Analyzing All Devices

The images_util.sh is a simple utility used for maintaining a large collection of Lineage install packages. These are used to research the binary blobs for each device. This uses the extractor.py script to do all the work. This script gets run at the top level to process each device sequentially.

For the Lineage 22.2 release, there are 210 packages. I use this script frequently, but your mileage may vary... you'll probably need to to edit a few paths for your system.

Package Cache

All the packages have been downloaded from Lineage, for the 22.2 release there are 210 supported devices. My cache has a directory structure like this:

images/
    /google
        Pixel_7a (linked to by lynx)
        Pixel_9_Pro (linked to by caiman)
        ...
    /motorola/
        Moto_G5 (linked to by rhode)
        Edge_40 (linked to by rtwo)
    ...

Under each device is the Lineage install zip files for the 22.2 and 22.1 releases. To make it easier to refer to a device by it's build, I create a symbolic link from the Lineage build name to the device name. To do this for all packages you can create the links this way:

[path]/image_utils.sh -l

Since there is no way to get the device model from the Lineage build, this also produces a config file called devices.list which can be used to bootstrap the database with this binding which simplifies later SQL queries by be able to relate a Lineage build with a device. This file also gets used by the other functions in the image_utils.sh utility instead of having each function scan for the Lineage zip files, with the advantage of making the code faster and more readable.

[path]/[import-device](import-device.md) -v -b devices.list

Options

--help (-h):    Display this help screen
--dryrun (-d):  Just dry run,don't execute anything
--remove (-r):  Remove all generated files
--link (-l):    Link build names to device name
--extract (-e): Extract files from package
--clone (-c):   Clone files
--unmount (-u): Unmount all device file-systems
--build (-b):   Build packages
--import (-i):  Import blobs into a database
-all (-a):      Do everything, which takes a long time

All of these options are just a wrapper used to process all 210 devices. The extractor script does all of the actual work. This makes it easy to rescan all the files as more analysis of each file is implemented.

--remove

Extracting files from the Lineage install packages generates many files and sub-directories. This option unmounts any mounted directories, and then remove all the files except the Lineage zip file.

--link

Traverse the directories containing Lineage install packages and link the Lineage build name to the device name to make it easy to navigate.

extract

Extract the files from the Lineage packages, which are filesystem images. This doesn't mount the filesystem, it only extracts the top level files.

clone

If the filesystem images have been extracted from the payload files. this will mount them and extract all the files listed in the proprietary-*files.txt files in Lineage. This clones all the files to the output directories ready to be imported into postgres. This is required before importing.

build

The uses the Lineage way of extracting files, so main fail some for some devices. If the extraction succeeds, it builds an install package. This may run for s very long time, use ccache to speed up recompiling.

import

Imports the metadata on all the devices from the cloned files into a postgres database. This assumes you are using the output of cloning using extractor. This also assume you have boot strapped the database with the devices.list file.