[tbb-bugs] #26401 [Applications/Tor Browser]: Rebase Orfox patches onto Tor Browser 8.0 for TBA

Tor Bug Tracker & Wiki blackhole at torproject.org
Thu Jul 26 16:24:39 UTC 2018

#26401: Rebase Orfox patches onto Tor Browser 8.0 for TBA
 Reporter:  sysrqb                            |          Owner:  tbb-team
     Type:  task                              |         Status:
                                              |  needs_revision
 Priority:  Very High                         |      Milestone:
Component:  Applications/Tor Browser          |        Version:
 Severity:  Normal                            |     Resolution:
 Keywords:  tbb-mobile, TorBrowserTeam201807  |  Actual Points:
Parent ID:  #26531                            |         Points:
 Reviewer:                                    |        Sponsor:

Comment (by sysrqb):

 Replying to [comment:36 gk]:
 > Okay, we are getting close, just the following comments/remarks are
 > commit 0b932f895e3b2866161048a6007e61e257dc395b -- not okay
 > There is no need for a `--enable-install-strip` as we already strip
 everything with `--enable-strip`.


 > commit 408e63083f15d59c98fee619a277dfe123429d8c -- not okay
 > we should follow the Android version as used in the resist
 fingerprinting mode, see:
 > https://dxr.mozilla.org/mozilla-
 > I.e. "Android 4.1.0" > "Android 6.0" which is way more popuplar than
 4.1.0 as well.

 Hrm. Yes, I see. I did consider using a higher Android version, but I
 thought it would be better if we used the lowest supported version number
 instead of the most popular. In any case, I don't feel strongly about
 this. I'll change the hard-coded UAS so it matches the RFP string.

 Also, https://dxr.mozilla.org/mozilla-

 > commit 7bc689d797439d6394853e421f71eda5d7d1631a -- not okay
 > +import android.content.IntentFilter; <- seems to be something not used
 > (probably part of the NetCipher upgrade on the previous branch?)

 Ugh. That was added in the wrong commit. The `import` should be in
 694067738b04b6a670d072c64a7afa4be63581b1 ("Orfox: add
 BroadcastReceiver..."). I'll move it.

 > `checkStartOrbot()` came before `processTabQueue()` before (and the if-
 clause around `!mHasResumed` got deleted). Doesn't the order matter here?
 (I don't know just for double-checking)

 Yeah, this was something I looked at while I was rebasing the commit.
 Calling `checkStartOrbot()` was moved as a fix for a crash

 I think the `!mHasResumed` clause was deleted in Orfox as a way of
 preventing the app from bringing itself to the foreground the next time it
 was paused (put into the background). After it was deleted, `mHasResumed`
 would never be set to true in `onResume()`, so when `onPause()` is called
 the `mHasResumed` conditional isn't executed and `Prompt:ShowTop` is never
 registered. I don't know why this was missed during the rebase - I
 might've been confused while resolving a merge conflict. I'll include this

Ticket URL: <https://trac.torproject.org/projects/tor/ticket/26401#comment:37>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online

More information about the tbb-bugs mailing list