clawhdf5-netcdf4: an unlimited dimension reports its current length

Dimension::size of an unlimited dimension was its dimension scale's extent, which netCDF-C leaves at 0, so it read 0 for a dimension holding records. It is now what netCDF-C reports (nc4_find_dim_len): the largest current extent of the variables using it in any group, found through the scale's REFERENCE_LIST, a coordinate variable's own extent included; 0 when nothing has been written. interop_tests::unlimited_dimension_lengths_match_netcdf4_python compares with netCDF4-python (variables of different lengths, one in a subgroup, an unwritten dimension, a coordinate variable shorter than another variable on its dimension, a subgroup's own dimension); before the fix it got time 0/6, rec 3/5, srec 0/1. The known-issues entry moves to Fixed; the crate README's warning goes. A new open entry records a related bug found meanwhile: variables' dimensions are matched by size, not DIMENSION_LIST. Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
2026-09-28 21:24:59 -05:00
parent 3eca5d8334
commit 31efac1ae1
5 changed files with 236 additions and 21 deletions
@@ -377,3 +377,87 @@ ds.close()
    let names = file.variable("name").unwrap().read_string().unwrap();
    assert_eq!(names, vec!["Oslo", "", "São Paulo", "x"]);
 }
+
+// ===========================================================================
+// Unlimited dimensions: the length netCDF-C reports
+// ===========================================================================
+
+/// An unlimited dimension's length is the largest extent of the variables
+/// using it, in any group (netCDF-C's `nc4_find_dim_len`), not its dimension
+/// scale's extent (which netCDF-C leaves at 0): variables of different
+/// lengths, one in a subgroup, a dimension no variable has written, a
+/// coordinate variable, a subgroup's own unlimited dimension. Compared with
+/// what netCDF4-python reports for the same file.
+#[test]
+fn unlimited_dimension_lengths_match_netcdf4_python() {
+    skip_if_no_netcdf4!();
+    let dir = tempfile::tempdir().unwrap();
+    let path = dir.path().join("unlimited.nc");
+    let script = format!(
+        r#"
+import netCDF4 as nc
+import numpy as np
+with nc.Dataset({path:?}, "w", format="NETCDF4") as f:
+    f.createDimension("time", None)
+    f.createDimension("empty", None)
+    f.createDimension("rec", None)
+    f.createDimension("x", 3)
+    f.createVariable("t", "f8", ("time", "x"))[0:2, :] = np.ones((2, 3))
+    f.createVariable("a", "i4", ("time",))[0:4] = np.arange(4)
+    f.createVariable("e", "i4", ("empty",))
+    f.createVariable("rec", "f4", ("rec",))[0:3] = [1, 2, 3]
+    f.createVariable("r", "f4", ("x", "rec"))[:, 0:5] = np.ones((3, 5))
+    g = f.createGroup("sub")
+    g.createVariable("c", "i4", ("time",))[0:6] = np.arange(6)
+    g.createDimension("srec", None)
+    g.createVariable("s", "i4", ("srec", "x"))[0:1, :] = np.ones((1, 3))
+with nc.Dataset({path:?}) as f:
+    for grp in (f, f.groups["sub"]):
+        for name, d in grp.dimensions.items():
+            print(grp.path, name, len(d), d.isunlimited())
+"#,
+        path = path.display().to_string()
+    );
+    let out = Command::new(python())
+        .args(["-c", &script])
+        .output()
+        .expect("failed to run python3");
+    assert!(
+        out.status.success(),
+        "{}",
+        String::from_utf8_lossy(&out.stderr)
+    );
+    let expected: Vec<String> = String::from_utf8(out.stdout)
+        .unwrap()
+        .lines()
+        .map(str::to_string)
+        .collect();
+    // time: t has 2 records, a 4 and sub/c 6; rec: the coordinate variable
+    // has 3, r 5; empty: nothing written.
+    for want in [
+        "/ time 6 True",
+        "/ empty 0 True",
+        "/ rec 5 True",
+        "/ x 3 False",
+        "/sub srec 1 True",
+    ] {
+        assert!(
+            expected.iter().any(|l| l == want),
+            "netCDF4 reports {expected:?}"
+        );
+    }
+
+    let file = NetCDF4File::open(&path).unwrap();
+    let sub = file.group("sub").unwrap();
+    let mut got = Vec::new();
+    for (path, dims) in [
+        ("/", file.dimensions().unwrap()),
+        ("/sub", sub.dimensions().unwrap()),
+    ] {
+        for d in dims {
+            let unlimited = if d.is_unlimited { "True" } else { "False" };
+            got.push(format!("{path} {} {} {unlimited}", d.name, d.size));
+        }
+    }
+    assert_eq!(got, expected);
+}