Merge pull request NCAS-CMS#700 from davidhassell/https-fix

davidhassell · web-flow · commit 972cdaadceec · 2023-10-09T17:49:00.000+01:00
Fix bug that prevented `https://` netCDF files from being read
diff --git a/Changelog.rst b/Changelog.rst
@@ -1,7 +1,7 @@
 version 3.15.4
 --------------
 
-**2023-??-??**
+**2023-10-??**
 
 * Record dimension coordinate cell characteristics
   (https://github.com/NCAS-CMS/cf-python/issues/692)
@@ -10,6 +10,10 @@ version 3.15.4
   1-d constructs whose axis is not in the data, even when the
   criterion was not matched
   (https://github.com/NCAS-CMS/cf-python/issues/691)
+* Fix bug that prevented "https://" netCDF files from being read
+  (https://github.com/NCAS-CMS/cf-python/issues/699)
+
+----
 
 version 3.15.3
 --------------
diff --git a/cf/read_write/read.py b/cf/read_write/read.py
@@ -5,6 +5,7 @@
 from numbers import Integral
 from os.path import isdir
 from re import Pattern
+from urllib.parse import urlparse
 
 from cfdm import is_log_level_info
 from numpy.ma.core import MaskError
@@ -883,7 +884,8 @@ def read(
         # Expand variables
         file_glob = os.path.expanduser(os.path.expandvars(file_glob))
 
-        if file_glob.startswith("http://"):
+        scheme = urlparse(file_glob).scheme
+        if scheme in ("https", "http"):
             # Do not glob a URL
             files2 = (file_glob,)
         else:
diff --git a/cf/test/test_read_write.py b/cf/test/test_read_write.py
@@ -917,6 +917,14 @@ def test_write_omit_data(self):
         self.assertFalse(g.array.count())
         self.assertTrue(g.construct("grid_latitude").array.count())
 
+    def test_read_url(self):
+        """Test reading urls."""
+        for scheme in ("http", "https"):
+            remote = f"{scheme}://psl.noaa.gov/thredds/dodsC/Datasets/cru/crutem5/Monthlies/air.mon.anom.nobs.nc"
+            # Check that cf can access it
+            f = cf.read(remote)
+            self.assertEqual(len(f), 1)
+
 
 if __name__ == "__main__":
     print("Run date:", datetime.datetime.now())
