jinsun has quit [Read error: Connection reset by peer]
jinsun has joined #pypy
derpydoo has joined #pypy
<fijal>
hexology: pretty high
derpydoo has quit [Ping timeout: 246 seconds]
derpydoo has joined #pypy
derpydoo has quit [Ping timeout: 246 seconds]
derpydoo has joined #pypy
lesshaste has joined #pypy
derpydoo has quit [Ping timeout: 268 seconds]
derpydoo has joined #pypy
lesshaste has quit [Ping timeout: 272 seconds]
jinsun has quit [Ping timeout: 255 seconds]
jinsun has joined #pypy
derpydoo has quit [Quit: derpydoo]
<mattip>
hexology: please let us know if it doesn't work
<mattip>
we implemented the popular parts of the C-API, but there are still some corners
<mattip>
btw, for c++ you might want to check out nanobind
<hexology>
mattip: i fired up the pypy bullseye-slim container on docker hub with --platform=linux/aarch and had no trouble building shapely (bindings to GEOS) and pyproj (bindings to Proj)
<hexology>
actually the one problem i had is that libproj itself is ancient in the bullseye repos, i need bookworm or sid and didn't want to figure out building it from source
<mattip>
cool FWIW, both are available as binary packages on conda-forge
<hexology>
i didn't do it on my mac because i'm waiting for the native m1 release. i expected that rosetta would have issues (library compiled for arm64, executable running in amd64 emulation?). i could not figure out which nightly build i should download either.
<mattip>
for nightlies, the build that ends in macosx_arm64.tar.bz2 is for M1 native arm64
<hexology>
iirc there are a lot of those on the download page
<hexology>
do i just take the "latest" one at the top?
<mattip>
yup, that is the last one built. We keep the older nightly builds around for historical reasons,
<hexology>
i didn't realize conda supported automatically using pypy like that. i assumed if i installed it in an env that all the python packages would still be installed using cpython
<mattip>
hmm, that is a failure in publicizing the information. Would expanding the information there help?
<mattip>
or maybe we should do a blog post about conda support
<mattip>
"hey, everyone, you should know there are more than 1000 binary packages for PyPy available on conda-forge
<mattip>
"
<tumbleweed>
make it a recommendation on the download page?
<hexology>
+1 to the blog and download page callout
jinsun__ has joined #pypy
jinsun has quit [Killed (osmium.libera.chat (Nickname regained by services))]
jinsun__ is now known as jinsun
jinsun__ has joined #pypy
jinsun has quit [Killed (molybdenum.libera.chat (Nickname regained by services))]