Let's get all the changes we've done so far on the main branch. Are we ready?

.github/pull_request_template.md

@@ -1,9 +1,12 @@
 <!-- Thank you for opening a PR! Please conform to the checklist below. If your PR is
 not completely ready yet, please open a "draft" PR. -->
 
-## PR checklist
+<details><summary>PR checklist</summary>
 
 - [ ] Has a descriptive title
 - [ ] Has a descriptive body
+- [ ] Includes a link to any issues closed by this PR, e.g. `Closes #5`.
 - [ ] Passes checks (to auto-fix a pre-commit.ci failure, add a comment
       containing "pre-commit.ci autofix")
+
+</details>

.github/workflows/ci.yml

@@ -0,0 +1,64 @@
+name: Continuous Integration
+
+on:
+  workflow_dispatch:
+  pull_request:
+  push:
+    branches:
+      - main
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+env:
+  # Many color libraries just need this to be set to any value, but at least
+  # one distinguishes color depth, where "3" -> "256-bit color".
+  FORCE_COLOR: 3
+
+defaults:
+  run:
+    # Enables steps to see the micromamba environment:
+    shell: bash -leo pipefail {0}
+
+
+jobs:
+  checks:
+    name: Check Python ${{ matrix.python-version }} on ${{ matrix.runs-on }}
+    runs-on: ${{ matrix.runs-on }}
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.8"]
+        runs-on:
+          - ubuntu-latest
+          - macos-latest
+          # - windows-latest
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - uses: mamba-org/setup-micromamba@v1
+        with:
+          micromamba-version: '1.5.6-0' # any version from https://github.com/mamba-org/micromamba-releases
+          environment-file: environment.yml
+          init-shell: >-
+            bash
+            powershell
+          cache-environment: true
+          post-cleanup: 'all'
+
+      - name: Type-check package
+        run: mypy
+
+      # - name: Test package
+      #   run: >-
+      #     python -m pytest -ra --cov --cov-report=xml --cov-report=term
+      #     --durations=20
+
+      # - name: Upload coverage report
+      #   uses: codecov/[email protected]
+      #   with:
+      #     token: ${{ secrets.CODECOV_TOKEN }}

Tb/nsidc-0080/.gitignore

@@ -2,3 +2,5 @@
 *
 # Except this file
 !.gitignore
+*.nc
+*.bin

antarctica_today/compute_mean_climatology.py

@@ -113,7 +113,7 @@ def compute_daily_climatology_pixel_averages(
     # Generate an empty MxNxT array with
     average_melt_array = numpy.zeros(
         (melt_array.shape[0], melt_array.shape[1], len(baseline_filler_dt_list)),
-        dtype=numpy.float,
+        dtype=float,
     )
 
     # Now, compute the average odds (0-1) of melt on any given day for any given pixel, over the baseline period.
@@ -123,7 +123,7 @@ def compute_daily_climatology_pixel_averages(
                 ((dt.month == bdt.month) and (dt.day == bdt.day))
                 for dt in dt_list_melt_season
             ],
-            dtype=numpy.bool,
+            dtype=bool,
         )
         # # print ("\t", [dt for i,dt in enumerate(dt_list_melt_season) if bdt_day_mask[i]])
         # if numpy.count_nonzero(bdt_day_mask) == 0:
@@ -286,9 +286,7 @@ def create_baseline_climatology_tif(
     num_years = int((end_date - start_date).days / 365.25)
     # print(num_years)
 
-    annual_sum_grids = numpy.empty(
-        model_array.shape[0:2] + (num_years,), dtype=numpy.int
-    )
+    annual_sum_grids = numpy.empty(model_array.shape[0:2] + (num_years,), dtype=int)
 
     if gap_filled:
         model_melt_days = model_array
@@ -308,7 +306,7 @@ def create_baseline_climatology_tif(
         # print(i, dt1, dt2)
 
         dates_mask = numpy.array(
-            [(dt >= dt1) & (dt <= dt2) for dt in datetimes], dtype=numpy.bool
+            [(dt >= dt1) & (dt <= dt2) for dt in datetimes], dtype=bool
         )
 
         # dt1_i = datetimes.index(dt1)
@@ -320,8 +318,8 @@ def create_baseline_climatology_tif(
     annual_std_array = numpy.std(annual_sum_grids, axis=2, dtype=numpy.float32)
 
     if round_to_integers:
-        annual_mean_array = numpy.array(numpy.round(annual_mean_array), dtype=numpy.int)
-        annual_std_array = numpy.array(numpy.round(annual_std_array), dtype=numpy.int)
+        annual_mean_array = numpy.array(numpy.round(annual_mean_array), dtype=int)
+        annual_std_array = numpy.array(numpy.round(annual_std_array), dtype=int)
 
     annual_mean_array[(ice_mask == 0)] = -1
     annual_std_array[(ice_mask == 0)] = -1
@@ -382,7 +380,7 @@ def create_partial_year_melt_anomaly_tif(
             ((dt >= first_dt_of_present_melt_season) and (dt <= current_datetime))
             for dt in dt_list
         ],
-        dtype=numpy.bool,
+        dtype=bool,
     )
     dts_masked = [dt for dt, mask_val in zip(dt_list, dt_mask) if mask_val]
 
@@ -492,7 +490,7 @@ def read_annual_melt_anomaly_tif(
     if verbose:
         print("Reading", anomaly_tif)
 
-    ds = gdal.Open(anomaly_tif, gdal.GA_ReadOnly)
+    ds = gdal.Open(str(anomaly_tif), gdal.GA_ReadOnly)
     if ds is None:
         return None
 
@@ -514,7 +512,7 @@ def get_baseline_climatology_array(fname=None, gap_filled=True):
         fname = base + "_gap_filled" + ext
 
     if os.path.exists(fname):
-        ds = gdal.Open(fname, gdal.GA_ReadOnly)
+        ds = gdal.Open(str(fname), gdal.GA_ReadOnly)
         if ds is None:
             raise FileNotFoundError("Could not open {0}".format(fname))
         array = ds.GetRasterBand(1).ReadAsArray()
@@ -592,7 +590,7 @@ def create_annual_melt_sum_tif(
 
         dates_mask = numpy.array(
             [((dt >= start_date) and (dt <= end_date)) for dt in dt_list],
-            dtype=numpy.bool,
+            dtype=bool,
         )
 
         # Skip any years for which there is no data.
@@ -602,13 +600,13 @@ def create_annual_melt_sum_tif(
         if gap_filled:
             # First add the floating-point values
             melt_array_year = numpy.sum(
-                melt_array[:, :, dates_mask], axis=2, dtype=numpy.float
+                melt_array[:, :, dates_mask], axis=2, dtype=float
             )
             # Round to integers
-            melt_array_year = numpy.array(melt_array_year.round(), dtype=numpy.int)
+            melt_array_year = numpy.array(melt_array_year.round(), dtype=int)
         else:
             melt_array_year = numpy.sum(
-                (melt_array[:, :, dates_mask] == 2), axis=2, dtype=numpy.int
+                (melt_array[:, :, dates_mask] == 2), axis=2, dtype=int
             )
 
         melt_array_year[ice_mask == 0] = -1
@@ -838,7 +836,7 @@ def _get_regional_tif_masks(tifname=antarctic_regions_tif):
     Regions will be 1..7. masks will each be an MxN array of (0,1) values.
     """
     # Read the ice mask tif, return the array.
-    regions_ds = gdal.Open(tifname, gdal.GA_ReadOnly)
+    regions_ds = gdal.Open(str(tifname), gdal.GA_ReadOnly)
     region_array = regions_ds.GetRasterBand(1).ReadAsArray()
     ndv = regions_ds.GetRasterBand(1).GetNoDataValue()
     unique_vals = [val for val in numpy.unique(region_array) if val != ndv]

antarctica_today/constants/grid.py

@@ -0,0 +1,10 @@
+from typing import Final
+
+# If you are using Arctic data or some other grid, change the DEFAULT_GRID_SHAPE below,
+# or just use the optional parameter when you call it.
+#
+# * For Antarctica, (rows, cols) = (332,316)
+# * For Arctic, (rows, cols) = (448, 304)
+#
+# Source: https://nsidc.org/data/polar-stereo/ps_grids.html
+DEFAULT_GRID_SHAPE: Final = (332, 316)

antarctica_today/constants/paths.py

@@ -0,0 +1,37 @@
+import os
+from pathlib import Path
+from typing import Final
+
+_this_dir: Final = Path(__file__).parent
+
+# The directory containing the main Python package:
+PACKAGE_DIR: Final = _this_dir.parent
+
+# The root of the Git repository (directory containing `.git`)
+REPO_DIR: Final = PACKAGE_DIR.parent
+
+# New data directories:
+# - Keep data included in the repo separate from runtime data. We don't want to require
+#   a specific directory structure that may not work on every computer. For example, on
+#   NSIDC VMs, we have limited direct storage, and need to use mounts to access larger
+#   storage devices.
+# - Use environment variables to enable override; in this case I think we only need one
+#   for the root storage directory. Default to an in-repo storage location so if the
+#   envvars are not populated, system pollution doesn't occur.
+# - Migrate more things iteratively :)
+_default_storage_dir = REPO_DIR
+STORAGE_DIR: Final = Path(
+    os.environ.get("ANTARCTICA_TODAY_STORAGE_DIR", _default_storage_dir)
+)
+DATA_DATABASE_DIR: Final = STORAGE_DIR / "database"
+# DATA_OUTPUT_DIR: Final = STORAGE_DIR / "output"
+
+# Legacy data directories
+DATA_DIR: Final = REPO_DIR / "data"
+DATA_QGIS_DIR: Final = REPO_DIR / "qgis"
+DATA_TB_DIR: Final = REPO_DIR / "Tb"
+DATA_PLOTS_DIR: Final = REPO_DIR / "plots"
+DATA_BASELINE_DATASETS_DIR: Final = REPO_DIR / "baseline_datasets"
+
+
+# TODO: Bring tb_file_data.py into constants