Ben Baumgold

Hi @ExpandingMan. Did you ever manage to find a solution to being able to read HDF5 files from an S3 object store? The [HDF5/S3 connector](https://www.hdfgroup.org/solutions/cloud-amazon-s3-storage-hdf5-connector/) may be useful here, but...

@mkitti - that’s what I suspected. Any idea if this is available/integrated with HDF5.jl? My understanding is virtual file drivers need to be selected at HDF5 compile-time. I presume we’ll...

As a work-around all nullable types can be converted to use Missing instead of Nothing, which seems to allow Python to read the Arrow files generated by Arrow.jl. The issue...

Related to #258 and [ARROW-15767](https://issues.apache.org/jira/browse/ARROW-15767)

I put up #277 to support appending to Arrow files that are either in Arrow IPC format (already supported thanks to @tanmaykm's #160) or Arrow file format (currently unsupported). @quinnj...

Maybe this code can be further simplified to reuse more of the Julia Base Enums module as follow? ```julia macro scopedenum(T::Union{Symbol,Expr}, syms...) # extract typename and basetype from T typename,...

I ran into the same issue recently. Any update here? Thanks.

@quinnj - thanks for your feedback. This PR only includes the minimal code changes to illustrate the approach. Since you seem to approve so far, I’ll take it one step...

On second thought I don't think this approach is viable since Julia uses Dec 31 0000 as Epoch: ```julia julia> DateTime(Dates.UTInstant(zero(Millisecond))) 0000-12-31T00:00:00 ``` Given that `Dates.UTInstant` uses an `Int64` to...

I wholeheartedly agree. Who do we need to convince in order to make a change?