IRC channel logs

2020-11-22.log

back to list of logs

<leoprikler>sneek: later tell a_v_p heads up "gdbus" is already claimed by the GNOME people IIRC (although potentially deprecated in favor of gio)
<sneek>Will do.
<daviid>sneek: later tell a_v_p nice, fwiw GNOME offers a low-level and high-level D-Bus support API, part of the Gio library, which guilers can (should be able to - might trigger a few bugs but ...) access using guile-gi or g-golf - see https://developer.gnome.org/gio/stable/ch01.html, https://developer.gnome.org/gio/stable/gdbus-lowlevel.html and https://developer.gnome.org/gio/stable/gdbus-convenience.html
<sneek>Got it.
<ane>how can I make a thunk like "(foobar)" in the C API?
***catonano_ is now known as catonano
<a_v_p>Good <time-of-day>, Guilers.
<sneek>Welcome back a_v_p, you have 2 messages!
<sneek>a_v_p, leoprikler says: heads up "gdbus" is already claimed by the GNOME people IIRC (although potentially deprecated in favor of gio)
<sneek>a_v_p, daviid says: nice, fwiw GNOME offers a low-level and high-level D-Bus support API, part of the Gio library, which guilers can (should be able to - might trigger a few bugs but ...) access using guile-gi or g-golf - see https://developer.gnome.org/gio/stable/ch01.html, https://developer.gnome.org/gio/stable/gdbus-lowlevel.html and https://developer.gnome.org/gio/stable/gdbus-convenience.html
<a_v_p>leoprikler: I'm using "gdbus" prefix only internally in Guile-DBus for now, public (Scheme) procedures are all start with "dbus-<something>". Do you think it may lead to issues?
<a_v_p>daviid: Alright, that's a good point. Although I don't think that my work and Guile-GI/GNU G-Golf fully overlap, because my library does not depend on Gnome and GTK.
<a_v_p>Also as I found recently there's AC-D-Bus, pure Scheme implementation of DBus protocol: https://gitlab.com/weinholt/ac-d-bus/
<daviid>a_v_p: you don't need to use gtk to access the dbus low or high level ai from gio, just gio, and you may even selectively import stricktly what you need ... i mentioned becaue writing and maintaining a good binding is difficult, dbus even more I think ... using the gio low/high levelapi is nearly 'free', if you do so using guile-gi or g-golf ...
<daviid>youalso have zero maintaince cost,since the gnome team does it for us ...
<daviid>a_v_p: anyway, do as you wish of course, but 'in your position' I would definitely at least try, before to further implement, either guile-gi or g-golf and see how your needs go using either the low, the high or both api's provided by the gio library -
<daviid>or ac-d-bus indeed ... too many options :):)
<tohoyn>sneek, botsnack
<sneek>:)
<tohoyn>lloda`: are you there?
***apteryx is now known as Guest26291
***apteryx_ is now known as apteryx
<mwette>a_v_p: I ran my ffi-helper on libdbus a few years ago. I have some demo's for that, which might be helpful. See https://git.savannah.nongnu.org/cgit/nyacc.git/tree/examples, looking at system/dbus.scm, and nyacc/lang/ffi-help/dbus*.
<mwette>a_v_p: ^-these examples use the auto-coded ffi from ffi/dbus.ffi. To get the associated .scm you'd have to download and run the ffi-helper. I think your approach will yield a more efficient, memory safe and portable implementation.
<a_v_p>mwette: Thanks!
<mwette>yw
***sneek_ is now known as sneek
<manumanumanu>Ahoy folks!
***sneek_ is now known as sneek
<manumanumanu>ArneBab: fyi, goof-loop just got :final expressions that work like in racket. Now I only want to add finalizers to :for -clauses so that I can do (in-file "path/to/file") that actually closes the file. I also added a generator-based interface so that I can do (in-indexed ...) and (in-cycle ...). That only has the same overhead as srfi-158 styled generators, which means it is pretty fast.
<manumanumanu>which was my chore for the day! good night everyone.