Re: [Rails] Re: POST-only logic in protect_from_forgery considered harmful?

2010-04-08 Thread Josh _
On Thu, Apr 1, 2010 at 7:20 AM, Jeff Lewis wrote: > This seems like a non-issue to me that can and should be handled by > the developer of the app, regardless of what lang/framework you're > using, by following basic best-practices for securing your app against > csrf or sql-injection or ... atta

[Rails] Re: POST-only logic in protect_from_forgery considered harmful?

2010-04-01 Thread Jeff Lewis
This seems like a non-issue to me that can and should be handled by the developer of the app, regardless of what lang/framework you're using, by following basic best-practices for securing your app against csrf or sql-injection or ... attack. So in your post example, if you didn't want to restrict

Re: [Rails] Re: POST-only logic in protect_from_forgery considered harmful?

2010-04-01 Thread Josh _
On Thu, Apr 1, 2010 at 1:08 AM, Frederick Cheung wrote: > Without getting into the debate about how idempotent GET requests > really are I'd suspect that these days most people are using restful > routes. If you use restful routes and remove the default route then > it's not possible invoke (eg)

[Rails] Re: POST-only logic in protect_from_forgery considered harmful?

2010-04-01 Thread Frederick Cheung
On Apr 1, 1:52 am, JSW wrote: > How many rails developers do we think put a POST-method validation > filter around all their form processing code, and yet expect > protect_from_forgery stills somehow protects the actions? > Without getting into the debate about how idempotent GET requests reall