Re: [Lift] Lift security vulnerability

2010-02-03 Thread Jeppe Nejsum Madsen
David Pollak feeder.of.the.be...@gmail.com writes: I'd like to get a sense of how important the community views this defect. Is it a backport the fix to every milestone and release yesterday or is it a fix it in 2.0-M2 or someplace in between. For me, it's fix it in 2.0-SNAPSHOT /Jeppe --

Re: [Lift] Lift security vulnerability

2010-02-03 Thread Timothy Perrett
+1 Fix it in head, no need to back-port; M2 is only around the corner. Cheers, Tim On 3 Feb 2010, at 09:49, Jeppe Nejsum Madsen wrote: David Pollak feeder.of.the.be...@gmail.com writes: I'd like to get a sense of how important the community views this defect. Is it a backport the fix to

Re: [Lift] Lift security vulnerability

2010-02-03 Thread Indrajit Raychaudhuri
1. Fix in head/master (2.0-SNAPSHOT) and prepone 2.0-M2. 2. Backport in 1.0.x branch and spin 1.0.4. We haven't marked 1.0.x 'unsupported' yet. Forcing apps to move to 2.0-M2 just for this vulnerability fix isn't fun. Cheers, Indrajit On 03/02/10 3:34 PM, Timothy Perrett wrote: +1 Fix it

Re: [Lift] Lift security vulnerability

2010-02-03 Thread Feng Zhang
I found that in the fix, \n is changed to \t, while \t to \n. Is this desired behavior? Thank you, Feng On Wed, Feb 3, 2010 at 9:20 AM, Indrajit Raychaudhuri indraj...@gmail.comwrote: 1. Fix in head/master (2.0-SNAPSHOT) and prepone 2.0-M2. 2. Backport in 1.0.x branch and spin 1.0.4. We

Re: [Lift] Lift security vulnerability

2010-02-03 Thread David Pollak
Thanks for pointing that out. There are other problems as well... I'll fix them (in both the Scala and Lift diffs) On Wed, Feb 3, 2010 at 7:39 AM, Feng Zhang sharpzh...@gmail.com wrote: I found that in the fix, \n is changed to \t, while \t to \n. Is this desired behavior? Thank you, Feng

[Lift] Lift security vulnerability

2010-02-02 Thread David Pollak
Folks, Turns out there's a security vulnerability in Lift. It's possible to insert control characters into input fields. When the control characters are sent back to the browser, the browser will choke. An example can be seen at http://demo.liftweb.net Go to that page, enter your name in the

Re: [Lift] Lift security vulnerability

2010-02-02 Thread Naftoli Gugenheim
If you scan the whole page wouldn't it affect performance? Or will you put a safeguard in the input field / processing query parameters? 2010/2/2 Naftoli Gugenheim naftoli...@gmail.com: Is that not a defect of the browsers? On Tue, Feb 2, 2010 at 7:57 PM, David Pollak

Re: [Lift] Lift security vulnerability

2010-02-02 Thread David Pollak
radically. Connected by MOTOBLURâ„¢ on T-Mobile -Original message- From: Naftoli Gugenheim naftoli...@gmail.com To: liftweb liftweb@googlegroups.com Sent: Wed, Feb 3, 2010 01:31:24 GMT+00:00 Subject: Re: [Lift] Lift security vulnerability If you scan the whole page wouldn't it affect

Re: [Lift] Lift security vulnerability

2010-02-02 Thread Naftoli Gugenheim
:24 GMT+00:00* Subject: *Re: [Lift] Lift security vulnerability If you scan the whole page wouldn't it affect performance? Or will you put a safeguard in the input field / processing query parameters? 2010/2/2 Naftoli Gugenheim : Is that not a defect of the browsers? On Tue, Feb 2, 2010

Re: [Lift] Lift security vulnerability

2010-02-02 Thread David Pollak
, 2010 01:31:24 GMT+00:00* Subject: *Re: [Lift] Lift security vulnerability If you scan the whole page wouldn't it affect performance? Or will you put a safeguard in the input field / processing query parameters? 2010/2/2 Naftoli Gugenheim : Is that not a defect of the browsers? On Tue, Feb