| Commit message (Collapse) | Author | Age |
|---|
| ... | |
| | |
| |
| |
| | |
Changelog update and versions bump
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
loaded in ESR 60
The infrastructure for loading extension-overrides.js no longer
exists in ESR60, so these prefs had to be moved.
Unfortunately, we can't just dump the contents of extension-overrides.js
into 000-tor-browser.js. The settings are actually partially built in
tor-browser-build to conditionally include various bridge strings as well
as localization settings. So, we piggy back off this existing build
process and just append the contents of the partially generated
extension-overrids.js to 000-tor-browser.js.
|
| |\ \
| |/
|/| |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| |/
|
|
|
|
| |
Starting with content sandboxing being enabled our seatbelt profiles
have been broken (see: #22000). We should remove them for now to avoid
a broken experience in the alphas.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit adds support for building the 64-bit version of NSIS, and
also bumps the version to 3.03. Doing this enables us to build MAR files
in a 64-bit container for the 64-bit version of Tor Browser; see bug
26363 and bug 24477.
The pe_checksum_fix.py doesn't work in a 64-bit container with the
bundled python-pefile version so we build its latest version to fix this
issue. This change is borrowed from commit bb32ec91b and updates
python-pefile to 2017.11.5.
The Debian package and the patches are no longer required as all changes
were merged upstream in 3.01-1. (See the nsis changelog in Debian.)
|
| |
|
|
|
|
|
| |
Mozilla switched from Bzip2 to LZMA as compression algorithm for MAR
files during th Firefox 56. That means esr52-based Tor Browsers don't
understand LZMA yet. Thus, we still use Bzip2 for the first esr60-based
MAR files and switch afterwards to LZMA.
|
| | |
|
| |\ |
|
| | |
| |
| |
| |
| |
| | |
Bumping Rust to 1.26.1 avoids intermittent Firefox compilation failures
we have seen, similar to those Mozilla got and documented in
https://bugzilla.mozilla.org/show_bug.cgi?id=1445413.
|
| | |
| |
| |
| |
| | |
sandboxed-tor-browser is not maintained anymore and Yawning requested to
not ship it anymore in our alphas once we switch to Firefox 60 ESR.
|
| |\ \ |
|
| | |/
| |
| |
| |
| |
| | |
The fix for https://bugzilla.mozilla.org/show_bug.cgi?id=1443823 makes
our workaround for https://bugzilla.mozilla.org/show_bug.cgi?id=1392604
obsolete, and we can remove it as it landed on esr60.
|
| |/ |
|
| | |
|
| |
|
|
| |
Also enable rust for all builds.
|
| |
|
|
|
| |
This commit adds support for building and bundling fxc2 which is
required for Windows cross-compilation on Linux.
|
| | |
|
| | |
|
| | |
|
| |\ |
|
| | |
| |
| |
| |
| | |
We were setting the minSupportedInstructionSet attribute for the win32
updates only. We now set it on all platforms.
|
| | | |
|
| |/
|
|
|
| |
Add the htaccess_rewrite_rules option which can be used to specify
rewrite rules to be added to the generated .htaccess files.
|
| | |
|
| | |
|
| |
|
|
|
|
| |
The 32bit toolchain is not able to link libxul anymore with selfrando
enabled as the linker is running out of address space. We therefore
disable selfrando for 32bit for now.
|
| |\ |
|
| | |
| |
| |
| |
| | |
At the same time we add a var/selfrando variable to make it easier to
enable/disable selfrando.
|
| |\ \ |
|
| | | |
| | |
| | |
| | | |
This commit fixes #26153 by bumping the selfrando version to 0.4.2
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
After building Firefox we now get a 'Tor Browser.app' instead of a
'TorBrowser.app'. This patch makes sure the additional whitespace in the
app name is correctly handled by the build script and the one that deals
with packaging the final bundle.
We need to ship a fix for the Firefox packaging step as well as
|./mach package| wants to build the final .dmg in that step, too, which
breaks: there is no .dmg creation tool available. Setting
`INNER_MAKE_PACKAGE` to `true` does not seem to work anymore. That part
of this patch is currently only a workaround to get the nightly builds
going. We should come up with a better solution that allows us to skip
that part of the Firefox packaging step on all supported platforms.
|
| | |
| |
| |
| |
| |
| | |
The snowflake build system is pretty complicated. We disable this pluggable
transport for now to get time to adapt it to the new macOS toolchain while
not blocking our nightly builds. See bug 26251 for work on that.
|
| | | |
|
| | | |
|
| | | |
|
| |\ \ |
|
| | | | |
|
| | | | |
|
| |\ \ \ |
|
| | |/ / |
|
| |\ \ \ |
|
| | | | | |
|
| |/ / /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Avoid duplicating Windows CFLAGS and LDFLAGS. Instead, use the flags
defined in rbm.conf.
This also means we are now using -fstack-protector-strong instead of
-fstack-protector-all for building openssl.
We update the CFLAGS/LDFLAGS defined in rbm.conf to move
var/flag_mwindows at the end, because openssl's Configure script does
not like having an argument starting with a space.
|
| |\ \ \ |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
CFLAGS and LDFLAGS were unused. We are now defining them in LOC, using
the definition of CFLAGS and LDFLAGS from rbm.conf.
We also don't hardcode anymore -j4.
|
