Re: Wire format examples
Felipe Gasper <felipe-c4PQK5fb7feI/[email protected]> Thu, 17 Oct 2024 23:48:08 -0400
| Newsgroups | gmane.comp.freedesktop.dbus |
|---|---|
| Message-ID | <[email protected]> |
> On Oct 17, 2024, at 9:09=E2=80=AFPM, Link Dupont <[email protected]> = wrote: >=20 >=20 >=20 >> On Oct 17, 2024, at 17:00, Lawrence D'Oliveiro = <ldo-kC51rjc0cqs1dNLGKZg6taU/[email protected]> wrote: >>=20 >> On Thu, 17 Oct 2024 11:17:17 -0400, Robert Middleton wrote: >>=20 >>> ... the fact that you can have nested objects and varied lengths of >>> data is what really trips me up. >>=20 >> And note that objects requiring greater alignment can occur inside >> parent objects with lesser alignment requirements. So the only way to >> know where the padding should be placed is to look at the message as = a >> whole. >=20 > In my exploration so far, this is by far the most opaque element of = the whole marshaling format to me. I haven=E2=80=99t found a pattern yet = that I can use to figure out when and where to pad the data buffer. The standard documents it variously, as I recall, though I also had to = do a fair bit of trial & error. D-Bus doesn=E2=80=99t lack for implementations; you could look around at = various ones for clarifications of the standard. -FG=