I am happy to close this; it would be lovely if the documentation was more explicit about this issue. I was certainly surprised even after a close reading of the docs.

Jamie

On Fri, Sep 2, 2022 at 10:07 AM Mathias Hauser @.***> wrote:

CAUTION: This email originated from outside of the University System. Do not click links or open attachments unless you recognize the sender and know the content is safe.

CAUTION: This email originated from outside of the University System. Do not click links or open attachments unless you recognize the sender and know the content is safe.

Jup, that's always the tradeoff - #1613 https://github.com/pydata/xarray/issues/1613 discusses a similar case.

— Reply to this email directly, view it on GitHub https://github.com/pydata/xarray/issues/6976#issuecomment-1235549943, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADBZR25KUX6CEF5ESPUJLKLV4ICYRANCNFSM6AAAAAAQCNUNFM . You are receiving this because you authored the thread.Message ID: @.***>

Jup, that's always the tradeoff - #1613 discusses a similar case.

Xarray passes the label indexers to the underlying pandas index:

```python import pandas as pd

"x" coordinate index

idx = pd.Index([2, 1, 0, 3, 5])

da.sel(x=slice(2, 3)) does this:

idx.slice_indexer(2, 3)

which returns slice(0, 4, None)

da.sel(x=[2, 3]) does this:

idx.get_indexer([2, 3])

which returns array([0, 3])

```

What might be possible is to warn if selecting with a slice and the index is not monotonic increasing or decreasing.

Is it always desirable? Asked differently, are there cases where one intentionally wants to select with a slice a non monotonic index? If yes, a warning might be annoying.

Maybe this could be clarified in the docs too?

So is this an expected behavior? I can work around it by explicitly creating the indices with arange() or the like. I do wonder if this is what is causing to_zarr() to fail even with compute=False? But I can work around that.

A smaller repro:

```python import numpy as np import xarray as xr

xr.DataArray(np.arange(5), dims="x", coords={"x": [2, 1, 0, 3, 5]})

da.sel(x=slice(2, 3))

```

Returns: <xarray.DataArray (x: 4)> array([0, 1, 2, 3]) Coordinates: * x (x) int64 2 4 5 3

python da.sel(x=[2, 3]) Returns <xarray.DataArray (x: 2)> array([0, 3]) Coordinates: * x (x) int64 2 3

Yes, good point. What might be possible is to warn if selecting with a slice and the index is not monotonic increasing or decreasing.