Arch Linux

lucas7_

Member

Registered: 2014-10-21

Posts: 10

Re: Inox Browser - yet another Chromium Privacy Spin-off

any news on the 32bit binary?

gcarq

Member

From: Austria

Registered: 2013-10-07

Posts: 40

Re: Inox Browser - yet another Chromium Privacy Spin-off

Updated to 44.0.2403.89. I had to update some patches because the code changed slightly, but nothing new so far.

Maybe change google's source code to completely remove all its abilities to identify the referer

But then you won't be able to post here on this forum At least I can't in firefox with network.http.sendRefererHeader set to 0.

Yes unfortunatly disabling the referer would breaking the user experience because many frameworks rely on it for navigation.

@lucas7_
Sorry about the 32bit binaries. its quite complicated to build chromium in a x86-chroot, because it seems many build scripts are using the arch from the running kernel.

cheers

sekret

Member

Registered: 2013-07-22

Posts: 289

Re: Inox Browser - yet another Chromium Privacy Spin-off

I'd love to see some more search engines added. DDG is fine, but I usually use startpage.com. Maybe we could collect for a reasonable list of say 3-5 added search engines?

sekret

Member

Registered: 2013-07-22

Posts: 289

Re: Inox Browser - yet another Chromium Privacy Spin-off

I was thinking ... Those patches should go bigger. E.g. if the f-droid people provide a chromium build one day, I'd really recommend they used your patches! So will you release those patches somewhere outside the AUR for other projects to be able to easily integrate them?

graysky

Wiki Maintainer

From: :wq

Registered: 2008-12-01

Posts: 10,643

Website

Re: Inox Browser - yet another Chromium Privacy Spin-off

Firstly, this seems like a nice project so thank you for the efforts to create it.  I posted a few questions to the AUR, but probably better to deal with here:

1) In 'local _chromium_conf=' your PKGBUILD has the following set which differs from the ARCH upstream. Is it needed for the patches?

-Duse_system_ssl=0

2) If you are disabling speech_input, why do you retain the dep of 'speech-dispatcher' and the following in your 'locaL-chromium_conf='?

-Dlinux_link_libspeechd=0
-Dlibspeechd_h_prefix=speech-dispatcher/

3) Why do you have the versioned dependency of 'python2-ply<3.5'?  44.0.2403.125 builds fine without it (currently 3.6-1 is in [extra]).

I was thinking ... Those patches should go bigger. E.g. if the f-droid people provide a chromium build one day, I'd really recommend they used your patches! So will you release those patches somewhere outside the AUR for other projects to be able to easily integrate them?

This is a good suggestion.  Github is a nice place for you to host files like patches.  For example: https://github.com/graysky2/kernel_gcc_patch

Last edited by graysky (2015-07-29 07:05:56)

---

CPU-optimized Linux-ck packages @ Repo-ck  • AUR packages • Zsh and other configs

gcarq

Member

From: Austria

Registered: 2013-10-07

Posts: 40

Re: Inox Browser - yet another Chromium Privacy Spin-off

In 'local _chromium_conf=' your PKGBUILD has the following set which differs from the ARCH upstream. Is it needed for the patches?

-Duse_system_ssl=0

No this is not needed, I experimented with the source code and got an error at some point and forgot to change this back.

If you are disabling speech_input, why do you retain the dep of 'speech-dispatcher' and the following in your 'locaL-chromium_conf='?

-Dlinux_link_libspeechd=0
-Dlibspeechd_h_prefix=speech-dispatcher/

I tried to remove this along with -Denable_web_speech=1, but somehow it ended in an compilation error. Did you compile it without speed-dispatcher?

Why do you have the versioned dependency of 'python2-ply<3.5'?  44.0.2403.125 builds fine without it (currently 3.6-1 is in [extra]).

The last time i checked the python2-ply project on github they fixed this issue in trunk, so I waited for a new ply version and didn't test 3.6-1 against chromium 44.0.2403.125.
Thanks for the input. I created an github project https://github.com/gcarq/inox-patchset and I will update the PKGBUILD later this week.

graysky

Wiki Maintainer

From: :wq

Registered: 2008-12-01

Posts: 10,643

Website

Re: Inox Browser - yet another Chromium Privacy Spin-off

@qcarq - Yes, built successfully in a clean chroot with multilib enabled:

--- a/PKGBUILD	2015-07-23 15:24:06.000000000 -0400
+++ b/PKGBUILD	2015-07-29 15:38:58.744479505 -0400
@@ -5,7 +5,7 @@
 # Contributor: Daniel J Griffiths <ghost1227@archlinux.us>
 
 pkgname=inox
-pkgver=44.0.2403.89
+pkgver=44.0.2403.125
 pkgrel=1
 _launcher_ver=2
 pkgdesc="Chromium Spin-off to enhance privacy by disabling data transmission to Google"
@@ -14,10 +14,10 @@ url="http://www.chromium.org/"
 license=('BSD')
 depends=('gtk2' 'nss' 'alsa-lib' 'xdg-utils' 'bzip2' 'libevent' 'libxss' 'icu'
          'libexif' 'libgcrypt' 'ttf-font' 'systemd' 'dbus' 'flac' 'snappy'
-         'speech-dispatcher' 'pciutils' 'libpulse' 'harfbuzz' 'libsecret'
+         'pciutils' 'libpulse' 'harfbuzz' 'libsecret'
          'libvpx' 'perl' 'perl-file-basedir' 'desktop-file-utils'
          'hicolor-icon-theme')
-makedepends=('python2' 'python2-ply<3.5' 'gperf' 'yasm' 'mesa' 'ninja')
+makedepends=('python2' 'gperf' 'yasm' 'mesa' 'ninja')
 makedepends_x86_64=('lib32-gcc-libs' 'lib32-zlib')
 optdepends=('kdebase-kdialog: needed for file dialogs in KDE'
             'gnome-keyring: for storing passwords in GNOME keyring'
@@ -43,7 +43,7 @@ source=(https://commondatastorage.google
         launcher-branding.patch
         disable-missing-key-warning.patch)
         
-sha256sums=('badd9aa979487b55bbcce4f54ec16926c6fa7eb46ec2536866200a5e68853a98'
+sha256sums=('78568ec613443b5b7be83216f2c86cb4e64e7398eee9bbc0a35d1c0cef402452'
             '7f91c81721092d707d7b94e6555a48bc7fd0bc0e1174df4649bdcd745930e52f'
             'ff3f939a8757f482c1c5ba35c2c0f01ee80e2a2273c16238370081564350b148'
             '46daf921ed7eaab175f5a86b09357c69c33a10ffe1d4e7c24476af510c1b28d0'
@@ -140,7 +140,6 @@ build() {
     -Dpython_ver=2.7
     -Dlinux_link_gsettings=1
     -Dlinux_link_libpci=1
-    -Dlinux_link_libspeechd=0
     -Dlinux_link_pulseaudio=1
     -Dlinux_strip_binary=1
     -Dlinux_use_bundled_binutils=0
@@ -149,7 +148,6 @@ build() {
     -Dicu_use_data_file_flag=0
     -Dlogging_like_official_build=1
     -Drelease_extra_cflags="$CFLAGS"
-    -Dlibspeechd_h_prefix=speech-dispatcher/
     -Dffmpeg_branding=Chrome
     -Dproprietary_codecs=1
     -Duse_gnome_keyring=0
@@ -164,7 +162,6 @@ build() {
     -Duse_system_libvpx=1
     -Duse_system_libxml=0
     -Duse_system_snappy=1
-    -Duse_system_ssl=0
     -Duse_system_xdg_utils=1
     -Duse_system_yasm=1
     -Duse_system_zlib=0

Last edited by graysky (2015-07-29 20:42:40)

---

CPU-optimized Linux-ck packages @ Repo-ck  • AUR packages • Zsh and other configs

sekret

Member

Registered: 2013-07-22

Posts: 289

Re: Inox Browser - yet another Chromium Privacy Spin-off

Really great seeing you guys working together!

BTW, I'm completely switched from firefox using inox + https everywhere + umatrix. Closest to perfect!

freddy88

Member

Registered: 2015-08-05

Posts: 1

Re: Inox Browser - yet another Chromium Privacy Spin-off

Hi,
could you add the patch to enable hardware acceleration through Vaapi ?

I built successfuly inox 44.0.2403.89 with this patch http://www.slackware.com/~alien/slackbu … aapi.patch

I know this is not the main reason to be of inox but no hardware video decode is a problem for most of us, right ?

Perfect Gentleman

Member

Registered: 2012-10-25

Posts: 222

Re: Inox Browser - yet another Chromium Privacy Spin-off

i've applied inox-patches, except branding patches, and vaapi patch to 44.0.2043.157.

Last edited by Perfect Gentleman (2015-08-25 15:55:32)

sekret

Member

Registered: 2013-07-22

Posts: 289

Re: Inox Browser - yet another Chromium Privacy Spin-off

Is this project already dead again? Because inox(-bin) on the AUR don't follow chromium's pkgver. The patches still apply for pkgver=44.0.2403.157, but I'm wondering if it would be better to go back to firefox, which I don't really want to do...

Perfect Gentleman

Member

Registered: 2012-10-25

Posts: 222

Re: Inox Browser - yet another Chromium Privacy Spin-off

gcarq, will you further make patches for newer Chromium versions?

gcarq

Member

From: Austria

Registered: 2013-10-07

Posts: 40

Re: Inox Browser - yet another Chromium Privacy Spin-off

Don't worry, it is not dead :-)
If everything compiles, an update should be there in 2h.

I will upgrade the patches at least every major release and if I have time some weekly builds too.

Perfect Gentleman

Member

Registered: 2012-10-25

Posts: 222

Re: Inox Browser - yet another Chromium Privacy Spin-off

happy to hear that.
applied patches,except branding, + vaapi patch to Chromium 45.0.2454.85, compiled without error, but i got segmentation fault.

Last edited by Perfect Gentleman (2015-09-03 14:19:06)

gcarq

Member

From: Austria

Registered: 2013-10-07

Posts: 40

Re: Inox Browser - yet another Chromium Privacy Spin-off

Same here, I have to check which patch fails this weekend. @Perfect Gentleman Do you have a clue?
I've pushed the updated patches too early to github :-/, if the bug is fixed I will update the PKGBUILD to use the github patches.

Perfect Gentleman

Member

Registered: 2012-10-25

Posts: 222

Re: Inox Browser - yet another Chromium Privacy Spin-off

@gcarq, honestly, no. i'm not a big expert in reading strace output.
firstly, i thought it was vaapi patch, then rebuild without it, but it didn't help.

sekret

Member

Registered: 2013-07-22

Posts: 289

Re: Inox Browser - yet another Chromium Privacy Spin-off

Glad to see/hear/read, that it's not dead!

orson

Member

Registered: 2015-09-08

Posts: 1

Re: Inox Browser - yet another Chromium Privacy Spin-off

@gcarq
I have just registered an account to say 'thank you'! I considered doing a similar project, and I have just found your patches. Well done, keep it up!

Perfect Gentleman

Member

Registered: 2012-10-25

Posts: 222

Re: Inox Browser - yet another Chromium Privacy Spin-off

gcarq, any progress ?

gcarq

Member

From: Austria

Registered: 2013-10-07

Posts: 40

Re: Inox Browser - yet another Chromium Privacy Spin-off

Jup, seg fault is fixed :-)
Updated packages to 45.0.2454.85

rEnr3n

Member

Registered: 2012-07-25

Posts: 39

Re: Inox Browser - yet another Chromium Privacy Spin-off

Anyone tried ublock on this? It kept crashing.

ERROR:extension_function.cc(396)] Bad extension message types.ChromeSetting.set
ERROR:extension_function_dispatcher.cc(112)] Terminating renderer because of malformed extension message.

gcarq

Member

From: Austria

Registered: 2013-10-07

Posts: 40

Re: Inox Browser - yet another Chromium Privacy Spin-off

@rEnr3n

This looks like a uBlock/Chromium issue: https://github.com/gorhill/uBlock/issues/566

As far as I can see Chromium exposes the WebRTC API altough it's disabled.

Perfect Gentleman

Member

Registered: 2012-10-25

Posts: 222

Re: Inox Browser - yet another Chromium Privacy Spin-off

merged, no errors, thanx, pal

rEnr3n

Member

Registered: 2012-07-25

Posts: 39

Re: Inox Browser - yet another Chromium Privacy Spin-off

@gcarq Strange. Thought this was an inox issue since ublock is running fine on my chromium.

sekret

Member

Registered: 2013-07-22

Posts: 289

Re: Inox Browser - yet another Chromium Privacy Spin-off

I use µmatrix and sometimes when I close inox and want to re-open it, it doesn't load. When I do killall -w inox, a popup shows which tells me that µmatrix has crashed. This happens about every 2 days or so. I always close inox, when there's no tab left, so it's quite often, about ... 5 times a day. So this happens about every 10 times I close inox. I can live with it, but could live better without it