You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lenya.apache.org by Thorsten Scherler <th...@wyona.com> on 2006/08/10 10:54:34 UTC
[discuss] AC - Implementation, to branch or not to branch?
Hi all,
I am ATM working on a major change of the AC implementation. I got code
from Edith that I needed to update to current API use, implementing
default implementations and then integrating into the existing classes.
This was not that easy since the starting code was half finished. Some
parts of the integration are still very basic or hardcoded to get to the
point where I do not have any exception anymore. The idea of the changes
are to have a node containing the credentials for the document. This are
stored and defined in the meta data of the document. You can protected
each document different for each user/group/... if you want.
The code now compiles but the functionality is still far from being
usable. My problem is that I still need to understand the changes
(architecture and implementation) in the new code without having worked
excessively with the old code (which makes the task not easier).
This leads to the subject of this mail. I would love that other people
can have a look and review this implementation and maybe point out which
parts have sharp corners.
Since I know I would break the AC for a couple of days I think it is
best to create a branch in the sandbox and apply what I have there. If
we finish the code and are happy with the result then we can merge it
again in the trunk.
This means on the other hand all AC work should be conducted in the
branch otherwise we will run into problems when merging back.
If we agree that a branch makes sense I would create such a branch
(ac_node). You can then just svn switch to the branch to see/debug the
changes.
wdyt?
salu2
--
Thorsten Scherler
COO Spain
Wyona Inc. - Open Source Content Management - Apache Lenya
http://www.wyona.com http://lenya.apache.org
thorsten.scherler@wyona.com thorsten@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: [discuss] AC - Implementation, to branch or not to branch?
Posted by Thorsten Scherler <th...@wyona.com>.
Hi all,
I created a branch now and added a first working prototype.
http://svn.apache.org/viewvc?rev=432549&view=rev
"Basic refactored AC prototype. You can now GRANT or DENY credentials on
a url basis. This is integrated into the current implementation. We
extend the *.acml files with an @method attribute to define whether or
not a certain accreditable can have a role. Till now I did not
cleaned/verify that some method can be romved due to this changes. The
sitetree usecase AC... needs an update and the whole stuff more testing.
Please do in lenya-trunk/src: svn switch
https://svn.apache.org/repos/asf/lenya/sandbox/ac-restricted-1.4-src
if you are keen to test and play around. Try to change the *acml in the
build (e.g. with vim) and see the effect after reload."
I will now work on updating the AC usecases, but would greatly
appreciated feedback regarding design and behavior of this modified AC
management.
TIA
salu2
El jue, 10-08-2006 a las 10:54 +0200, Thorsten Scherler escribió:
> Hi all,
>
> I am ATM working on a major change of the AC implementation. I got code
> from Edith that I needed to update to current API use, implementing
> default implementations and then integrating into the existing classes.
>
> This was not that easy since the starting code was half finished. Some
> parts of the integration are still very basic or hardcoded to get to the
> point where I do not have any exception anymore. The idea of the changes
> are to have a node containing the credentials for the document. This are
> stored and defined in the meta data of the document. You can protected
> each document different for each user/group/... if you want.
>
> The code now compiles but the functionality is still far from being
> usable. My problem is that I still need to understand the changes
> (architecture and implementation) in the new code without having worked
> excessively with the old code (which makes the task not easier).
>
> This leads to the subject of this mail. I would love that other people
> can have a look and review this implementation and maybe point out which
> parts have sharp corners.
>
> Since I know I would break the AC for a couple of days I think it is
> best to create a branch in the sandbox and apply what I have there. If
> we finish the code and are happy with the result then we can merge it
> again in the trunk.
>
> This means on the other hand all AC work should be conducted in the
> branch otherwise we will run into problems when merging back.
>
> If we agree that a branch makes sense I would create such a branch
> (ac_node). You can then just svn switch to the branch to see/debug the
> changes.
>
> wdyt?
>
> salu2
--
Thorsten Scherler
COO Spain
Wyona Inc. - Open Source Content Management - Apache Lenya
http://www.wyona.com http://lenya.apache.org
thorsten.scherler@wyona.com thorsten@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: [discuss] AC - Implementation, to branch or not to branch?
Posted by Jörn Nettingsmeier <po...@uni-duisburg.de>.
Thorsten Scherler wrote:
> Hi all,
>
> I am ATM working on a major change of the AC implementation. I got code
> from Edith that I needed to update to current API use, implementing
> default implementations and then integrating into the existing classes.
>
> This was not that easy since the starting code was half finished. Some
> parts of the integration are still very basic or hardcoded to get to the
> point where I do not have any exception anymore. The idea of the changes
> are to have a node containing the credentials for the document. This are
> stored and defined in the meta data of the document. You can protected
> each document different for each user/group/... if you want.
>
> The code now compiles but the functionality is still far from being
> usable. My problem is that I still need to understand the changes
> (architecture and implementation) in the new code without having worked
> excessively with the old code (which makes the task not easier).
>
> This leads to the subject of this mail. I would love that other people
> can have a look and review this implementation and maybe point out which
> parts have sharp corners.
>
> Since I know I would break the AC for a couple of days I think it is
> best to create a branch in the sandbox and apply what I have there. If
> we finish the code and are happy with the result then we can merge it
> again in the trunk.
>
> This means on the other hand all AC work should be conducted in the
> branch otherwise we will run into problems when merging back.
>
> If we agree that a branch makes sense I would create such a branch
> (ac_node). You can then just svn switch to the branch to see/debug the
> changes.
please create a branch. and since this ac change has not been mentioned
in the 1.4 status discussion, i'd say it's post-1.4.0 stuff.
otoh, i do not know how the current ac code plays with andreas' uuid
stuff - i guess ac might be broken anyway, but i haven't got my test
setup to run yet.
if ac *is* currently broken, and if your changes have a reasonable
chance of fixing that and being usable for testers within a few days,
then (sigh) hack away...
regards,
jörn
--
"Open source takes the bullshit out of software."
- Charles Ferguson on TechnologyReview.com
--
Jörn Nettingsmeier, EDV-Administrator
Institut für Politikwissenschaft
Universität Duisburg-Essen, Standort Duisburg
Mail: pol-admin@uni-due.de, Telefon: 0203/379-2736
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: [discuss] AC - Implementation, to branch or not to branch?
Posted by Thorsten Scherler <th...@wyona.com>.
El jue, 10-08-2006 a las 12:17 +0200, Michael Wechner escribió:
> Thorsten Scherler wrote:
>
> >
> >This means on the other hand all AC work should be conducted in the
> >branch otherwise we will run into problems when merging back.
> >
> >If we agree that a branch makes sense I would create such a branch
> >(ac_node). You can then just svn switch to the branch to see/debug the
> >changes.
> >
> >wdyt?
> >
> >
>
> isn't that rather a case for the sandbox?
Yes
El jue, 10-08-2006 a las 10:54 +0200, Thorsten Scherler escribió:
> Since I know I would break the AC for a couple of days I think it is
> best to create a branch in the sandbox and apply what I have there. If
> we finish the code and are happy with the result then we can merge it
> again in the trunk.
;)
salu2
> Michi
>
> >salu2
> >
> >
>
--
Thorsten Scherler
COO Spain
Wyona Inc. - Open Source Content Management - Apache Lenya
http://www.wyona.com http://lenya.apache.org
thorsten.scherler@wyona.com thorsten@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org
Re: [discuss] AC - Implementation, to branch or not to branch?
Posted by Michael Wechner <mi...@wyona.com>.
Thorsten Scherler wrote:
>
>This means on the other hand all AC work should be conducted in the
>branch otherwise we will run into problems when merging back.
>
>If we agree that a branch makes sense I would create such a branch
>(ac_node). You can then just svn switch to the branch to see/debug the
>changes.
>
>wdyt?
>
>
isn't that rather a case for the sandbox?
Michi
>salu2
>
>
--
Michael Wechner
Wyona - Open Source Content Management - Apache Lenya
http://www.wyona.com http://lenya.apache.org
michael.wechner@wyona.com michi@apache.org
+41 44 272 91 61
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org