Re: Horrible compiletimes and memory usage while compiling a parser with X3

On 11/07/2016 02:44 AM, Peter Gerell wrote:
> On 07/11/2016 12:38 PM , Joel de Guzman wrote:
>> On 07/11/2016 1:26 AM, Mikael Asplund wrote:
>>> This is the reason we're still on 1.60 (with a small
>>> const-patch from 1.61), since later version have link problems which are just too hard
>>> to debug.
>
>> It would be great if you post a test file that exhibits this problem you mention.
>> This linking problem doesn't look good and must be fixed.
>
> Hi Joel.
> This is the problem Mikael is referring to.
> http://melpon.org/wandbox/permlink/X8Awj4wIZ2JqMWIw
>
> If KEEP_CHAR is defined the code builds with both boost 1.60 and 1.61.
> Without KEEP_CHAR it fails with 1.61.
>
> The following commit introduced the problem.
> https://github.com/boostorg/spirit/commit/a8e391bd99dddb3f9ece84bdb1bb9236b0a37cf7
>
> The problem seems to be that the attribute types are short-circuited for sequences of one
> even though we explicitly request a specific binding using FUSION_DEFINE_FOR_STRUCT(ast::Start, count);
[snip]

More specifically, the change on line 178 causes the
problem.  That change caused r_pass on line 303 to be
derived from pass_through_sequence_attribute instead of
pass_sequence_attribute_front, as was the case with
boost_1_60_0.  That change, in turn, caused r_attr on line
303 to just copy attr, which was type ast::Start, instead of
the type type of the boost_1_60_0 r_attr which was type
ast::Count.

Since parser.right has type:

   rule<count_class, ast::Count, false>

the call on line 312 of detail/sequence.hpp:

   && parser.right.parse(first, last, context, rcontext,
   r_attr))

will, ultimately, cause parse_rule to be called with the
wrong r_attr(the one with type ast::Start instead of
ast::Count), and, since the expression.cpp only has:

   BOOST_SPIRIT_INSTANTIATE(count_type, iterator_type,
   x3::unused_type);

which only instantiates (according to the definition of that
macro in rule.hpp:173) the parse_rule specializaton for
attr=ast::Count, the link error happens.

Why not revert the change on detail/sequence.hpp:178 to
solve the problem?

-regards,
Larry



------------------------------------------------------------------------------
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi