Google Code-in

Look at Bug 2498. This bug report contains two patches.

Your task is to check that these patches still apply cleanly, the documentation in the first patch is clear, and verify the claimed performance:

1. Run bench-simulator as shown in the bug report, and save the performance table as shown in the bug report
2. Verify that the two patches apply cleanly
3. Check that the Doxygen builds without errors on this file (see the Manual section on Building Doxygen Docs for how to do this). Does the edited description make sense? Does it completely describe the new approach?
4. Re-run bench-simulator and show the performance comparison before and after the patch.

Bonus: it would be nice to switch between the old version and the new version by a configuration variable:

1. Add a boolean configuration variable to the class, called m_reverse. The original code should be the default with m_reverse = false; the new version is true.
2. Add a public function bool SetReverse (bool reverse). This should return the current value of m_reverse, then set it to the argument value.
3. Where the patch makes changes to calendar-scheduler.cc, make those depend on m_reverse. Note that most of these are inside loops. Instead of testing m_reverse each iteration of the loop, insert the old and new version of the loop, so the m_reverse test selects which loop version to use.