- From: John Wong <gokoproject@gmail.com>
- Date: Fri, 17 Jan 2014 18:15:11 -0500
- To: "public-webappsec@w3.org" <public-webappsec@w3.org>
- Message-ID: <CACCLA55p=nYSFXqguBpwV_k_ZTfYvdr4aRgaASLo4e+Ny0_7ZQ@mail.gmail.com>
I happen to have this discussion again on #security with the Mozillians today. Issue: the 1.0 spec does not seem to specify whether directive names should be case insensitive or not (a quick glance on 1.1 draft also seems to be case as well). Correct me if I am wrong. Someone told me that since the grammar follows the ABNF, the following is implicit in CSP spec: ABNF strings are case insensitive and the character set for these strings is US-ASCII. http://www.ietf.org/rfc/rfc5234.txt It is worthwhile that we determine whether CSP directives should be case insensitive or not and write that into the spec explicitly. For Firefox's bug, please see https://bugzilla.mozilla.org/show_bug.cgi?id=938652 Thanks. Yeuk Hon On Mon, Oct 28, 2013 at 3:29 PM, John Wong <gokoproject@gmail.com> wrote: > Hi, > > The UA algorithm states we can take source expressions case insensitive. > What about directive names themselves? For example, 'self' and 'SELF' are > acceptable. [1] > > > If the source expression is a case insensitive match for 'self'(including the quotation marks) > > [1]: http://www.w3.org/TR/CSP/#parsing-1 > > Thanks. > > Yeuk Hon >
Received on Friday, 17 January 2014 23:15:38 UTC