Re: [boost] [teeks99][test] Test failures on MSVC 8-11

22 Nov 2015

...
-----Original Message-----
From: Boost [mailto:boost-bounces@lists.boost.org] On Behalf Of Bjorn Reese
Sent: 22 November 2015 16:21
To: boost@lists.boost.org
Subject: Re: [boost] [teeks99][test] Test failures on MSVC 8-11
On 11/21/2015 09:54 PM, Raffi Enficiaud wrote:
...
#ifndef BOOST_NO_CXX11_HDR_INITIALIZER_LIST
static std::map const char_type{{
     {'<' , "lt"},
     {'>' , "gt"},
     {'&' , "amp"},
     {'\'', "apos"},
     {'"' , "quot"}
     }};
#else /.../
I would replace the std::map with a switch statement in this case. Then there is no need for an
#ifndef.
...
...
http://www.boost.org/development/tests/develop/developer/output/igazt
anaga-gcc-5-1c++14-boost-bin-v2-libs-log-test-attr_function-test-gcc-
mngw-5-1c+-dbg-dbg-symbl-off-thrd-mlt.html
and maybe because of "-Werror=sign-compare" ? or again this is hiding
something else that was cut off in the report...
The problem is that if we test an unsigned integer against a literal integer value, then we a
warning in
the first test but not the second:
unsigned int value = 1;
   BOOST_REQUIRE(value == 1); // No sign compare warning
   BOOST_REQUIRE_EQUAL(value, 1); // Sign compare warning
The quick fix is for all Boost.Test users to do:
BOOST_REQUIRE_EQUAL(value, 1U);
but this is inconvenient if you do not know the signedness of the type, e.g. when you use a size()
member function on some container.
I think that Boost.Test behaviour is correct/best here and that users should be providing the right
type for the test, even if a bit more work and/or thought.

For size() items one should best use size_t or size_type?

And in general  BOOST_REQUIRE_EQUAL rather than BOOST_REQUIRE
is usually much more helpful in telling *why* the test has failed.

Paul

---
Paul A. Bristow
Prizet Farmhouse
Kendal UK LA8 8AB
+44 (0) 1539 561830