launchpad-reviewers team mailing list archive

[Colin Watson <cjwatson@xxxxxxxxxxxxx>]

Colin Watson has proposed merging ~cjwatson/launchpad:builtin-pow into launchpad:master.

Commit message:
Revert looptuner tests to using builtin pow()

Requested reviews:
  Launchpad code reviewers (launchpad-reviewers)

For more details, see:
https://code.launchpad.net/~cjwatson/launchpad/+git/launchpad/+merge/386523

We switched to math.pow because Twisted Conch used gmpy which overrode __builtin__.pow.  But Twisted Conch no longer uses gmpy, so that's no longer a problem.
-- 
Your team Launchpad code reviewers is requested to review the proposed merge of ~cjwatson/launchpad:builtin-pow into launchpad:master.

diff --git a/lib/lp/services/doc/looptuner.txt b/lib/lp/services/doc/looptuner.txt
index aeb91cb..8c83471 100644
--- a/lib/lp/services/doc/looptuner.txt
+++ b/lib/lp/services/doc/looptuner.txt
@@ -324,9 +324,7 @@ compensates, then finally converges towards it.
 If the cost function is highly regular and predictable, the oscillation will
 be a neat alternation of oversized and undersized chunks.
 
-We use math.pow here, since __builtin__.pow is overridden by twisted.conch.
-
-    >>> body = CostedLoop(lambda c: math.pow(1.2, c), 50)
+    >>> body = CostedLoop(lambda c: pow(1.2, c), 50)
     >>> loop = CostedTuner(body, goal_seconds, 1)
     >>> loop.run()
     start
@@ -350,9 +348,7 @@ We use math.pow here, since __builtin__.pow is overridden by twisted.conch.
 With more extreme exponential behaviour, the overshoot increases but the
 effect remains the same:
 
-We use math.pow here, since __builtin__.pow is overridden by twisted.conch.
-
-    >>> body = CostedLoop(lambda c: math.pow(3, c), 50)
+    >>> body = CostedLoop(lambda c: pow(3, c), 50)
     >>> loop = CostedTuner(body, goal_seconds, 1)
     >>> loop.run()
     start