Hi list,
as announced in my post from 28. October (see below) I have committed
a patch to perl-ldap's SVN that allows to get attribute values from LDAP
directories and LDIF files correctly encoded as Perl strings.
This is especially interesting for people who have strings with non-ASCII
values in their directories.
The patch add an additional attribute called binary to Net::LDAP's new()
constructor and search() method that contains a regular expression which
matches all attribute names that are not to be converted to Perl strings.
If the binary attribute is not given at all perl-ldap's behaviour is the same
as before.
If it is given in new() it is automatically inherited by each search() unless
specified differently there.
Here is an example:
=== 8< snip >8 ===
use Net::LDAP;
# un-comment ithis if you have your locale set to UTF-8
#binmode(STDOUT, ":encoding(utf8)");
my $ldap = Net::LDAP->new('localhost', version => 3);
$ldap->bind('cn=Administrator', password => '****');
my $search = $ldap->search(base => 'ou=EIS,o=Dir4E',
filter => '(sn=*)',
attrs => [ qw/sn givenName mail/ ],
binary => qr/;binary|^jpegPhoto/io
);
print $search->error() if ($search->code());
while (my $entry = $search->shift_entry()) {
print lc($entry->get_value('sn')), " ",
lc($entry->get_value('givenName)), "\n";
}
=== 8< snip >8 ===
It is currently only lightly tested (but hey, this is what SVN is for ;-)
The adventourous of you out there, please test and give me feedback.
Have fun with it
Peter
On Friday, 28. October 2005 09:19, Peter Marschall wrote:
> I am trying to get non-ACII characters in attributes/objects that I get
> back from directories using Net::LDAP correctly encoded in Perl-unicode
> (I.e. so that uc("Hägar") results in "HÄGAR" and not in "HäGAR" or even
> errors)
> Unfortunately I have not found any solution that worked as I wanted.
>
> Then I tried to take another route:
> Why not let Convert::ASN1 do the dirty work of upgrading the UTF8 strings
> to Perl-unicode. For the other direction (sending Perl-uncode chars to LDAP
> directories) it already does it.
>
> Thinking about it a bit longer I came to the conclusion that Convert::ASN1
> might be the wrong place to do it since there may be some discrepancies
> between the ASN1 encoding and the programmers' needs (i.e. sometimes one
> may need to have the raw byte-value of an attribute)
> And besides, this change deep in Convert::ASN1 might break applications
> that use Net::LDAP.
>
> So I thought about adding a parameter 'binary' to Net::LDAP->new() and
> Net::LDIF->new() that contains a list / pattern of all the attribute names
> whose values should NOT be upgraded to Perl-unicode.
> Then, whenever a search is performed or an LDIF file is read
> Net::LDAP::Search and Net::LDIF take care of upgrading the values of the
> attributes that are not in the binary list to Perl-unicode.
> (I chose the negative selection 'binary' to get the list shorter, because
> IMHO most anttributes in an LDAP directory are text)
>
> To keep compatibility with existing installations, the presence of
> the 'binary' attribute is required for Net::LDAP::Search / Net::LDIF to
> behave the way described above.
>
> What do you think about it ?
>
> Or did I completely miss something very basic and all this stuff above is
> completely unnecessary ?
--
Peter Marschall
eMail: [EMAIL PROTECTED]
