On 08/13/2010 04:24 PM, Adam Young wrote:
On 08/12/2010 09:35 AM, Pavel Zůna wrote:
On 2010-08-12 14:38, Rob Crittenden wrote:
Pavel Zůna wrote:
On 2010-08-12 04:46, Rob Crittenden wrote:
Pavel Zůna wrote:
setattr and addattr can now be used both to set all values of
ANY attribute. the last setattr always resets the attribute to
the specified value and all addattr append to it.

user-mod testuser --setattr=title=msc
title: msc
user-mod testuser --setattr=title=msb
title: msb
user-mod testuser --addattr=title=msc
title: msb, msc
user-mod testuser --setattr=title=
user-mod testuser --setattr=title=msc --addattr=msb
title: msc, msb
user-mod testuser --setattr=title=ing --addattr=bc
title: ing, bc
user-mod testuser --setattr=title=doc
title: doc

It's not very user friendly, but it's going to be used very very
rarely in special conditions in the CLI and we can use it to save
lots of JSON-RPC roundtrips in the webUI.


It was my intention when I added addattr and setattr that one couldn't set already-defined params this way. They were silently ignored. So you
couldn't do:

user-mod testuser --setattr=givenname=Jeff

This would be possible with this patch. Was that intentional?

BTW I have the start of a test suite for this functionality.


Yes, it is intentional. I forgot to mention it in the description. I'm
using setattr/addattr for everything in the webUI - it makes the code a
lot simpler.

Doesn't that invalidate all the validators we have in the plugins? This
is why I disallowed it.


It does, but I see these options as something only experienced users, who need to set something we don't support directly, will use. Sometimes they might want to disable the validators, if they know what they're doing. We could also make the setattr/addattr handler in frontend.py detect if a there's a validator available and use it.

Validators in the webUI is still something we need to figure out. Adam was proposing having validators in the form of regex strings, which is not a bad idea as it's easy to implement on any platform/language. On the other hand, I don't know if it's good enough for all parameters we have.

Hmm. There's a lot to think about here actually. I'll make it my homework for the weekend. :)


Freeipa-devel mailing list

Here's my version, that just calls the parameter prior to updating the attr. I tested it with:

[r...@ipa ~]# ipa user-mod --setattr uidnumber=555 kfrog
Modified user "kfrog"
  User login: kfrog
  First name: Kermit
  Last name: Frog
  Home directory: /home/kfrog
  Login shell: /bin/sh
  UID: 555
  Groups: ipausers
[r...@ipa ~]# ipa user-mod --setattr uidnumber=frog kfrog
ipa: ERROR: invalid 'uidnumber': must be an integer

Freeipa-devel mailing list
Pushed to Master. ACKed in IRC by rcrit, and based on a pzuna patch reviewd by both myself and rcrit

Since this was a diff to a Patch ACK
Freeipa-devel mailing list

Reply via email to