* PosteMobile is now a full MVNO and it has its own mnc
* Entries extracted from platina-user 10 QKQ1.190910.002 V12.0.3.0.QDTMIXM release-keys
Change-Id: I70d314c558074f2564e88c761c22bc842b973d99
* CoopVoce is now a full MVNO, thus deprecating older ESP entries (still kept here because not all users changed SIM card yet)
* Entries extracted from platina-user 10 QKQ1.190910.002 V12.0.3.0.QDTMIXM release-keys
Change-Id: I9a59ba54926e27ce64d5adbca42fb7b058436e6f
* Ignore the block devices in case their mount points are symlinks.
This is common on devices where maintainers have chosen not to use
real partitions because of their size being too small to be useful
Also `continue` instead of `break`. Oops.
Change-Id: I3e27abe510219066ecacd81d099220ac8e119f9f
There are addon.d scripts that rely on the value of ADDOND_VERSION
to determine if they're being called from a-only vs a/b backuptool.
If they declare ADDOND_VERSION=3, they shall stop doing that;
otherwise offer them the same environment, that is unset ADDOND_VERSION
for a-only backuptool.
Change-Id: I1be21eda2e6ec9837b3080bb5e7fbe5241318eaa
* Also: add a lot of additional info for many of the numbers based on
the comments from the commit prior to this
Change-Id: I937637c8f002d193ed6b23ac60c1f350d7d2eec7
* The plan was to support odm & oem too, but it turned out
unnecessarily complicated due to their symlink setup,
while being most likely unused.
The partitions were removed from the list of supported ones
in a-only backuptool, but were forgotten in ab backuptool.
Change-Id: I58a01cdc3f5c4239048b204f3313f4bf697dd60a
For scripts declaring ADDOND_VERSION=3 automatically mount
vendor, product, system_ext and others (when they're dedicated partitions).
Also expose the get_output_path() function to get the path to where
a file is mounted in case it lives in a dedicated partition.
ab exapmles:
get_output_path "system/product/priv-app/MyApp.apk" = "/postinstall/product/priv-app/MyApk.apk"
get_output_path "system/app/MySystemApp.apk" = "/postinstall/system/app/MySystemApp.apk"
a-only examples:
get_output_path "/mnt/system/system/product/priv-app/MyApp.apk" = "/mnt/system/system/product/priv-app/MyApp.apk"
******************************************************************
Instead of cycling all scripts for each stage, run
pre-backup -> backup -> post-backup in quick succession
(and likewise for restore), to ensure backwards compatibility
with scripts that wrongly assumed their environment not to
change between steps.
This is needed because we want to undo any mounting done for V3
scripts when executing V2 scripts. If a V2 script did mounting in
pre-restore and expected things to still be mounted in restore,
we would break their (yes incorrect) assumption.
Change-Id: I73fbad6f45824fed99e4482128769435348588f5
FscryptSetDirectoryPolicy no longer tries to infer the action from the
filename. Well mostly; it still assumes top-level directories in /data
should be encrypted unless the mkdir arguments say otherwise, but
it warns.
Bug: 26641735
Test: boot, check log messages
Change-Id: Id6d2cea7fb856f17323897d85cf6190c981b443c
Adding "default" to the type list, will allow to connect to mobile internet via LTE again.
According to the user, even if enforcing LTE here, it won't work unless adding "default".
I couldn't find any official resources, just that this APN seems to relate to CDMA and other people fixed the issue by adding the APN manually:
https://help.ting.com/hc/en-us/community/posts/206433277/comments/216019607
Therefore trusting the reporter in the ticket below that default is enough:
https://gitlab.com/LineageOS/issues/android/-/issues/2819
Change-Id: I931f5355af74e197020032a1581dd11abf84f10c
In Q, we added a new bugreport API that asks for user consent before
showing the "Share" notification for the bugreport. We will be migrating
all bugreporting clients to use the API in R.
In the new API workflow we are moving away from broadcasts and using
callbacks to communicate with dumpstate. Finished bugreport notification
broadcast used to share the final path of the bugreport but API now uses
file descriptors instead. Hence, moving away from broadcasts.
The full bugreport in this case can be retrieved from dumpstate directory
using 'adb pull' or finding and attaching the file directly to gmail or
betterbug.
Remove outputfile flag since it's unused.
Bug: 135186519
Bug: 137825702
Test: Build and flash to the device. Create a bugreport using combo
keys, share notification does not pop up. Full bugreport created in
bugreports/ dir
Change-Id: I3c15123a6a4ed62ae9cb87f5dd07db60b1416dfc
Using cp instead of mv during the restore step resulted in double the
amount of free space and inodes being used.
For example, a nano arm64 opengapps install uses around 500M. But during
a backup/restore process when the Updater is used, it temporarily uses
1G on the other slot. That causes failed updates on devices that
don't have that much free space on the system partition.
Change-Id: I57ba2e0f52d328d4ff91d47a3db45f47bb8402af
Extracted from sony stock rom firmware 'H8266_Customized DE_1313-6279_52.1.A.3.49_R3C'
E-Plus got merged into O2
Synced the newest O2 changes
Change-Id: I67e6987769d0bcc271e58e7416468faa902a870b
* There is no reference to ANDROID_CACHE now, and
/persist does not exist on most devices, neither
can we write to it.
Change-Id: I91af1e6f571ced317d195e3a7901bf4c269486a1
Backup/restore functionality was broken in the
Ia1f4ae95c9e4dae4df844853e81c264bc838f177 change
because of incorrect check of the function's result.
check_prereq() function refactored to return 0 if
backuping/restoration is possible. Any work should be
performed only if check_prereq() succeeds.
Change-Id: Ic977dba675df58a228ef4b882b25beb66cc9d2c6
For non AB devices system partition should be unmounted
if check_prereq function fails.
This patch also refactors backuptool a bit for AB devices
in order to look same as backuptool for non AB devices.
Change-Id: Ia1f4ae95c9e4dae4df844853e81c264bc838f177
* This check was supposed to check whether the script
addon.d version was lower than backuptool's
* Given that the backuptool addon.d version is 1, this
isn't going to happen ever making this check completely
unuseful
Change-Id: I2464749b52bf4e8825e0b4ef42500ee7d3bbfa61
* For some odd reasons executing `cd /system/addon.d` makes the system
hang and unmount error:
umount: /system_root: Device or resource busy
* Don't change directory to not allow the system partition hang
Change-Id: I3d30bdc59c2f05d16823e99046c1dce2e1e6eb73
* If any of these two function gets run on a recovery mounting system
to /system, /system/addon.d won't exist while /system/system/addon.d will.
* Run the functions with $S as argument to make this work correctly
Change-Id: I02e7b91429a9e74d28bdb77e56955dad97ca75ac
* The path /postinstall exists only for A/B, causing:
grep: /postinstall/tmp/addon.d/*sh: No such file or directory
Change-Id: Ia07b3029e949c3e08302457cd08798a4dde00ef6
* Dynamically mount system to the path chosen by the recovery through backuptool
* This can be helpful because of the fragmentation that will happen with system mount in recovery after Q
Change-Id: I2d1e775efcf87e33319bc7790d1e54bca72116d3
