Re: [boost] [build] bootstrap.sh is still broken

Subject: Re: [boost] [build] bootstrap.sh is still broken
From: Steven Watanabe (watanabesj_at_[hidden])
Date: 2013-10-21 13:57:08

AMDG

On 10/21/2013 10:32 AM, Vladimir Prus wrote:
> On 21.10.2013 21:03, Stephan T. Lavavej wrote:
>> [STL]
>>> http://lists.boost.org/boost-build/2013/10/27025.php
>>
>> [Rene Rivera]
>>> Could you post the diff patches as attachments? It's almost
>>> impossible to
>>> deal with inline diffs to apply these. Even though I can't test any of
>>> this. I'm willing to apply the patches.
>>
>> Thanks! Download this: http://nuwen.net/stuff/boost-bootstrap.patch
>>
>> As I explained in my mail, the diff for boost_1_54_0/bootstrap.sh is a
>> horrible hack and should not be applied. It fixes mingw but breaks
>> everyone
>> else. I would love to get a real fix for this, but someone would have to
>> investigate.
>>
>> The diffs for boost_1_54_0/tools/build/v2/engine/build.jam and
>> boost_1_54_0/tools/build/v2/engine/build.sh should be applied - they fix
>> mingw and should not affect anyone else.
>>
>> [Vladimir Prus]
>>> And on the mingw topic, I might have missed something, but I think the
>> major
>>> issue is still mingw vs. gcc naming inconsistency between bootstrap.sh
>>> and b2 proper - which is ugly, but not quite a showstopper?
>>
>> That's what my boost_1_54_0/bootstrap.sh diff is hacking around. It's a
>> showstopper in the sense that it completely breaks the build, and I
>> haven't
>> figured out how to evade it with command-line arguments.
> 
> Ok. Rene, Steven, how about we s/mingw/gcc throughout b2 engine source?
> 

I agree in principle, but it's not quite that easy.
mingw is treated slightly differently from gcc
(windows sources, not unix).

In Christ,
Steven Watanabe