>          '',
>>>          { PrintError =>   0, AutoCommit =>   0, RaiseError =>   1, },
>>>          );
>>>      print {*STDERR} 'Connected !', "\n";
>>>      $dbh->do(q(alter session set nls_territory = 'GERMANY'));
>>>      my $sql =<<'END_SQL';
>>> SELECT  ltrim(rtrim(to_char(0, 'L')))
>>> FROM    dual
>>> END_SQL
>>>      my $sth = $dbh->prepare($sql);
>>>      $sth->execute;
>>>      my ($val);
>>>      $sth->bind_columns(\($val));
>>>      while ($sth->fetch) {
>>>          print {*STDERR} 'val=[', $val, '] len=', length($val), "\n";
>>>      }
>>>      if (Encode::is_utf8($val)) {
>>>          print {*STDERR} ' utf8 is on', "\n";
>>>      } else {
>>>          print {*STDERR} ' utf8 is off', "\n";
>>>      }
>>>      $sth->finish;
>>>      $sql =<<'END_SQL';
>>> declare
>>>      l_ret       varchar2(10);
>>> begin
>>>      select  ltrim(rtrim(to_char(0, 'L')))
>>>      into    l_ret
>>>      from    dual;
>>>      --
>>>      :ret := l_ret;
>>> end;
>>> END_SQL
>>>      undef $val;
>>>      $sth = $dbh->prepare($sql);
>>>      $sth->bind_param_inout(':ret', \$val, 100);
>>>      $sth->execute;
>>>      print {*STDERR} 'val=[', $val, '] len=', length($val), "\n";
>>>      if (Encode::is_utf8($val)) {
>>>          print {*STDERR} ' utf8 is on', "\n";
>>>      } else {
>>>          print {*STDERR} ' utf8 is off', "\n";
>>>      }
>>>      $dbh->disconnect;
>>>      return 0;
>>> }
>>>
>>> exit main();
>>>
>> Fails for me to with the very latest DBD::Oracle from subversion trunk.
>>
>> Is it a bug? I guess so as DBD::Oracle is supposed to do the right thing.
>>
>> Try this change:
>>
>> Index: dbdimp.c
>> ==============================**==============================**=======
>> --- dbdimp.c    (revision 15130)
>> +++ dbdimp.c    (working copy)
>> @@ -3197,7 +3197,14 @@
>>                 if (SvPVX(sv)) {
>>                         SvCUR_set(sv, phs->alen);
>>                         *SvEND(sv) = '\0';
>> -                       SvPOK_only_UTF8(sv);
>> +            SvPOK_only_UTF8(sv);
>> +            if (CSFORM_IMPLIES_UTF8(SQLCS_**IMPLICIT)) {
>> +#ifdef sv_utf8_decode
>> +                sv_utf8_decode(sv);
>> +#else
>> +                SvUTF8_on(sv);
>> +#endif
>> +            }
>>                 }
>>                 else {  /* shouldn't happen */
>>                         debug = 2;
>>
>> It seems to work for me but I don't really understand why DBD::Oracle code
>> calls SvPOK_only_UTF8 as I cannot see why UTF8 status of the scalar would
>> already be on.
>>
>> BTW, that code originates from a patch in April 2002 so it is OLD code -
>> http://www.mail-archive.com/**[email protected]/msg10046.**html<http://www.mail-archive.com/[email protected]/msg10046.html>
>>
>> You should probably rt this at rt.cpan.org.
>>
>> Martin
>> --
>> Martin J. Evans
>> Easysoft Limited
>> http://www.easysoft.com
>>
>
> ---------------------------------------------------------------------------------------
> This email is intended solely for the use of the addressee and may
> contain information that is confidential, proprietary, or both.
> If you receive this email in error please immediately notify the
> sender and delete the email.
> ---------------------------------------------------------------------------------------
>