Progetto Esteso Raccolta Automatica Configurazioni hardware Organizzate Tramite Tarallo Autonomamente
Program to gather data, display it and update T.A.R.A.L.L.O. automatically.
Multiple use cases are possible. More details on each program are provided in the How to install and run section.
Launch peracotta
: that is the GUI that allows you to gather data, parse it and display it.
Options can be configured in the left pane, then after Generate is pressed data is displayed in the right pane. Some basic editing is possible (add and remove items and features, edit feature values). The result can be saved as a JSON or uploaded directly to tarallo.
Launch peracruda
from the terminal: this a script that gathers and parses data, however it offers no editing capabilities.
At the end, you can save data as a JSON or upload it to tarallo directly.
The saved JSON can be uploaded to tarallo or imported from the peracotta
GUI e.g. on another computer, to review and edit it before upload.
If you are desperate and the entire system is unstable due to failing hardware or any other reason, run scripts/generate_files.sh
. That's the most bare-bones way to gather data: the script takes a single (optional) parameter for the output directory, and generates some txt files. No parsing is done.
You can load those files in peracruda
(-f
option) or peracotta
(File > Load raw files
option) and continue from there.
sudo apt install libxcb-cursor-dev
pip install peracotta
generate_files.sh requires root permissions. There are a few ways around this, none of which is terribly secure:
This is what is done automatically: both peracruda and peracotta will run sudo generate_files.sh
where needed.
peracotta will also ask you the sudo password, however keep in mind that is done with a QMessageBox and the password is stored in memory in plaintext. If you don't trust our code, don't type it there.
If you have configured sudo with NOPASSWD, at least for the generate_files.sh script, you can add this to .env:
export GENERATE_FILES_ASK_SUDO_PASSWORD=0
so it will not ask your password anymore but still use sudo.
This is not recommended, but you can run everything as root and disable sudo. This is particularly useful if you don not have sudo at all.
Add this to your .env:
export GENERATE_FILES_USE_SUDO=0
Run sudo generate_files.sh /path/to/output/directory
then load the raw files in peracruda or peracotta. This is probably the safest way, considering that generate_files.sh is pretty short so you can inspect it before running. Everything else will work as usual and won't require root permissions.
Same as before, until the pip install
part. Just install requirements-dev.txt
instead:
pip install -r requirements-dev.txt
This will allow you to run tests: pytest -vv tests
Some markers are also available, e.g. you can run pytest -m gui
to just test the gui, or pytest -m 'not gui'
to test everything else. See pytest.ini
for a list of markers.
If requirements change:
- install the correct version of the requirements (e.g. a new library or a new version of an already installed library)
- with the virtual environment activated, run
pip freeze > requirements-dev.txt
and manually edit the file (add the-r requirements.txt
line and remove non-dev requirements)
If you can't run generate_files.sh because you don't have access to sudo
, such as on our development VM, you can look at tests/source_files
for examples.
This script basically runs sudo ./generate_files.sh
and collect data into an appropriate json for T.A.R.A.L.L.O, but it does so in an interactive manner, so you can either pass it the arguments, or it will ask you for them nicely.
You can also pass as the -f/--files argument the directory where generate_files.sh dropped its files.
You can find the usage below, but keep in mind that the three most important arguments are:
- the path for files generation: if none given, it will default to a tmp directory, and if it exists, you will be asked whether you want to overwrite it
-g | -c | -b
: one of these tells the script where the GPU (or graphics card if it's not integrated) is located. If none of them is given, a menu with the same choices will appear during the execution.--code CODE
and--owner OWNER
: these two parameters are used to add some more information directly into the output json file.-f
to read files from the path instead of callinggenerate_files.sh
again.
usage: peracruda [-h] [-f] [--code CODE] [--owner OWNER] [-p PARSERS]
[-g | -c | -b | --gpu-none] [-v]
[path]
Parse the files generated with generate_files.sh and get all the possible info
out of them
positional arguments:
path optional path where generated files are stored
options:
-h, --help show this help message and exit
-f, --files reuse previously generated files
--code CODE set the code assigned by T.A.R.A.L.L.O
--owner OWNER set a owner
-p PARSERS, --parsers PARSERS
which parsers to use
-v, --verbose print some warning messages
GPU Location:
-g, --gpu computer has dedicated GPU
-c, --cpu GPU is integrated inside the CPU
-b, --motherboard GPU is integrated inside the motherboard
--gpu-none There's no GPU at all
Just need to run it with ./peracotta
or from your file manager. It does everything the terminal based version does and more, all through a GUI.
This will create some txt files with data related to the computer, that will be parsed by launching
peracruda
with -f/--files argument. The hard work is powered by the many read_X.py
scripts, which are the actual
parsers.
Install dependencies on Debian-based distributions (Debian, Ubuntu, Xubuntu, etc):
sudo apt install pciutils i2c-tools mesa-utils smartmontools dmidecode
These are the actual programs that generate the files that we parse.
There are many read_something.py scripts in the parsers
directory: these are used internally by the other scripts.
They can also be launched from the command line, with very basic parameters.
They can also be imported as libraries.
This directory contains some images and other files used by the GUI.
Fan icons created by juicy_fish - Flaticon
More icons created by Kirill Kazachek - Flaticon
Down arrow icons created by Freepik - Flaticon