Torbutton INFO: tor SOCKS: http://f4amtbsowhix7rrf.onion/reports/r/tor-browser-2017-06-05/test_nightly-linux-x86_64 via --unknown--:b3bb40be24aec95a7d8c35608707fc28
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information
Child items 0
Show closed items
No child items are currently assigned. Use child items to break down this issue into smaller parts.
Linked items 0
Link issues together to show that they're related.
Learn more.
You were asking about new issues so persistently on tor-qa (as if you lack them :) that I couldn't pass by :) How many new issues will be enough for you? What are you going to do with all of them?
You were asking about new issues so persistently on tor-qa (as if you lack them :) that I couldn't pass by :) How many new issues will be enough for you? What are you going to do with all of them?
We'll try to get them fixed. Just let them coming.
Thanks for tracking this down. Giorgio: could you have a look at that one? I guess the intended behavior is: if we need to issue a request due to clicking on a javascript: link then it should adhere to our first-party isolation. That probably means NoScript itself should not issue that request as this is treated as a browser internal request which gets put onto the catch-all circuit (due to lack of URL bar domain information).
It's the "Attempt to fix Javascript links" (noscript.fixLinks in !about:config) feature.
Do you need me to modify the HEAD XHR preflight request in order to fit into your navigation restrictions, or to omit it outright? Or would it be better for you to just flip the preference?
It's the "Attempt to fix Javascript links" (noscript.fixLinks in !about:config) feature.
Do you need me to modify the HEAD XHR preflight request in order to fit into your navigation restrictions, or to omit it outright? Or would it be better for you to just flip the preference?
It seems to me flipping the preference might not be that bad. Richard: can you take a look whether that could be a viable way for solving our issue (without introducing other unintended ones)?
So the noscript.fixLinks will disable the custom onclick handler (which is what does the above described behaviour) but also disables a custom onchange handler (for select and option elements).
However, for Tor Browser that's a good thing, as it has a similar feature whereby it will automatically try to navigate to a selected option if it looks like a URL (the threshold for 'looks like a URL' is even lower though: value contains '/' or '.' and does not contain '@'). This URL will try to be navigated through the same code-path, so would have the same browser internal request gk mentioned.
Updating TorButton to set turn off the noscript.fixLinks option should work, will have a patch up in a bit.
Confirmed, noscript.fixLinks now set to 'false' by default in Linux rbm build, --unknown-- domain no longer sent when clicking links on provided page, and URL redirect vulnerability no longer occurs when clicking on javascript:XXX links.
Looks good. The needs_revision is only for making a proper git patch. http://tbaggery.com/2008/04/19/a-note-about-git-commit-messages.html is not unreasonable as a guideline and we try to follow Mozilla by starting the first line with "Bug XXXXX" ("XXXXX" being the bug number the commit relates to).
Trac: Status: needs_review to needs_revision Keywords: TorBrowserTeam201709R deleted, TorBrowserTeam201709 added