Boost logo

Boost :

Subject: Re: [boost] Bugs in regression/library_status.cpp on Windows
From: Robert Kawulak (robert.kawulak_at_[hidden])
Date: 2011-07-24 18:17:36


Could anybody please apply that patch or suggest what can I do (or whom to
contact directly) to get this fixed? BTW, I wonder how it's possible that
most of the bugs were introduced in one commit half a year ago and nobody
even noticed - it would suffice to run this app on any platform to get
obvious errors while on Windows it wouldn't even compile. Am I really the
only person using this tool? :P

I've also found another minor problems:
- when building it, "$(BOOST_ROOT)\tools\regression\build\bin" directory is
created but it's not marked with "ignore" in SVN;
- char_xlate functor only translates '/' character in URLs while on Windows
the character used is '\' (and it also causes problems in some browsers).
The solution is very simple - add "|| c = '\\'" in line 75.

I'm attaching a patch file combining the previous fixes and the '\'

Best regards,

> -----Original Message-----
> From: Robert Kawulak [mailto:Robert.Kawulak_at_[hidden]]
> Sent: Saturday, July 09, 2011 11:35 PM
> To: 'boost_at_[hidden]'
> Cc: 'bdawes_at_[hidden]'; 'blelbach_at_[hidden]'
> Subject: Bugs in regression/library_status.cpp on Windows
> Hello,
> I've stumbled upon several bugs in
> tools/regression/src/library_status.cpp when compiling on Windows with
> gcc 4.5 and msvc 10.0. I'm attaching a patch file with all necessary
> fixes. The bugs are:
> - using path::native() instead of path::string() - the former is
> wstring on Windows,
> - calling path::string() twice to obtain begin() and end() iterators -
> on Windows path::string() returns a temporary, so the iterators point
> to different objects,
> - using whole path for each node in build_node_tree() instead of only
> the directory name,
> - using whole path instead of only the test name in do_table_body(),
> - the link to links file is relative to test directory, not to the
> results output directory,
> - generation of closing "</a>" without opening "<a>".
> Best regards,
> Robert

Boost list run by bdawes at, gregod at, cpdaniel at, john at