Re: [boost] [ratio] Request for permission to merge to release branch

Subject: Re: [boost] [ratio] Request for permission to merge to release branch
From: Vicente J. Botet Escriba (vicente.botet_at_[hidden])
Date: 2011-11-02 07:57:35

Le 02/11/11 05:31, Rene Rivera a écrit :
> On 11/1/2011 11:22 PM, Eric Niebler wrote:
>> On 11/1/2011 2:05 PM, Vicente J. Botet Escriba wrote:
>>> Hi,
>>>
>>> I want to merge https://svn.boost.org/trac/boost/changeset/75240 to the
>>> release branch. The modifications are minor and correct to tickets:
>>>
>>> https://svn.boost.org/trac/boost/ticket/5724 and
>>> https://svn.boost.org/trac/boost/ticket/6082.
>>
>> I'd love to answer this, but Trac is giving me "OperationalError:
>> database is locked". Rene, I thought we fixed this. Any thoughts?
>
> I didn't have a problem earlier.. Or now. But that's expected "random" 
> behavior for this. And the problem was only ameliorated, not solved. 
> It will only be solved when we move servers and subversion, and move 
> trac to a real database, or move to something other than trac.
>
> PS. Even after looking at the the above changeset.. I don't feel 
> knowledgeable enough to say if this is a safe merge or not.
>
>
Hi,

after merging the change in my local repository the differences are the 
following

pwd
/Users/viboes/boost/rel/boost/ratio
iMac-de-Vicente-Botet-Escriba:ratio viboes$ svn diff
Index: detail/overflow_helpers.hpp
===================================================================
--- detail/overflow_helpers.hpp    (revision 75242)
+++ detail/overflow_helpers.hpp    (working copy)
@@ -242,7 +242,12 @@
 >
 >::type type;
    };
-
+  template <class R, boost::intmax_t D>
+  struct ratio_add<R, ratio<0,D> >
+  {
+    typedef R type;
+  };
+
    template <class R1, class R2>
    struct ratio_subtract
    {
@@ -269,6 +274,12 @@
 >::type type;
    };

+  template <class R, boost::intmax_t D>
+  struct ratio_subtract<R, ratio<0,D> >
+  {
+    typedef R type;
+  };
+
    template <class R1, class R2>
    struct ratio_multiply
    {

This just avoids an instantiation that makes a division by 0.

I don't really think it is risky at all.

Best,
Vicente