It seems to me that allowing this to be modified is fairly harmless. Would you expect it to fallback to __acl__ if it doesn't find the custom method/attribute? Or would you expect it to be changed wholesale? The issue in changing it wholesale and still calling it the ACLAuthorizationPolicy is that it may be difficult to use with any third party code but that's kind of an issue with any code written to a particular policy.
On Fri, Jul 17, 2015 at 10:19 AM, Iain Duncan <iainduncanli...@gmail.com> wrote: > This sounds like it would actually be helpful for me too for the project > we're about to embark on, if I'm understanding the OP correctly. > > my two cents Canadian! > iain > > On Thu, Jun 11, 2015 at 10:04 AM, Chris McDonough <chr...@plope.com> wrote: >> >> Id like to hear other people speak up who have needed the same knob. If >> this is something a few people have needed, I'd ask them if their life would >> be much better with a knob on aclauthorizationpolicy instead of a custom >> policy given that they would also need to document it and justify it in >> the.pyramid docs. >> >> On June 11, 2015 11:28:25 AM EDT, Christian Theune >> <christian.the...@googlemail.com> wrote: >>> >>> Hi, >>> >>> I am implementing an application that is structured in way where I have a >>> business model and multiple Pyramid applications on top of that, ending up >>> with 4 processes, two of them providing APIs (XML-RPC) and two of them >>> providing UIs. >>> >>> I'm happy using the ACLAuthenticationPolicy but I noticed that my __acl__ >>> methods are getting convoluted as they have to know about the various >>> processes. I wanted to split this up and wondered why the >>> ACLAuthorizationPolicy can't be customized to use different attributes for >>> the lookup. >>> >>> I'm making separate instances of the ACLAuthorizationPolicy in those >>> applications and thus I started out to customize it, but had to copy the >>> whole class. The ACL system itself seems fine for me, but putting the >>> variations of ACLs that I'm dealing with into the same __acl__ is ... well >>> ... convoluted. >>> >>> Chris pointed out that "adding knobs" isn't that fashionable - but I >>> AFAICT I'm happy with the ACLAuthorizationPolicy "except this one bit". I >>> know this isn't necessarily a strong argument but maybe somone can tell me >>> whether I'm doing something wrong or maybe help me to work towards a "no >>> knobs needed" solution. :) >>> >>> Cheers, >>> Theuni >> >> >> -- >> Sent from my Android device with K-9 Mail. Please excuse my brevity. >> >> -- >> You received this message because you are subscribed to the Google Groups >> "pylons-devel" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to pylons-devel+unsubscr...@googlegroups.com. >> To post to this group, send email to pylons-devel@googlegroups.com. >> Visit this group at http://groups.google.com/group/pylons-devel. >> For more options, visit https://groups.google.com/d/optout. > > > -- > You received this message because you are subscribed to the Google Groups > "pylons-devel" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to pylons-devel+unsubscr...@googlegroups.com. > To post to this group, send email to pylons-devel@googlegroups.com. > Visit this group at http://groups.google.com/group/pylons-devel. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "pylons-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to pylons-devel+unsubscr...@googlegroups.com. To post to this group, send email to pylons-devel@googlegroups.com. Visit this group at http://groups.google.com/group/pylons-devel. For more options, visit https://groups.google.com/d/optout.
