Re: [Boost-bugs] [Boost C++ Libraries] #4421: fix for #4400

Subject: Re: [Boost-bugs] [Boost C++ Libraries] #4421: fix for #4400
From: Boost C++ Libraries (noreply_at_[hidden])
Date: 2015-05-19 10:09:37


#4421: fix for #4400
--------------------------------------+---------------------------
  Reporter: Wolf Lammen <ookami1@…> | Owner: no-maintainer
      Type: Patches | Status: new
 Milestone: Boost 1.44.0 | Component: preprocessor
   Version: Boost 1.44.0 | Severity: Problem
Resolution: | Keywords:
--------------------------------------+---------------------------

Comment (by ookami1@…):

 I do see your point, but let me comment on it nevertheless. This is not
 meant to talk you into applying this patches. It is fine with me, and
 actually I was quite surprised and pleased to see somebody care about this
 report at all after such a long time.

 Usually, you would do parameter checking in order to enforce an API. This
 cannot really be done in preprocessor code, as this is very expensive and
 requires lots of macro replacements, unduely slowing down compiling.

 So it is up to users to carefully use the preprocessor code. Apart from
 sloppy mindsets found all over the place, correct handling is not made
 easy. Who really sees at a glance that BOOST_PP_SEQ_REST_N must not be
 called with an index pointing to the last element. And are there other
 places with obscure restrictions? Then we have the exceptions (see the
 documentation for BOOST_PP_SEQ_NIL) which may fool people into thinking,
 they can do likewise somewhere else.

 On top, it does not help that BOOST_PP_SEQ_REST_N in reality behaves as
 expected in the vast majority of (and practically all relevant) cases. So
 code appears to be correct, but it is not.

 I proposed my patches because I felt it is more appropriate to be fault
 tolerant in such a situation.

 cheers Wolf

 Replying to [comment:9 eldiener]:
> Fixed in the latest preprocessor code in the 'develop' branch.
>
> While I appreciate the patches I did not use them. It is an undefined
 result to call BOOST_PP_SEQ_REST_N with an 'n' which is equal to the size
 of the 'seq'. The fact that this returns nothing is an implementation
 detail, since a seq cannot be empty and BOOST_PP_SEQ_REST_N clearly states
 that it expands to a seq. I will update the documentation accordingly. I
 am not willing to change BOOST_PP_SEQ_REST_N to accomodate a situation
 which should be undefined.

-- 
Ticket URL: <https://svn.boost.org/trac/boost/ticket/4421#comment:10>
Boost C++ Libraries <http://www.boost.org/>
Boost provides free peer-reviewed portable C++ source libraries.

This archive was generated by hypermail 2.1.7 : 2017-02-16 18:50:18 UTC