arrow

Feature Request

Many locales have the bare minimum when it comes to test cases. While I understand it can be tedious and repetitive to write out test case

We now have native ODBC support upstream. This has to be exposed in polars similarly to existing IO readers and writers.

Is your feature request related to a problem? Please describe.
While reviewing PR #9817 to introduce DataFrame.diff, I noticed that it is restricted to acting on numeric types.

A time-series diff is probably a very common user need, if provided a series of timestamps and seeking the durations between observations.

Pandas supports diffs on non-numeric types like timestamps:

TPC-DS has many queries with IN predicates where all elements are constants. It's a low-hanging fruit if we could implement an InSet function for this all constants value case.

While implementing this, we could either use a hashtable or a chain of if-elif-else, depending on the length and the type of the constants array.

Q8:

 WHERE substr(ca_zip, 1, 5) IN (
               '2412

We no longer need to control the number of concurrent kernels, since now we control the number of concurrent tasks

Note sure if it could be interesting but:

When registering a table:

addr: 0.0.0.0:8084
tables:
  - name: "example"
    uri: "/data/"
    option:
      format: "parquet"
      use_memory_table: false

add in options:
glob

pattern: "file_typev1*.parquet"

or regexp

pattern: "\wfile_type\wv1\w*.parquet"

It would allow selecting in uri's with different exte

It would be helpful to have Fletchgen output warnings for unused metadata fields that start with fletcher_. For example, (this happened to me) when someone adds fletchgen_epc to Schema metadata instead of Field metadata.

Problem description

Reading a dataset with eager's read functionality raises a ValueError when providing columns.

Example code (ideally copy-pastable)

import pandas as pd

from tempfile import TemporaryDirectory
from functools import partial
from storefact import get_store_from_url

from kartothek.io.eager import store_dataframes_as_dataset, read_dataset_as_data

Motivation:

1. Improved compile times (at least by 2x compared to arrow-rs).
2. Faster Parquet impl
3. Projects are migrating to arrow2 (including Datafusion and Polars)