gridded_data speed up and test publish

Author: oskarhurst

.github/workflows/publish-to-test-pypi.yml

@@ -1,4 +1,4 @@
-name: Deployment
+name: Test Deployment
 
 # The deployment workflow should only run when changes are merged into the `main` branch.
 # Since a branch protection rule prevents directly pushing to `main` this workflow will
@@ -11,6 +11,11 @@ on:
 
   # Allow the workflow to be manually triggered from the Actions tab.
   workflow_dispatch:
+    inputs:
+      tag_name:
+        description: 'If you want to run a release, type an existing or new tag (eg. v1.2.0)'
+        required: false
+        default: ''
 
 jobs:
   # Build and test the distribution before deploying to the various platforms. This allows
@@ -142,7 +147,9 @@ jobs:
     runs-on: ubuntu-latest
 
     # Require a tag push to publish a release distribution.
-    if: startsWith(github.ref, 'refs/tags/')
+    if: >
+      startsWith(github.ref, 'refs/tags/')
+      || (github.event_name == 'workflow_dispatch' && github.event.inputs.tag_name != '')
 
     # The distribution will only be published if the tests have passed.
     needs:

src/hecdss/gridded_data.py

@@ -1,6 +1,7 @@
 # import pandas as pd
 import numpy as np
 import math
+import time
 
 NULL_INT = -3.4028234663852886e+38
 
@@ -114,12 +115,12 @@ def range_limit_table(self, minval, maxval, range_, bins, datasize, data):
             self.rangeLimitTable[i] = minval + step * i
 
         self.rangeLimitTable[bins - 1] = maxval
-
         # Exceedance
-        for idx in range(datasize):
-            for jdx in range(bins):
-                if data[idx] >= self.rangeLimitTable[jdx]:
-                    self.numberEqualOrExceedingRangeLimit[jdx] += 1
+        sorted_data = np.sort(data)
+        n = len(sorted_data)
+        for jdx in range(bins):
+            idx = np.searchsorted(sorted_data, self.rangeLimitTable[jdx], side="left")
+            self.numberEqualOrExceedingRangeLimit[jdx] = n - idx
 
     def update_grid_info(self):
         """
@@ -135,7 +136,7 @@ def update_grid_info(self):
 
         self.data = np.nan_to_num(self.data, nan=NULL_INT)
         self.numberOfRanges = math.floor(1 + 3.322 * math.log10(n) + 1)
-        flat_data = self.data.flatten()
+        flat_data = self.data.ravel()
         self.range_limit_table(self.minDataValue, self.maxDataValue, bin_range, self.numberOfRanges, n, flat_data)
 
     @staticmethod