You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think it would be a lfie-changing feature for some projects - basically, this would enable to export data from a Vec<Struct> almost straight to a pd.DataFrame, how cool is that?
Wonder what are the thoughts on this? Has this been considered? Otherwise, I have a bit of free time and so I could probably give it a try. This would require a separate proc-macro crate (rust-numpy-derive), and it will probably require a bit of reworking the existing dtype constructs.
(A bit of pre-history: I've implemented support for numpy record arrays in pybind11 a while ago, and then I've implemented support for custom user structs type mapping in hdf5 crate via proc macros - I think adding support for recarrays here should be somewhere inbetween the two.)
The text was updated successfully, but these errors were encountered:
I think it would be a lfie-changing feature for some projects - basically, this would enable to export data from a
Vec<Struct>
almost straight to apd.DataFrame
, how cool is that?Wonder what are the thoughts on this? Has this been considered? Otherwise, I have a bit of free time and so I could probably give it a try. This would require a separate proc-macro crate (rust-numpy-derive), and it will probably require a bit of reworking the existing dtype constructs.
(A bit of pre-history: I've implemented support for numpy record arrays in
pybind11
a while ago, and then I've implemented support for custom user structs type mapping inhdf5
crate via proc macros - I think adding support for recarrays here should be somewhere inbetween the two.)The text was updated successfully, but these errors were encountered: