Remove default parallelizable parameter in NoJump transformation (#4604)

* Fixes #4259 * Remove parallelizable parameter in NoJump transformation (and always set it to False because the transformation is inherently not parallelizable with split-apply-combine) * update CHANGELOG
MDAnalysis · May 24, 2024 · 2c6e21f · 2c6e21f
1 parent 0f04e3e
commit 2c6e21f
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 3 deletions.
diff --git a/package/CHANGELOG b/package/CHANGELOG
@@ -21,6 +21,7 @@ The rules for this file:
  * 2.8.0
 
 Fixes
+ * Fix #4259 via removing argument `parallelizable` of `NoJump` transformation.
  * Fix doctest errors of analysis/pca.py related to rounding issues 
    (Issue #3925, PR #4377)
  * Convert openmm Quantity to raw value for KE and PE in OpenMMSimulationReader.

diff --git a/package/MDAnalysis/transformations/nojump.py b/package/MDAnalysis/transformations/nojump.py
@@ -99,11 +99,10 @@ def __init__(
         self,
         check_continuity=True,
         max_threads=None,
+    ):
         # NoJump transforms are inherently unparallelizable, since
         # it depends on the previous frame's unwrapped coordinates
-        parallelizable=False,
-    ):
-        super().__init__(max_threads=max_threads, parallelizable=parallelizable)
+        super().__init__(max_threads=max_threads, parallelizable=False)
         self.prev = None
         self.old_frame = 0
         self.older_frame = "A"