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
It would be nice to be able to specify additional arguments for specific format in write_object. The desired state would be to be able to run something similar to
df.to_csv(file_name, index=False)
Similarly to how engine is solved I can imagine either expanding format_options or adding a new argument, e.g. format_kwargs and then have smth like
I have to admit that I don't like this :), including the already present engine etc. This is a wrong abstraction, where we are putting fsql in front of existing different incompatible layers, forcing it to adopt all their API idiosyncracies
what I'd like to do instead is to get a get_file_descriptor(url) which you would then use in any function you like (df.to_csv, df.to_parquet, etc); or, in case we'll ever support partitioned writing, to have write_partitioned_df(url_base, partition_columns, single_partition_writer: Callable[[fd, df], None])
It would be nice to be able to specify additional arguments for specific
format
inwrite_object
. The desired state would be to be able to run something similar toSimilarly to how
engine
is solved I can imagine either expandingformat_options
or adding a new argument, e.g.format_kwargs
and then have smth likeThe text was updated successfully, but these errors were encountered: