To be able to run tuhi-kete, we need a tuhi server running. Provide a
script that spawns one if none is running.
Running simple 'python3 setup.py --install' started to go havoc by
complaining that /app/lib/python3.6/site-packages/easy-install.pth
was read-only.
The solution mentioned in https://github.com/flatpak/flatpak-builder/issues/5
didn't do the trick.
So using https://stackoverflow.com/questions/6301003/stopping-setup-py-from-installing-as-egg/27175492#27175492
as a trick to install the dependencies without the egg part
To start tuhi-kete:
$> flatpak run --command=tuhi-kete org.freedesktop.tuhi
Note that currently the fetched files are not available outside of the
sandbox
Related to #62
"pair" is alrady taken by Bluetooth and since we have a bluetooth device here,
it can cause confusion. Use "register" instead, with an explanation in the
README for the more paranoid of us.
Fixes#67
While running the app in flatpak, there are a few deprecation warnings
coming in:
PyGIDeprecationWarning: GObject.SIGNAL_RUN_FIRST is deprecated;
use GObject.SignalFlags.RUN_FIRST instead
PyGIDeprecationWarning: GObject.property is deprecated;
use GObject.Property instead
PyGIDeprecationWarning: GObject.MainLoop is deprecated;
use GLib.MainLoop instead
In the normal mode (without -v) we'd see the errors, but not the messages when
we reconnect. Elevate the "connected to" message to info so it shows up.
And slight rewording, which may or may not be better.
'list' and 'listen' are just too close to each other, meaning the
completion stops at 'list' while the lazy developer/tester want to
do a 'listen'.
Change the name of the less used command so we have separate spaces
for completion
Better keep this in a public place where it won't vanish from our disks.
Note that this is a python 2 script (because btsnoop), and it should not
be part of the installation, ever.
Tuhi raise -EACCES if the UUID stored is not valid anymore.
Instead of shouting some obscure error, we should notify the
user that a pairing need to happen or the device will not talk
to us.
Also fix a small sign mistake in calling os.strerror() for general
errors
Fixes#6
workaround for cumberness of having to type timestamps.
Instead of having some fancy formatted timestamp, help the user
by displaying the translated value
Let's the developers of tuhi to consume more energy in their monitor
to light up more pixel when displaying single quotes whereas double
quotes.
It also makes the whole bit more uniform.
Fixes#32
We are using argparse in the various prompt function that already
provide a NameSpace class. We can just reuse it instead of rebuilding
everything.
Given that we enforce/parse the type of the arguments, there is no
need to do more processing.
And let us handle the local time zone conversions in the client, because
having timestamps in localtime in the daemon is confusing and error-prone (how
is the client supposed to know what localtime was?)
Modified version from
https://stackoverflow.com/questions/384076/how-can-i-color-python-logging-output
Modifications (from that post) include:
* make flake8 happy by removing whitespaces and whatnot
* move everything into a class
* add the LIGHT_* colors
* remove the magic 30+ offset, assign the real color codes instead
* remove the $BG feature, we don't need background colors
Errors that occur during shutdown because one of the things hasn't been set up
in the expected way do not matter. For the shell we still require a manual
user exit though.
File "./tools/tuhi-kete.py", line 344, in stop
self.manager.disconnect(self.s1)
AttributeError: 'Searcher' object has no attribute 's1'
File "./tools/tuhi-kete.py", line 267, in _on_name_vanished
self.mainloop.quit()
AttributeError: 'NoneType' object has no attribute 'quit'
If a log message comes in the background, it will clear the current
command.
Rewrite the current line buffer from readline so we get more user
friendly.
We overwrite the current formatter to display or not
the prompt depending if we are in the prompt or not.
To prevent races between the background events and our
current configuration, we acquire/release the lock on
the current handler.
Only list and listen commands are currently implemented.
The Ctrl-C handling has been a little bit tricky. The default GLib
mainloop tends to add its own SIGINT handler, which prevents us to
gracefully handle the KeyboardInterrupt exception during cmdloop().
So we need to create the mainloop in TuhiKeteShellWorker directly,
but bypassing the GLib.Mainloop() python facility.
For the various commands, it is easier if we have a common interface
than just a simple function call.
Each current command runs something before the mainloop is created, and
then something after the mainloop is terminated. Having such a worker
allows us to have only one place where we start the mainloop, meaning
that the interactive prompt will not try to start it more than once,
and most above, will not kill it in the middle of a command while other
commands are still running.
Benjamin Tissoires