This fixes a problem where some preferences had the wrong default value.
Also see bug 27472 where we made a similar fix for Android.

boklm authored Sep 17, 2018 and Georg Koppen committed Sep 18, 2018

On CentOS/RHEL 6, add the bundled directory from the firefox package to
the LD_LIBRARY_PATH.

In addition, we don't reset anymore the previous LD_LIBRARY_PATH value,
but only add new directories to it. This allows users to set LD_LIBRARY_PATH
before starting the browser if additional libraries are needed.

Version bumps

Kathleen Brade authored Mar 26, 2018 and Georg Koppen committed Sep 03, 2018

When doing Moat things, use a separate profile for the secondary browser
(instead of trying to use the main meek helper profile).

boklm authored Aug 17, 2018 and Georg Koppen committed Sep 03, 2018

Set mar_compression to xz for the alpha, and bzip2 for the stable.

Version bumps and Changelog update

Sukhbir Singh authored Aug 21, 2018 and Georg Koppen committed Aug 29, 2018

Since https://bugzilla.mozilla.org/show_bug.cgi?id=1463749 is fixed in
Firefox 60.2 ESR, we re-enable the verification of langpack checksums.

This new versions picks up a fix for #26548.

Add the option `mar_compression` to update_responses_config.yml to
select the compression format used in the mar files.

Currently we are using bzip2 for the previous alpha, xz for the new
alpha, and bzip2 for both the previous and new stable.

Fixed missing items in Changelog, correct sha256sum for new NoScript

Versions bump and Changelog update

boklm authored Aug 15, 2018 and Georg Koppen committed Aug 16, 2018

In a previous patch for bug 25485 we changed how we set LD_LIBRARY_PATH
in start-tor-browser. However this change is not taken into account when
the browser is restarted by the updater. To avoid this issue we replace
the firefox binary by a wrapper setting LD_LIBRARY_PATH.

boklm authored May 29, 2018 and Georg Koppen committed Aug 15, 2018

We still need the other half of the fix for actually providing higher
entropy (see bug 12968, comment:17ff. for discussion).