Closed AMDmi3 closed 5 years ago
It seems that all parsers support the field (where it can be used as is) now. Here's current list of extracted values:
arch | families |
---|---|
aarch64 | {openmandriva,sisyphus} |
all | {luarocks} |
any | {arch,msys2} |
armv7hl | {openmandriva} |
athlon | {sisyphus} |
e2k | {sisyphus} |
e2kv4 | {sisyphus} |
i486 | {slitaz} |
i586 | {openmandriva,pclinuxos,sisyphus} |
i686 | {msys2,sisyphus} |
linux-x86_64 | {luarocks} |
macosx-x86 | {luarocks} |
macosx-x86_64 | {luarocks} |
noarch | {openmandriva,pclinuxos,salix,sisyphus,void} |
pentium3 | {sisyphus} |
pentium4 | {sisyphus} |
ppc64 | {adelie} |
ppc64le | {sisyphus} |
rockspec | {luarocks} |
src | {aix,centos,fedora,luarocks,mageia,mer,openmandriva,opensuse,pld,rosa} |
win32-x86 | {luarocks} |
x86_64 | {alpine,arch,kwort,msys2,openmandriva,pclinuxos,salix,sisyphus,slitaz,solus,void} |
The field is not used anywhere yet and is not displayed on the site, so I'm going to disable it now. If it's needed at some point (for example, when we parse binary Debian repos in addition to source unes, or when mingw:*
support is improved (repology/repology-rules#10) it can be easily enabled, but it'll probably need some normalization and a policy for handling absent arch (mostly source-based repos, e.g. we could manully fill arch as src
there). Also, same repos define a list of [supported] arches. These can go to the same, or other field depending on how arch
is used.
all
, any
, and explicit list of arches32bits
/64bits
architecturesx86_64_linux
x86
and x86_64
noarcch
or list of repos)
After #788 is implemented, we need to fill this field in more parsers.