Mixing async and sync I/O leads to a multitude of complications, and
generally speaking it's much more convenient to stick to one paradigm
or the other. Since axum (and many other HTTP servers) use async,
and since async is a convenient model for performing operations like
"handle an MPSC message or file read, whichever happens first", let's
commit to an async interface.
This crate's code is nearly entirely autogenerated using an ASN.1
parsing tool called hampi. The code in src/lte_rrc.rs shouldn't be
manually modified, and should only be regenerated using hampi (or a
similar tool).
Instead of reading/writing to a single QMDL file, we now can manage
a directory of several files, and have the ability to start/stop writing
to them on the fly.
This commit also adds graceful exiting to the server, so we can perform
cleanup steps when the server's exiting.
Add tools for launching a root shell on the device, and for sending
serial commands to the device. Extend the make.sh script to push those
and configure a root shell. Commands can now be executed as root via:
adb shell rootshell -c \"touch /tmp/test\"
allowing automatic configuration of the tooling.
This commit does a couple things:
1. breaks out the pcap streaming logic into its own module
2. bundles wavehunter/static files into the binary for easy distribution
3. serves those static files
4. serves dynamic json representing system and diag stats
I also threw together the world's ugliest website to display all this.
1. Main binary now parses a config toml and uses its paths for the
output pcap and qmdl files
2. Previously we were relying on DiagDeviceError for several things
that aren't a diag device. This modularizes that error in a way
that both improves the error descriptions, and also allows for
better separation of concerns.
Will Greenberg