changeset 38517:6db38c9d7e00

revlog: efficient implementation of 'descendant'

Iterating over descendants is costly, because there are no "parent -> children"
pointers. Walking the other way around is much more efficient, especially on
large repositories, where descendant walks can cost seconds. And the other hand,
common ancestors code follows links in the right direction and has a compiled
implementation.

In real life usage, this saved up to 80s during some pull operations, where
descendant test happens in extension code.

--- a/mercurial/revlog.py	Thu Jun 21 23:56:51 2018 +0100
+++ b/mercurial/revlog.py	Fri Jun 22 00:05:20 2018 +0100
@@ -1376,16 +1376,14 @@
         return c
 
     def descendant(self, start, end):
+        """True if revision 'end' is an descendant of revision 'start'
+
+        A revision is considered as a descendant of itself."""
         if start == nullrev:
             return True
         elif start == end:
             return True
-        for i in self.descendants([start]):
-            if i == end:
-                return True
-            elif i > end:
-                break
-        return False
+        return start in self._commonancestorsheads(start, end)
 
     def commonancestorsheads(self, a, b):
         """calculate all the heads of the common ancestors of nodes a and b"""