Re: Script for mass updates

["Ulrich Windl" <Ulrich.Windl@rz.uni-regensburg.de>]

(forgot to reply to the list)
>>> Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de> schrieb am 06.02.2017 um
08:14
in Nachricht <58983099.ED38.00A1.0@rz.uni-regensburg.de>:
>>>> "Ralf Mattes" <rm@mh-freiburg.de> schrieb am 01.02.2017 um 17:23 in
Nachricht
> <45d3-58920b80-2b-2963e90@255443456>:
> 
> > Am Mittwoch, 01. Februar 2017 16:52 CET, Jephte Clain 
> > <jephte.clain@univ-reunion.fr> schrieb:
> > 
> >> using michaël's filter, you could try this:
> >>
> >> ldapsearch [options]
'(&(ou:dn:=people)(!(ou=system))(!(displayName=*)))'
> >> cn | awk '
> >> /^dn:/ {
> >>   print
> >>   print "changetype: modify"
> >>   print "replace: displayName"
> >>   next
> >> }
> >> /^cn:/ {
> >>   sub(/^cn/, "displayName")
> >> }
> >> { print }
> >> ' | ldapmodify [options]
> >>
> >> we aren't doing your homework, are we? :-)
> > 
> > Danger, Will Robinson!
> > This will only work for dn values that aren't encoded.
> > That's a trivial job for perl or python (or whatever). AWK operates
> > on character streams and that's a bad fit for LDIF.
> 
> Zwo questions:
> 
> 1: Doesn't /^dn:/ also match "dn::" (likewise /^cn:/ and "cn::"?
> 2: If /^cn:/ matches "cn::", wouldn't the sub() replace "cn::" with 
> "displayName::"?
> 
> So everything would still work IMHO.
> 
> We don't use displayName like that; we only us it if someone with an 
> academic title insists on having his "Prof." or "Dr." (or any collection of

> that in his/her name).
> 
> Regards,
> Ulrich
> 
> Thought of the day: Trump vs. America; who will win?
> 
> > 
> >  Cheers, Ralf Mattes
> > 
> >> regards,
> >> Jephté
> >>
> 
> 
> 
>