rexexp across a ldap ldif file, collecting cn: guid: memberUID:

khippy has asked for the wisdom of the Perl Monks concerning the following question:

Hi perlmongers,

my current problem is the following:

I have a ldif input file from an ldap database, which I want to get the groupnames, guid and members from.

The structure is like this:

<some irrelevant data or the same like downwards>

dn: cn=groupname1,dc=domain,dc=com
cn: groupname1
gidNumber: 122
memberUid: member1
memberUid: member2
memberUid: member3
memberUid: member4
memberUid: member6
memberUid: member7
userPassword:: Kg==
objectClass: top
objectClass: posixGroup
creatorsName: uid=cyrus,dc=domain,dc=com
createTimestamp: 20040408083004Z
modifiersName: uid=cyrus,dc=domain,dc=com
modifyTimestamp: 20040408083004Z

<more irrelevant data or the same like above>

dn: cn=groupname2,dc=domain,dc=com
cn: groupname2
gidNumber: 113
userPassword:: Kg==
objectClass: top
objectClass: posixGroup
creatorsName: uid=cyrus,dc=domain,dc=com
createTimestamp: 20031208140152Z
memberUid: member1
memberUid: member2
memberUid: member3
memberUid: member4
memberUid: member6
memberUid: member7
description: some irrelevant description
modifiersName: uid=cyrus,dc=domain,dc=com
modifyTimestamp: 20040404112251Z

<more irrelevant data or the same like above>
[download]

Each group of information is separated from the others with an empty line as you can see.

Now I want to regexp across the file and collect from each group of information starting with

dn: cn=<groupname>,dc=domain,dc=com
[download]

the following data:

cn: <groupname><br>
gidNumber: <guid><br>
memberUid: <membername>
[download]

The aim is to collect this data to use for another cyrus imap mailserver getting the groups information from an mysql database.

How can I do this?

--

there are no silly questions

killerhippy

Comment on rexexp across a ldap ldif file, collecting cn: guid: memberUID: Select or Download Code

Replies are listed 'Best First'.
Re: rexexp across a ldap ldif file, collecting cn: guid: memberUID: by bart (Canon) on Aug 21, 2004 at 20:44 UTC
Like I said in the CB: read the file in paragraph mode and extract the sections with `/^PATTERN/m`. Example code: `#! perl -w $/ = ""; # paragraph mode while(<DATA>) { if(/^dn:/) { # skip irrelevant paragraphs my($cn) = /^cn:\s(.)/m; my($gid) = /^gidNumber:\s(\d+)/m; my(@member) = /^memberUid:\s(.+)/mg; print "cn: $cn\ngid: $gid\nmembers: @member\n\n"; } } __DATA__ * Your sample data follows *` [download] The printout I get is: `cn: groupname1 gid: 122 members: member1 member2 member3 member4 member6 member7 cn: groupname2 gid: 113 members: member1 member2 member3 member4 member6 member7` [download]	[reply] [d/l] [select]
Re^2: rexexp across a ldap ldif file, collecting cn: guid: memberUID: by khippy (Scribe) on Aug 21, 2004 at 21:00 UTC
Thank you for your code and hint. Updated, see downwards I tried the code as is, and I get a lot of `Use of uninitialized value in concatenation (.) or string at ./data li +ne 9, <DATA> chunk <number of data line>.` [download] and a lot of empty hits `cn: gid: members:` [download] mixed up data `cn: username gid: 102 members:` [download] but also hits, which are right. This might be lead from dn: being inside irrelevant data blocks. How to face a block, and only check it out, if it starts with "dn: cn" at the very first line? This would disregard irrelevant data blocks. Update: `if(/^dn: cn/) { # skip irrelevant paragraphs` [download] does exactly what I need. Paragraph mode is the one! supercool -- there are no silly questions killerhippy	[reply] [d/l] [select]