Bug#1029957: g-ir-scanner: produces FHS violations by putting architecture-dependent files in /usr/share

[Helmut Grohne]

Hi Simon,

Thank you!

On Sun, Jan 29, 2023 at 06:39:27PM +0000, Simon McVittie wrote:
> Control: forwarded -1 https://gitlab.gnome.org/GNOME/gobject-introspection/-/issues/323

It is impressive just how far you moved this issue already. I fear there
is little to add.

In particular, I was unaware how this already affecs RPM distributions
due to the silent file overwrite issue in RPM.

> To minimize regressions, I think we probably want "most" GIR XML files
> to remain in /usr/share, with only the ones that genuinely vary between
> architectures moved into an architecture-specific place. That would make
> their handling similar to C/C++ headers, which similarly are mostly
> placed in an architecture-independent location (/usr/include) but can
> be relocated into an architecture-specific location (/usr/include/TUPLE)
> when necessary.

In that forwarded issue, you hint a possible way forward where Debian
would deviate from upstream. Given upstreams unwillingness to even
consider this problem combined with this being a real problem affecting
multiple distribution families, I think moving forward with a
Debian-specific solution as you depicted would be the best of options
available. I would expect that adding a Debian-specific location would
be a patch that rebases easily.

In any case, we can now centrally block (in the BTS sense) all requests
for M-A:same on gir packages on this bug to avoid repeated discussions
and users pinging for progress.

Helmut