On Jan 22, 2008 7:43 PM, Vahid Moghaddasi <[EMAIL PROTECTED]> wrote: > # egrep "^bin:|^adm:|^daemon:|^root:|^rpm:" /etc/passwd > root:x:0:0:Linux root:/root:/bin/bash > bin:x:1:1:Linux bin:/bin:/sbin/nologin > daemon:x:2:2:Linux daemon:/sbin:/sbin/nologin > rpm:x:37:37:Linux :/var/lib/rpm:/sbin/nologin > > daemon is missing. > # grep daemon passwd.all > daemon:x:2:2:Linux daemon:/sbin:/sbin/nologin > daemon:x:1:1:System account #8545740:/: > but both get filtered out.
It looks as if those two both have UIDs and usernames that were in the list above. Doesn't that mean that they *should* be filtered out? > Why should be abything missing? I meant the code to keep the first > copy of the duplicate and throw any identical after that. For example, > if there are three "root" with UID "0" just keep the first and discard > the other two. So, you're throwing away those two because they're identical, in one parameter, to the ones you've seen. That's why. What about this data? fred:x:42:42:Fred Flintstone:/home/fred:/bin/bash barney:x:99:99:Barney Rubble:/home/barney:/bin/bash fred:x:69:69:Fred Astaire:/home/fred:/bin/bash dino:x:69:69:Dino:/home/dino:/bin/bash As I understand your code, it will drop Fred Astaire for having a username that had been seen before. Then, it will drop Dino, for having a UID that had been seen before, even though the line with that UID was itself dropped. Is that what you want it to do? Maybe you can invent a better solution. For example, instead of dropping users whose UID has been used elsewhere, perhaps they could be assigned new UIDs; I imagine that would make the users happier. Duplicate usernames aren't so simple as that, but aren't generally too difficult to handle programmatically. Cheers! --Tom Phoenix Stonehenge Perl Training -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] http://learn.perl.org/
