docs/references.bib (1)

50-59: Nice addition; consider adding DOI/URL metadata.

Adding a DOI (and optionally the URL) improves reproducibility and doc builds that auto-link citations.

dascore/core/patch.py (1)

256-256: Group transform aliases with the other transform methods.

For consistency/readability, place this alongside the other transform method mappings (lines 346+). Also prefer transform.radians_to_strain to match nearby aliases.

Apply this diff:

-    radians_to_strain = dascore.transform.radians_to_strain

Add near the other transform mappings:

@@
     tau_p = transform.tau_p
+    radians_to_strain = transform.radians_to_strain

tests/test_transform/test_strain.py (1)

195-197: Guard against divide-by-zero in ratio check.

Random data may contain zeros; mask to avoid spurious inf/NaN.

Apply this diff (in addition to the above):

-        const = out.data / rad_patch.data
-        assert np.allclose(const, expected_const)
+        mask = rad_patch.data != 0
+        const = out.data[mask] / rad_patch.data[mask]
+        assert np.allclose(const, expected_const)

dascore/transform/__init__.py (1)

13-16: Fix linter issues: wrap long import and declare re-exports.

Addresses E501 and F401 by line-wrapping and making the re-exports explicit.

Apply this diff:

-from .strain import velocity_to_strain_rate, velocity_to_strain_rate_edgeless, radians_to_strain
+from .strain import (
+    velocity_to_strain_rate,
+    velocity_to_strain_rate_edgeless,
+    radians_to_strain,
+)
+
+# Re-exported API
+__all__ = [
+    "differentiate",
+    "rfft",
+    "dft",
+    "idft",
+    "integrate",
+    "spectrogram",
+    "dispersion_phase_shift",
+    "tau_p",
+    "velocity_to_strain_rate",
+    "velocity_to_strain_rate_edgeless",
+    "radians_to_strain",
+]

dascore/transform/strain.py (2)

214-218: Docstring: clarify “strain (rate)” → “strain”.

Equation 3 converts phase (radians) to strain, not strain rate.

Apply this diff:

-    Convert data in radians to strain (rate).
+    Convert data in radians to strain.

---

260-263: Numerical constant matches Eq. 3; consider explicit typing for gauge_length.

Optional: annotate gauge_length: float | None for clarity.

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

💡 Knowledge Base configuration:

• MCP integration is disabled by default for public repositories
• Jira integration is disabled by default for public repositories
• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

tests/test_transform/test_strain.py (1)

188-197: Replace magic number with formula-derived constant (Eq. 3).

Hard-coding 11.6e-9 is brittle; compute from the same defaults the implementation uses.

-        # Ensure the constant published in @lindsey2020broadband eq 3 works
-        # because the test data also have 10m gauge length.
-        expected_const = 11.6e-9
-        const = out.data / rad_patch.data
-        assert np.allclose(const, expected_const)
+        # Compute the expected constant from Eq. 3 using defaults and GL=10 m.
+        wl = 1550.0e-9
+        n = 1.445
+        zeta = 0.79
+        Lg = rad_patch.attrs.gauge_length
+        expected_const = wl / (4 * np.pi * n * Lg * zeta)
+        const = out.data / rad_patch.data
+        assert np.allclose(const, expected_const)

dascore/transform/strain.py (1)

246-254: Radian presence check is brittle and may miss aliases; also fails on non-str inputs.

Prefer a lowercase substring check on the original units string; accept “rad”, “radian(s)”.

-    # If units doesn't contain radians just return so function is idempotent
-    quant = dc.get_quantity(patch.attrs.data_units)
-    if str(dc.get_unit("radians")) not in str(quant):
+    # If units don't contain any radian component, return idempotently.
+    units_str = str(patch.attrs.get("data_units", "")).lower()
+    if "rad" not in units_str:
         msg = (
             f"Patch {patch} has no radians in its data_units, "
             f"skipping strain conversion."
         )
         warnings.warn(msg)
         return patch

tests/test_transform/test_strain.py (3)

175-187: Fixture is fine; consider modern RNG API.

Use numpy’s Generator for forward-compatibility and reproducibility.

-        rand = np.random.RandomState(42)
-        data = rand.random(random_patch.data.shape) - 0.5
+        rng = np.random.default_rng(42)
+        data = rng.random(random_patch.data.shape) - 0.5

---

199-208: Strengthen positive path check.

After the ParameterError case, also verify scaling is correct for an explicit gauge_length.

         out = patch.radians_to_strain(gauge_length=1)
         assert isinstance(out, dc.Patch)
+        wl, n, zeta = 1550.0e-9, 1.445, 0.79
+        expected = wl / (4 * np.pi * n * 1 * zeta)
+        assert np.allclose(out.data / rad_patch.data, expected)

---

209-217: Idempotence assertion is good; optionally also assert no history appended.

Ensures zero-op transform truly leaves Patch untouched.

         assert out is patch
+        assert list(getattr(out.attrs, "history", ())) == list(
+            getattr(patch.attrs, "history", ())
+        )

dascore/transform/strain.py (2)

214-215: Doc: this converts to strain, not strain-rate.

Fix the wording.

-    Convert data in radians to strain (rate).
+    Convert data in radians to strain.

---

266-268: Consider setting data_type="strain" on output.

Aligns attrs with the unit change.

-    new_attrs = patch.attrs.update(data_units=new_units)
+    new_attrs = patch.attrs.update(data_units=new_units, data_type="strain")

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

💡 Knowledge Base configuration:

• MCP integration is disabled by default for public repositories
• Jira integration is disabled by default for public repositories
• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.