https://bugs.kde.org/show_bug.cgi?id=465818
Harald Sitter <sit...@kde.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Latest Commit|https://invent.kde.org/plas |https://invent.kde.org/plas
|ma/flatpak-kcm/commit/3d2a3 |ma/flatpak-kcm/commit/5d6a1
|55aec552f49394975550aa36d3e |c0536da8114f1f9ba9d341e3b41
|3634302c |9839d4e3
--- Comment #4 from Harald Sitter <sit...@kde.org> ---
Git commit 5d6a1c0536da8114f1f9ba9d341e3b419839d4e3 by Harald Sitter.
Committed on 17/02/2023 at 13:44.
Pushed by sitter into branch 'Plasma/5.27'.
always use frontend strings but convert to backend strings when saving
this resolves various localization problems where we got confused
between frontend strings (localized) and backend strings (english-y
config file strings). in an attempt to keep this change fairly
conservatively the entire code base now should assume that Value
instances inside the Permission object are localized and must be
delocalized for writing to file.
for filesystem entries that was kind of working already and has been
extended to cover (hopefully) all branches where writeouts occur.
for dbus entries this is now also the behavior and assisted by back and
forth conversion helper functions.
dbus entries also had their config file manipulation adjusted to be a
tad easier to read and a lot more reliable
further refactoring is advised: ideally there'd be a sensible split
between frontend and backend strings so the backend code only ever deals
with backend strings and the conversion for displaying happens
separately from that. also the config file writing/management seems
overly fragile at a glance. the config file management possibly should
get split out of the model so it is easier to comprehensively unit test
it. there should just never be a case where the config breaks because of
us :|
FIXED-IN: 5.27.1
(cherry picked from commit 3d2a355aec552f49394975550aa36d3e3634302c)
M +90 -49 flatpakpermission.cpp
https://invent.kde.org/plasma/flatpak-kcm/commit/5d6a1c0536da8114f1f9ba9d341e3b419839d4e3
--
You are receiving this mail because:
You are watching all bug changes.
