[sos-devel] Should _expand_copy_spec() return broken symlinks

[Louis Bouchard]

Hello,

I just ran into this while testing 3.4. Ubuntu Trusty has a bug where
/etc/blkid.tab turns out to be a broken symlink.

But add_copy_spec() does this :

> for _file in files:
>                 current_size += os.stat(_file)[stat.ST_SIZE]
>                 if sizelimit and current_size > sizelimit:
>                     limit_reached = True
>                     break
>                 self._add_copy_paths([_file])

files comes from :

> self._expand_copy_spec(copyspec)

and _expand_copy_spec() will return /etc/blkid.tab, since the file exists, but
os.stat() will fail since there is nobody at the other end of the symlink.

Obviously, it is an O/S bug that needs to be fixed, but should
_expand_copy_spec() return something when there is no _real_ file at the other end ?

If so, I'm happy to fix this, as it breaks my packaging tests. Otherwise, I'll
fix the packaging accordingly.

Kind regards,

...Louis

-- 
Louis Bouchard
Software engineer, Cloud & Sustaining eng.
Canonical Ltd
Ubuntu developer                       Debian Maintainer
GPG : 429D 7A3B DD05 B6F8 AF63  B9C4 8B3D 867C 823E 7A61