working version

.streamlit/config.toml

@@ -0,0 +1,9 @@
+[server]
+headless = true
+address = "0.0.0.0"
+port = 5000
+enableCORS = true
+enableXsrfProtection = true
+
+[browser]
+gatherUsageStats = false

Dockerfile

@@ -0,0 +1,11 @@
+FROM mambaorg/micromamba:2.0.4
+
+WORKDIR /app
+COPY ogs-env.yml /tmp/ogs-env.yml
+
+RUN micromamba env create -y -f /tmp/ogs-env.yml && \
+    micromamba clean --all --yes
+
+COPY . /app
+EXPOSE 5000
+CMD ["micromamba", "run", "-n", "ogs_env", "streamlit", "run", "rom.py", "--server.address=0.0.0.0", "--server.port=5000"]

fetch_data.py

@@ -1,25 +1,51 @@
+import os
 import copernicusmarine
 import numpy as np
 from datetime import datetime, timedelta
 import netCDF4 as nc
 from scipy.interpolate import RegularGridInterpolator, NearestNDInterpolator
-import os
-import matplotlib.pyplot as plt
 from scipy.ndimage import gaussian_filter
-masks_red=np.load("masks_red.npy")
-masks=np.load("masks.npy")
-
+import matplotlib.pyplot as plt
+from pathlib import Path
+import shutil
+
+BASE_DIR = Path(__file__).resolve().parent
+REQUIRED_DIR = BASE_DIR / "required_data"
+DAILY_DIR = BASE_DIR / "daily_data"
+BACKUP_DIR = BASE_DIR / "daily_backup"
+
+DAILY_DIR.mkdir(exist_ok=True)
+BACKUP_DIR.mkdir(exist_ok=True)
+
+def clear_directory(directory: Path) -> None:
+    for item in directory.iterdir():
+        if item.is_dir():
+            shutil.rmtree(item)
+        else:
+            item.unlink()
+
+existing_daily_files = list(DAILY_DIR.glob("*"))
+if existing_daily_files:
+    clear_directory(BACKUP_DIR)
+
+    for src in existing_daily_files:
+        dst = BACKUP_DIR / src.name
+        if src.is_dir():
+            shutil.copytree(src, dst)
+        else:
+            shutil.copy2(src, dst)
+
+masks_red=np.load(REQUIRED_DIR / "masks_red.npy")
+masks=np.load(REQUIRED_DIR / "masks.npy")
 print(masks_red.shape)
-latitude_red=np.load("latitude_red.npy")
-longitude_red=np.load("longitude_red.npy")
-
+latitude_red=np.load(REQUIRED_DIR / "latitude_red.npy")
+longitude_red=np.load(REQUIRED_DIR / "longitude_red.npy")
 
 yesterday = datetime.today() - timedelta(days=1)
 formatted_date = yesterday.strftime('%Y-%m-%d')
 
 depths=[0,10,30,60,100]
 
-
 c=np.concatenate([x.reshape(latitude_red.shape[0],longitude_red.shape[0],1) for x in np.meshgrid(latitude_red,longitude_red,indexing="ij")],axis=2)
 
 c=c.reshape(-1,2)
@@ -34,60 +60,95 @@
 
 }
 
-if not os.path.exists("depths.nc"):
+depths_nc_path = DAILY_DIR / "depths.nc"
+if not depths_nc_path.exists():
     copernicusmarine.subset(
             dataset_id="cmems_mod_med_phy_anfc_4.2km_static",
             minimum_longitude=11,
             maximum_longitude=18,
             minimum_latitude=42,
             maximum_latitude=47,
+            minimum_depth=0,
+            maximum_depth=120,
             start_datetime=formatted_date,
             end_datetime=formatted_date,
-            output_filename = "depths.nc",
+            output_filename=str(depths_nc_path),
             force_download=True
-            )
+        )
 
-d_dataset=nc.Dataset("depths.nc").variables["depth"][:]
+d_dataset = nc.Dataset(str(depths_nc_path)).variables["depth"][:]
 
+vars_list =["Chla","N1p","N3n","S","T"]
 
+for i, depth in enumerate(depths):
+    for var in vars_list:
+        depth_tmp = d_dataset[np.argmin(d_dataset - depth)]
 
-vars=["Chla","N1p","N3n","S","T"]
+        nc_filename = DAILY_DIR / f"{var}_{depth}.nc"
+        npy_filename = DAILY_DIR / f"{var}_{depth}.npy"
+        png_filename = DAILY_DIR / f"miao_{var}_{depth}.png"
 
-for i in range(len(depths)):
-    for var in vars:
-        depth=depths[i]
-        depth_tmp=d_dataset[np.argmin(d_dataset-depth)]
-        if not os.path.exists("{}_{}.nc".format(var,depth)):
+        if not nc_filename.exists():
             copernicusmarine.subset(
-            dataset_id=data[var][1],
-            variables=[data[var][0]],
-            minimum_longitude=11,
-            maximum_longitude=18,
-            minimum_latitude=42,
-            maximum_latitude=47,
-            start_datetime=formatted_date,
-            end_datetime=formatted_date,
-            minimum_depth=depth_tmp,
-            maximum_depth=depth_tmp,
-            output_filename = "{}_{}.nc".format(var,depth),
-            force_download=True
+                dataset_id=data[var][1],
+                variables=[data[var][0]],
+                minimum_longitude=11,
+                maximum_longitude=18,
+                minimum_latitude=42,
+                maximum_latitude=47,
+                start_datetime=formatted_date,
+                end_datetime=formatted_date,
+                minimum_depth=depth_tmp,
+                maximum_depth=depth_tmp,
+                output_filename=str(nc_filename),
+                force_download=True,
             )
-        a=nc.Dataset("{}_{}.nc".format(var,depth))
-        longitude=a.variables["longitude"][:].data
-        latitude=a.variables["latitude"][:].data
-        tmp=a.variables[data[var][0]][0,0].data
-        tmp[tmp>1e+10]=np.nan
-        rg=RegularGridInterpolator([latitude,longitude],tmp,method="nearest",fill_value=np.nan)
-        miao=rg(c).reshape(latitude_red.shape[0],longitude_red.shape[0])
+
+        a = nc.Dataset(str(nc_filename))
+        longitude = a.variables["longitude"][:].data
+        latitude = a.variables["latitude"][:].data
+        tmp = a.variables[data[var][0]][0, 0].data
+        a.close()
+
+        tmp[tmp > 1e10] = np.nan
+
+        rg = RegularGridInterpolator(
+            [latitude, longitude],
+            tmp,
+            method="nearest",
+            fill_value=np.nan,
+        )
+        miao = rg(c).reshape(latitude_red.shape[0], longitude_red.shape[0])
+
         valid_mask = ~np.isnan(miao)
-        valid_pts   = c[valid_mask.ravel()]
-        valid_vals  = miao[valid_mask]
+        valid_pts = c[valid_mask.ravel()]
+        valid_vals = miao[valid_mask]
         nn = NearestNDInterpolator(valid_pts, valid_vals)
-        miao = nn(c).reshape(latitude_red.shape[0],longitude_red.shape[0])
+        miao = nn(c).reshape(latitude_red.shape[0], longitude_red.shape[0])
+
         miao = gaussian_filter(miao, sigma=1.5, mode='nearest')
         miao = gaussian_filter(miao, sigma=1.5, mode='nearest')
-        plt.imshow(np.ma.masked_array(miao,masks_red[i]))
-        plt.savefig("miao.png")
-        miao=miao.reshape(-1)[np.logical_not(masks_red[i]).reshape(-1)]
-        np.save("{}_{}.npy".format(var,depth),miao)
 
+        plt.imshow(np.ma.masked_array(miao, masks_red[i]))
+        plt.savefig(png_filename)
+        plt.close()
+
+        miao_masked_flat = miao.reshape(-1)[np.logical_not(masks_red[i]).reshape(-1)]
+        np.save(npy_filename, miao_masked_flat)
+
+expected_files = []
+expected_files.append(depths_nc_path)
+for depth in depths:
+    for var in vars_list:
+        expected_files.append(DAILY_DIR / f"{var}_{depth}.nc")
+        expected_files.append(DAILY_DIR / f"{var}_{depth}.npy")
+        expected_files.append(DAILY_DIR / f"miao_{var}_{depth}.png")
+
+for path in expected_files:
+    if not path.exists():
+        backup_path = BACKUP_DIR / path.name
+        if backup_path.exists():
+            shutil.copy2(backup_path, path)
+            print(f"[RECOVER] Restored {path.name} from backup.")
+        else:
+            print(f"[WARNING] Missing {path.name} in both daily_data and daily_backup.")