[pull] main from pydata:main

HOW_TO_RELEASE.md

@@ -28,7 +28,7 @@ upstream        https://github.com/pydata/xarray (push)
     Then run
 
     ```sh
-    pixi run release-contributors
+    pixi run -e release release-contributors
     ```
 
     and copy the output.

doc/api-hidden.rst

@@ -396,6 +396,7 @@
    plot.imshow
    plot.pcolormesh
    plot.scatter
+   plot.lines
    plot.surface
 
    CFTimeIndex.all

doc/api/plotting.rst

@@ -10,6 +10,7 @@ Dataset
    :toctree: ../generated/
    :template: autosummary/accessor_method.rst
 
+   Dataset.plot.lines
    Dataset.plot.scatter
    Dataset.plot.quiver
    Dataset.plot.streamplot
@@ -32,6 +33,7 @@ DataArray
    DataArray.plot.hist
    DataArray.plot.imshow
    DataArray.plot.line
+   DataArray.plot.lines
    DataArray.plot.pcolormesh
    DataArray.plot.step
    DataArray.plot.scatter

doc/user-guide/io.md

@@ -1244,6 +1244,35 @@ ds.to_zarr(
 The number of chunks on Tair matches our dask chunks, while there is now only a single
 chunk in the directory stores of each coordinate.
 
+(io.zarr.rectilinear-chunks)=
+
+### Variable-sized (rectilinear) chunks
+
+Zarr v3 supports _rectilinear_ chunk grids, where chunk sizes vary along one
+or more dimensions. This is useful when natural data boundaries (yearly
+chunks of a daily time series, per-tile spatial extents) don't align to a
+regular grid. Reading such arrays requires `zarr-python >= 3.2`, with the
+experimental feature enabled:
+
+```python
+import zarr
+
+with zarr.config.set({"array.rectilinear_chunks": True}):
+    roundtrip = xr.open_zarr("rectilinear.zarr", zarr_format=3)
+    roundtrip.chunks["x"]  # e.g. (10, 20, 30)
+```
+
+```{note}
+xarray can currently only *read* rectilinear-chunked Zarr V3 arrays, not
+write them. Rectilinear arrays must be created with `zarr-python` directly
+(or another tool, such as Icechunk) before being opened with `xr.open_zarr`.
+This also means a Dataset opened from a rectilinear-chunked store cannot be
+written back with {py:meth}`Dataset.to_zarr`, including writing to a
+`region` of the store -- both raise a `TypeError` naming the affected
+variable. Write support is tracked in
+[GH11279](https://github.com/pydata/xarray/pull/11279).
+```
+
 ### Groups
 
 Nested groups in zarr stores can be represented by loading the store as a

doc/user-guide/plotting.md

@@ -689,6 +689,59 @@ ds.plot.scatter(x="A", y="B", z="z", hue="y", markersize="x", row="x", col="w");
 For more advanced scatter plots, we recommend converting the relevant data variables
 to a pandas DataFrame and using the extensive plotting capabilities of `seaborn`.
 
+### Lines
+
+{py:func}`xarray.plot.lines` calls matplotlib.collections.LineCollection under the hood,
+allowing multiple lines being drawn efficiently. It uses similar arguments as
+{py:func}`xarray.plot.scatter`.
+
+Let's return to the air temperature dataset:
+
+```{code-cell}
+airtemps = xr.tutorial.open_dataset("air_temperature")
+air = airtemps.air - 273.15
+air.attrs = airtemps.air.attrs
+air.attrs["units"] = "deg C"
+
+air.isel(lon=10).plot.lines(x="time", hue="lat");
+```
+
+Make it a little more transparent:
+
+```{code-cell}
+air.isel(lon=10).plot.lines(x="time", hue="lat", alpha=0.2);
+```
+
+Zoom in a little on the x-axis, and compare a few latitudes and longitudes,
+group them using `hue` and `linewidth`. The `linewidth` kwarg works in
+a similar way as `markersize` kwarg for scatter plots, it lets you vary the
+line's size by variable value.
+
+```{code-cell}
+:tags: [remove-stderr]
+
+air_zoom = air.isel(time=slice(1200, 1500), lat=[5, 10, 15], lon=[10, 15])
+air_zoom.plot.lines(x="time", hue="lat", linewidth="lon", add_colorbar=False);
+```
+
+Lines can modify the linestyle but does not allow markers. Instead combine {py:func}`xarray.plot.lines`
+with {py:func}`xarray.plot.scatter`:
+
+```{code-cell}
+air.isel(lat=10, lon=10)[:200].plot.lines(x="time", color="k", linestyle="dashed")
+air.isel(lat=10, lon=10)[:200].plot.scatter(x="time", color="k", marker="^");
+```
+
+Switching to another dataset with more variables we can analyse in similar
+fashion as {py:func}`xarray.plot.scatter`:
+
+```{code-cell}
+:tags: [remove-stderr]
+
+ds = xr.tutorial.scatter_example_dataset(seed=42)
+ds.plot.lines(x="A", y="B", hue="y", linewidth="x", row="x", col="w");
+```
+
 ### Quiver
 
 Visualizing vector fields is supported with quiver plots: