[RAS] processing aborts

Thomas Krichel krichel at openlib.org
Sat Oct 24 08:07:00 CDT 2009


  Thomas Krichel writes

> The insert of the two lines appears to have worked, but the nightly
> job has interrupted processing. The job is running on channel 0.
> If it finishes after 15:00 we have a problem beacues I will
> leave and not return until Sunday morning. We need somebody
>  to make the DNS change back when the process is finished.

  Actually, this concerns RAS and should have been sent 
  to RAS-run. I am sorry for my oversight. 

  To summarize, processing of CitEc aborted. I added two 
  encoding statements. 

  In 

/home/aras/acis/lib/ACIS/Citations/Input.pm

      $prev_ostring =
        my $ost = $_->{ostring};
      #tok 2009-10-23
      $ost=Encode::encode('utf8',$ost);
      #tok 2009-10-23 
      my $md5 = Digest::MD5::md5_base64( $ost );
      my $clid = "$srcdocsid-$md5";
      delete $index->{$clid};

  In 

/home/aras/acis/lib/RePEc/Index/Collection/CitationsAMF.pm

 my $ctx = Digest::MD5->new;
  $ctx -> add( "$id\n" );
  foreach ( @$self ) {
    if ( not ref( $_ ) ) { next; }
    my $s = " ";
    $s .= $_->{ostring};
    if ( $_->{trgdocid} ) {
      $s .= '@';
      $s .= $_->{trgdocid};
    }
    $s .= "\n";
    #tok 2009-10-23
    $s=Encode::encode('utf8', $s);
    #tok 2009-10-23 
    $ctx -> add( $s );
  }
  return $ctx->b64digest;

  This is not innocent code, I think. 


  Cheers,

  Thomas Krichel                    http://openlib.org/home/krichel
                                http://authorclaim.org/profile/pkr1
                                               skype: thomaskrichel



More information about the RAS-run mailing list