> [: Albert Astals Cid :] > After reading qt5 code it seems you can use XDG_DATA_DIRS to inject > directories in GenericDataLocation, well, it's not that bad i guess
I've just used this fact in a unit test initialization (via qputenv), to point to the location of test-specific catalogs. So it works as such. But now I wonder how should the proposed directory binding method combine with this. Should it have *lower* priority than XDG_DATA_DIRS? It seems to me that otherwise the original feature would be lost, the user would not be able to override the location any more. On the other hand, maybe the directory binding method should not be added at all. For example, I've located the piece of code that Sebastian referred to, and that should be ported by simply deleting all insertCatalog calls. (In general, by removing all traces of translation handling for out-of-library code: applets, dataengines, etc.) -- Chusslove Illich (Часлав Илић)
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Kde-frameworks-devel mailing list Kde-frameworks-devel@kde.org https://mail.kde.org/mailman/listinfo/kde-frameworks-devel
