Merge pull request #74 from ucgmsim/no_compress_broadband

Do not compress broadband waveforms

Author: lispandfound

tests/test_realisation.py

@@ -567,6 +567,35 @@ def test_seeds() -> None:
     )
 
 
+def test_seeds_from_random(tmp_path: Path) -> None:
+    """Test read from realisation or random"""
+    realisation_file = tmp_path / "test_realisation.json"
+    # Create a file with unrelated content
+    initial_content = {"other_key": "some_value"}
+    with open(realisation_file, "w") as f:
+        json.dump(initial_content, f)
+
+    seeds = realisations.Seeds.read_from_realisation_or_random(realisation_file)
+    assert all(
+        0 <= seed <= 2 ** (struct.Struct("i").size * 8 - 1) - 1
+        for seed in seeds.to_dict().values()
+    )
+
+    realisation_file = tmp_path / "test_realisation_1.json"
+    # Create a file with unrelated content
+    seeds = realisations.Seeds(
+        nshm_to_realisation_seed=0,
+        rupture_propagation_seed=0,
+        genslip_seed=0,
+        srfgen_seed=0,
+        hf_seed=0,
+    )
+    seeds.write_to_realisation(realisation_file)
+
+    seeds_read = realisations.Seeds.read_from_realisation_or_random(realisation_file)
+    assert seeds == seeds_read
+
+
 def test_velocity_model_1d(tmp_path: Path) -> None:
     velocity_model_1d = realisations.VelocityModel1D(
         model=pd.DataFrame(

workflow/realisations.py

@@ -242,8 +242,8 @@ class Seeds(RealisationConfiguration):
     """HF seed."""
 
     @classmethod
-    def read_from_realisation_or_defaults(
-        cls, realisation_ffp: Path, *args: list[Any]
+    def read_from_realisation_or_random(
+        cls, realisation_ffp: Path
     ) -> Self:  # *args is to maintain compat with superclass (remove this and see the error in mypy).
         """Read seeds configuration from a realisation file or generate random seeds if not present.
 
@@ -256,8 +256,6 @@ def read_from_realisation_or_defaults(
         ----------
         realisation_ffp : Path
             The realisation filepath to read from.
-        *args : list
-            Ignored arguments.
 
         Returns
         -------

workflow/scripts/bb_sim.py

@@ -154,9 +154,11 @@ def combine_hf_and_lf(
     )
     hf = hf.sel(station=common_stations)
     lf = lf.sel(station=common_stations)
-
     vs30_df = pd.read_csv(
-        station_vs30_ffp, sep=r"\s+", header=None, names=["station", "vsite"]
+        station_vs30_ffp,
+        sep=r"\s+",
+        header=None,
+        names=["station", "vsite"],  # type: ignore[invalid-argument-type]
     ).set_index("station")
     vs30_df["vsite"] = vs30_df["vsite"].astype(np.float32)
     vs30_df = vs30_df.loc[common_stations]
@@ -218,8 +220,6 @@ def combine_hf_and_lf(
         engine="h5netcdf",
         encoding={
             "waveform": {
-                "compression": "zlib",  # Use zlib compression.
-                "complevel": 5,  # Compress to level 5 (of 9).
                 "fletcher32": True,  # Add Fletcher-32 checksums for long-term storage.
             }
         },

workflow/scripts/check_srf.py

@@ -83,7 +83,11 @@ def check_srf(
     )
     mu = velocity_model["mu"].iloc[indices].values
     srf_magnitude = mag_scaling.mom2mag(
-        (srf_file.points["area"].values * srf_file.points["slip"].values * mu).sum()
+        (
+            np.array(srf_file.points["area"].values)
+            * np.array(srf_file.points["slip"].values)
+            * mu
+        ).sum()
     )
     logger = log_utils.get_logger("__name__")
 

workflow/scripts/gcmt_to_realisation.py

@@ -161,17 +161,24 @@ def gcmt_to_realisation(
     ].set_index("PublicID")
 
     if gcmt_event_id in gcmt_solutions.index:
-        solution = gcmt_solutions.loc[gcmt_event_id]
-        latitude = solution["Latitude"]
-        longitude = solution["Longitude"]
-        centroid_depth = solution["CD"]
-        magnitude = solution["Mw"]
-        nodal_plane_1 = NodalPlane(
-            solution["strike1"], solution["dip1"], solution["rake1"]
-        )
-        nodal_plane_2 = NodalPlane(
-            solution["strike2"], solution["dip2"], solution["rake2"]
-        )
+        latitude = gcmt_solutions.at[gcmt_event_id, "Latitude"]
+        longitude = gcmt_solutions.at[gcmt_event_id, "Longitude"]
+        centroid_depth = gcmt_solutions.at[gcmt_event_id, "CD"]
+        magnitude = gcmt_solutions.at[gcmt_event_id, "Mw"]
+        strike1 = gcmt_solutions.at[gcmt_event_id, "strike1"]
+        dip1 = gcmt_solutions.at[gcmt_event_id, "dip1"]
+        rake1 = gcmt_solutions.at[gcmt_event_id, "rake1"]
+        strike2 = gcmt_solutions.at[gcmt_event_id, "strike2"]
+        dip2 = gcmt_solutions.at[gcmt_event_id, "dip2"]
+        rake2 = gcmt_solutions.at[gcmt_event_id, "rake2"]
+        assert isinstance(strike1, float | int)
+        assert isinstance(dip1, float | int)
+        assert isinstance(rake1, float | int)
+        assert isinstance(strike2, float | int)
+        assert isinstance(dip2, float | int)
+        assert isinstance(rake2, float | int)
+        nodal_plane_1 = NodalPlane(strike1, dip1, rake1)
+        nodal_plane_2 = NodalPlane(strike2, rake2, dip2)
     elif gcmt_event_id in automated_gcmt_solutions:
         solution = automated_gcmt_solutions[gcmt_event_id]
         latitude = solution["location"]["latitude"]
@@ -203,6 +210,7 @@ def gcmt_to_realisation(
     # Calculate dip direction from strike (strike + 90 degrees for right-hand rule)
     dip_direction = (selected_nodal_plane.strike + 90) % 360
 
+    assert isinstance(magnitude, float)
     length, width = magnitude_scaling.magnitude_to_length_width(
         scaling_relation, magnitude, rake
     )
@@ -217,6 +225,7 @@ def gcmt_to_realisation(
         length_m = length_km * 1000  # Convert km to meters
         width_m = width_km * 1000  # Convert km to meters
 
+        assert isinstance(centroid_depth, float)
         source_geometry = sources.Point.from_lat_lon_depth(
             point_coordinates=np.array(
                 [latitude, longitude, centroid_depth * 1000]

workflow/scripts/generate_rupture_propagation.py

@@ -81,7 +81,7 @@ def generate_rupture_propagation(
     strategy : RuptureStrategy
         The rupture propagation strategy to use. Default is `RuptureStrategy.RANDOM`.
     """
-    seeds = realisations.Seeds.read_from_realisation_or_defaults(realisation_ffp)
+    seeds = realisations.Seeds.read_from_realisation_or_random(realisation_ffp)
     source_config = realisations.SourceConfig.read_from_realisation(realisation_ffp)
     faults = source_config.source_geometries
 

workflow/scripts/generate_station_coordinates.py

@@ -75,7 +75,10 @@ def generate_fd_files(
 
     # retrieve in station names, latitudes and longitudes
     stations = pd.read_csv(
-        stat_file, delimiter=r"\s+", comment="#", names=["lon", "lat", "name"]
+        stat_file,
+        delimiter=r"\s+",
+        comment="#",
+        names=["lon", "lat", "name"],  # type: ignore[invalid-argument-type]
     )
 
     x, y = proj(lat=stations["lat"].values, lon=stations["lon"].values).T

workflow/scripts/generate_velocity_model_parameters.py

@@ -387,10 +387,12 @@ def pgv_target(
             for magnitude in magnitudes.magnitudes.values()
         )
     )
-    return np.interp(
-        total_magnitude,
-        velocity_model_parameters.pgv_interpolants[:, 0],
-        velocity_model_parameters.pgv_interpolants[:, 1],
+    return float(
+        np.interp(
+            total_magnitude,
+            velocity_model_parameters.pgv_interpolants[:, 0],
+            velocity_model_parameters.pgv_interpolants[:, 1],
+        )
     )
 
 

workflow/scripts/hf_sim.py

@@ -279,7 +279,7 @@ def run_hf(
     None
         The function does not return any value. It writes the HF output directly to `out_file`.
     """
-    seeds = Seeds.read_from_realisation_or_defaults(realisation_ffp)
+    seeds = Seeds.read_from_realisation_or_random(realisation_ffp)
 
     domain_parameters = DomainParameters.read_from_realisation(realisation_ffp)
     metadata = RealisationMetadata.read_from_realisation(realisation_ffp)
@@ -294,7 +294,7 @@ def run_hf(
         station_file,
         delimiter=r"\s+",
         header=None,
-        names=["longitude", "latitude", "name"],
+        names=["longitude", "latitude", "name"],  # type: ignore[invalid-argument-type]
     ).set_index("name")
     station_hashes = np.array(
         [stable_hash(name) for name in stations.index], dtype=np.int32
@@ -306,7 +306,9 @@ def run_hf(
     stations["seed"] = np.int32(seeds.hf_seed) ^ station_hashes
     velocity_model_path = work_directory / "velocity_model"
     velocity_model.write_velocity_model(velocity_model_path)
-    nt = int(np.float32(domain_parameters.duration) / np.float32(hf_config.dt))  # Match Fortran's single-precision for consistent nt calculation
+    nt = int(
+        np.float32(domain_parameters.duration) / np.float32(hf_config.dt)
+    )  # Match Fortran's single-precision for consistent nt calculation
     waveform = np.empty((3, len(stations), nt), dtype=np.float32)
 
     hf_input_template = build_hf_input(

workflow/scripts/nshm2022_to_realisation.py

@@ -329,7 +329,7 @@ def generate_realisation(
         )
         raise typer.Exit(code=1)
     faults_info = db.get_rupture_fault_info(rupture_id)
-    seeds = Seeds.read_from_realisation_or_defaults(realisation_ffp)
+    seeds = Seeds.read_from_realisation_or_random(realisation_ffp)
     np.random.seed(seed=seeds.nshm_to_realisation_seed)
     source_config = SourceConfig(faults)