>From a release engineer perspective (without java knowledge) I agree with
Micah, I'd rather make a patch release for an older version if needed but
modernize the codebase and simplify CI!
On Sat, Oct 7, 2023 at 5:27 AM Micah Kornfield
wrote:
> I think given the stability of Arrow Java,
I think given the stability of Arrow Java, dropping support probably makes
sense. If a bug comes up or consumers really need to new features we can
always make a patch release of an older version.
On Thu, Oct 5, 2023 at 3:13 PM Dane Pitkin
wrote:
> I also learned today that Apache Spark has
+1!
On Fri, Oct 6, 2023, 8:03 PM Matt Topol wrote:
> +1
>
> On Fri, Oct 6, 2023, 6:55 PM Benjamin Kietzman
> wrote:
>
> > +1
> >
> > On Fri, Oct 6, 2023, 17:27 Felipe Oliveira Carvalho >
> > wrote:
> >
> > > Hello,
> > >
> > > I'm writing to propose "+vl" and "+vL" as format strings for
+1
On Fri, Oct 6, 2023, 6:55 PM Benjamin Kietzman wrote:
> +1
>
> On Fri, Oct 6, 2023, 17:27 Felipe Oliveira Carvalho
> wrote:
>
> > Hello,
> >
> > I'm writing to propose "+vl" and "+vL" as format strings for list-view
> and
> > large list-view arrays passing through the Arrow C data interface
+1
On Fri, Oct 6, 2023, 17:27 Felipe Oliveira Carvalho
wrote:
> Hello,
>
> I'm writing to propose "+vl" and "+vL" as format strings for list-view and
> large list-view arrays passing through the Arrow C data interface [1].
>
> The previous proposal was considered a bad idea because existing
Hello,
I'm writing to propose "+vl" and "+vL" as format strings for list-view and
large list-view arrays passing through the Arrow C data interface [1].
The previous proposal was considered a bad idea because existing parsers of
these format strings might be looking at only the first `l` (or
Agreed, it's unfortunately not just a simple tradeoff. We have discussed
this a bit in [1] and in several other threads around this topic. If we say
that Arrow is about interchange and not execution, so we shouldn't adopt
the pointer version that DuckDB uses, that means we're also making
Le 06/10/2023 à 17:54, Felipe Oliveira Carvalho a écrit :
Hello,
Since existing C Data Interface integrations sometimes don't parse beyond
the first `l` (or `L`) I'm going to start a new [VOTE] thread with Dewey's
suggestion:
Regardless of which format string we choose for ListView, a bug
Hello,
Since existing C Data Interface integrations sometimes don't parse beyond
the first `l` (or `L`) I'm going to start a new [VOTE] thread with Dewey's
suggestion:
+vl and +vL
If anyone objects to that and has a different suggestion, reply here so I
don't have to spam the list with too many
> I feel the broader question here is what is Arrow's intended use case -
interchange or execution
The line between interchange and execution is not always clear. For
example, I think we would like Arrow to be considered as a standard for UDF
libraries.
On Fri, Oct 6, 2023 at 7:34 AM Mark
For the index vs pointer question - DuckDB went with pointers as they are more
flexible, and DuckDB was designed to consume data (and strings) from a wide
variety of formats in a wide variety of languages. Pointers allows us to easily
zero-copy from e.g. Python strings, R strings, Arrow
You might be able to find relevant examples in the Delta lake Rust
implementation[1] which I believe features DataFusion catalog integration
You might also find the "Catalogs, Schemas, and Tables" section of the
library developers guide[2] helpful.
[1] https://github.com/delta-io/delta-rs
[2]
Given I don't see any input from the DuckDB / Velox development team (this
discussion seems primarily Arrow developers) I have filed a ticket in
DuckDB requesting their consideration[1] and tried to bump the attention of
the existing ticket in Velox[2]. Perhaps their input will provide a way
Worth noting that here were some minor changes made to the spec while
the vote was active:
- The "uniform_dimensions" metadata key was removed, since this can
also be inferred from the "uniform_shape" information
- The shape of non-constant dimensions in the "uniform_shape" entry is
now
Hey All,
We have 4 binding +1 votes, no non-binding +1 votes, and no -1 votes, so
the vote passes.
Thanks everyone for your work and participation on this!
As a follow up we will:
[ ] merge changes to the format (
https://github.com/apache/arrow/pull/37166/files)
[ ] merge C++ and Python
15 matches
Mail list logo
