Hi, On Tuesday 08 March 2005 22:53, Korsani wrote: > It is not blocking, but rather anoying. > Try the script and you'll see: > Attempt to use reference as lvalue in substr at > /usr/share/perl5/Net/LDAP/LDIF.pm line 348 [or 346, I don't remember] > > It is due that at this line $_[0] could be something like 'do bless > ...'. Code, in other words. That means that if you try to 'print > $_[0]' you won't see the code, bu substr while complain.
I tested the program and was able to reproduce the error, but I am not sure whether I should consider this an error in Net::LDAP::LDIF. The cause of the trouble is that the construct "cn=$cn,$base" with $cn being a Unicode::String object is still a Unicode::String object. This object gets fed into Net::LDAP::Entry and, when written using Net::LDAP::LDIF, leads to the undesired warning. As a workaround I suggest changing the line $entry->dn( "cn=$cn,$base" ); to $entry->dn( 'cn='."$cn".",$base" ); This change makes the argument to $entry->dn() a string instead of a Unicode::String object and avoids the problem. What do other people on the list think about it ? Is it a bug in Net::LDAP::LDIF ? Is the bug in Net::LDAP::Entry for not properly checking its input ? Or is it even a bug in Unicode::String [incomplete implementation of substr()] ? CU PEter -- Peter Marschall eMail: [EMAIL PROTECTED]
