On 25.5.2016 18:17, Martin Basti wrote:
On 25.05.2016 16:07, Jan Cholasta wrote:
On 25.5.2016 15:03, David Kupka wrote:
On 18/05/16 08:01, Jan Cholasta wrote:
On 16.5.2016 16:35, Martin Basti wrote:
On 09.05.2016 14:07, Jan Cholasta wrote:
On 6.5.2016 14:32, Martin Basti wrote:
On 28.04.2016 14:45, Jan Cholasta wrote:
I have pushed my thin client WIP branch to GitHub:
All commits up to "ipalib: use relative imports for cross-plugin
imports" should be good for review. The rest is subject to change
(WARNING: I will force push into this branch).
I did not test it yet, I just checked the code
* automount: do not inherit automountlocation_tofiles from
* dns: move all dnsrecord code called on client to a single class *
* dns: do not rely on server data structures in code called on
you forgot to increment VERSION
This was deliberate, as it will no longer be necessary to bump
for backward compatible changes after this whole patchset is merged.
But we're not there yet, so fixed.
How we should handle VERSION after your patches?
* otptoken: fix import of DN *
* otptoken_yubikey: fix otptoken_add_yubikey arguments *
you forgot to increment VERSION
Did you find out why this was issue?
- del answer['value'] # Why does this cause an error if
- del answer['summary'] # Why does this cause an error if
The command definition was not complete, it was missing has_output.
* vault: move client-side code to the module level *
* vault: copy arguments of client commands from server
you forgot to increment Version
* ipalib: use relative imports for cross-plugin imports *
1) Missing explanation for future generations why this change is
in commit message
The other commits I will check later.
* frontend: remove the unused Command.soft_validate method
* frontend: perform argument value validation only on server
* frontend: do not forward argument defaults to server
I'm not a fan of returning None in promt_param function, but I havent
found anything better to use.
It always returned None for unset params.
* ipalib: optimize API.txt
this contains a lot of black magic, but because this is mainly copy of
current to proper places, LGTM
It's actually mostly cut-n-paste.
* makeaci: load additional plugins using API.add_module
Looks good, but I miss explanation why is this change needed
The next change would be impossible without this.
* plugable: replace API.import_plugins with new API.add_package
* ipalib, ipaserver: migrate all plugins to Registry-based
* ipalib, ipaserver: fix incorrect API.register calls in docstrings
* plugable: remove the unused deprecated API.register method
* plugable: switch API to Registry-based plugin discovery
* frontend: merge baseldap.CallbackRegistry into Command
*frontend: move the interactive_prompt callback type to Command
first batch of 30 patches from Honza's trac-4739 branch
(https://github.com/jcholast/freeipa/commits/trac-4739) is ready to be
pushed into master.
All upto "frontend: allow commands to have an argument named `name`" got
over numerous test&fix cycles in last week+ and is working as expected
Thanks for the review.
Honzo, please rebase and push them, thanks!
Attaching the patches for reference.
Pushed to master: 2b50fc617024f18a81e97f30f75ed1b9221c4055
Guys, you forgot to test it with newer pylint.
I don't see any "BuildRequires: newer pylint" in the spec file.
LGTM, although the commit message is wrong - this is not related to thin
client at all, PublicError and PublicMessage had .kw since forever.
Manage your subscription for the Freeipa-devel mailing list:
Contribute to FreeIPA: http://www.freeipa.org/page/Contribute/Code