IRC channel logs

<ArneBab>ieure: I don’t think that’s too hard to do, but it sure is too hard to find. When I started with Guile, I wrote the guile-basics with a lot of basic tasks that should be easy to find but are not: https://www.draketo.de/proj/guile-basics/ -- some of these are now part of py2guile: https://www.draketo.de/proj/py2guile/

<ArneBab>I hope that with Naming & Logic the start will be easier now: https://www.draketo.de/software/programming-scheme

<sneek>wb tohoyn!

<tohoyn>sneek: botsnack

<sneek>tohoyn, you have 3 messages!

<sneek>tohoyn, daviid says: I reviewed your coe, here is a version of the dbus-server-test.scm https://bpa.st/67JA, and here is the dbus-client-test.scm https://bpa.st/NA6Q - for some resaon, dbus isn't properly filling ffi-args when strings and/or pointers (not always as least) - I did ask why/for-help in instrospection, let's see

<sneek>tohoyn, daviid says: all ffi-arg are treated the same, see g-golf-callback-closure-marshal and ffi-arg->cb-arg in (g-golf hl-api callback), then (g-golf support ffi) for the decodong of their values, in particular ffi-arg-string->scm and ffi-arg-pointer->scm - for some still isterious reasons, dbud does not fill those properly, at least afaict - I extensively use callback in another app, but all callback args are numbers, and they

<sneek>tohoyn, daviid says: i forgot to mention, i patched g-golf to proper handle GDBus prefix methods and funnction names, devel branch ...

<sneek>:)

<mwette>sneek: bstruct?

<mwette>sneek: software?

<mwette>sneek: botsnack

<sneek>:)

<mwette>/me, in nyacc dev-3.02 branch, has added support in the ffi-helper for alternate backends (from github.com/mwette/guile-fhbe). Currently bytestructures seems to be working; bstructs will be a tough one.

<mwette>ACTION , hmmm

IRC channel logs

2025-12-21.log