[bt-devel] MacOS Build Status

[Gregory Hellings]

On Sep 6, 2008, at 10:09, Martin Gruner <mg.pub at gmx.net> wrote:

> Can you give me a short description how this is traditionally done?  
> I'm still
> a cmake newbie...
> Thanks.

I'm rather new myself and may not have time to pull up the proper  
syntax for a few days, but putting some thing like this near the front  
of the CMakeLists.txt:

If(Build == Release)
   If(install prefix == "")
      Define( install prefix /usr/local)
   endif
Elif(build == Development)
...
Else #default to debug
....
Endif

I'm guessing there's already a similar structure just to parse the  
build type somewhere? Or do you have cmake handle that by default? I'm  
not sure if that's how the best practices are defined, but it seems  
the most straightforward to me.

Again, I will be in and out of touch for most of the next week or so.  
I am presenting at the Conference for Intelligent Data Understanding  
this upcoming week, and don't know how much free time to expect while  
I'm there.



>
>
> mg
>
> Am Samstag, 6. September 2008 15:39:03 schrieb Gregory Hellings:
>> Martin,
>>
>> On Sep 6, 2008, at 8:05, Martin Gruner <mg.pub at gmx.net> wrote:
>>> HI Greg,
>>>
>>> -pedantic-errors and -werror are set if you do not compile in
>>> release mode
>>> (build_release.sh). I see that I need to change CMakeLists.txt to
>>> change the
>>> default behaviour.
>>
>> Perhaps having three build methods would help? Debug, release and
>> devel? I'd prefer to be able to do the debug build for now, since  
>> this
>> is still the testing phase. If you also set the default install  
>> prefix
>> for each of those 3, then you could do away with the .sh scripts.
>>
>> --Greg
>>
>>> mg
>>>
>>> Am Freitag, 5. September 2008 22:24:35 schrieb Gregory Hellings:
>>>> Sent from my iPhone
>>>>
>>>> On Sep 5, 2008, at 11:25, "Eeli Kaikkonen" <eeli.kaikkonen at gmail.com 
>>>> >
>>>>
>>>> wrote:
>>>>> I configured my mailboxes with Thunderbird and this post probably
>>>>> wasn't sent. I subscribed to the list with my gmail account, I  
>>>>> hope
>>>>> this comes through now.
>>>>>
>>>>>
>>>>> Martin Gruner wrote:
>>>>> [  0%] Building CXX object CMakeFiles/bibletime.dir/
>>>>> bibletime_automoc.o
>>>>> /opt/local/include/qt4-mac/QtCore/qglobal.h:747: error: ISO C++  
>>>>> does
>>>>> not support 'long long'
>>>>>
>>>>>
>>>>> Hi Greg,
>>>>>
>>>>> as you can see this is an error in Qt's header files. Something is
>>>>> wrong there.
>>>>>
>>>>> mg
>>>>>
>>>>>
>>>>> Or the compiler is too strict. What c++ compiler and version do  
>>>>> you
>>>>> have? Does Mac compiler have different set of default compiler
>>>>> switches?
>>>>>
>>>>> The problem is one of these lines:
>>>>>
>>>>> #if defined(Q_OS_WIN) && !defined(Q_CC_GNU)
>>>>> #  define Q_INT64_C(c) c ## i64    /* signed 64 bit constant */
>>>>> #  define Q_UINT64_C(c) c ## ui64   /* unsigned 64 bit constant */
>>>>> typedef __int64 qint64;            /* 64 bit signed */
>>>>> typedef unsigned __int64 quint64;  /* 64 bit unsigned */
>>>>> #else
>>>>> #  define Q_INT64_C(c) static_cast<long long>(c ## LL)     /*  
>>>>> signed
>>>>> 64 bit constant */
>>>>> #  define Q_UINT64_C(c) static_cast<unsigned long long>(c ##  
>>>>> ULL) /*
>>>>> unsigned 64 bit constant */
>>>>> typedef long long qint64;           /* 64 bit signed */
>>>>> typedef unsigned long long quint64; /* 64 bit unsigned */
>>>>> #endif
>>>>>
>>>>>
>>>>> For me the g++ compiler switches "-ansi -pedantic" give the same
>>>>> error output with "typedef long long qint64;int main(){}" test
>>>>> program.
>>>>
>>>> The g++ version on the Mac has the same behavior. It's version 4.01
>>>> build 5480. However, the flag -pedantic-errors has the same result
>>>> and
>>>> is enabled in the Makefile in my Bibletime build directory. Is  
>>>> that a
>>>> flag BT added or Cmake? Can it be filtered out?
>>>>
>>>> Greg
>>>>
>>>>> --Eeli Kaikkonen
>>>>>
>>>>> _______________________________________________
>>>>> bt-devel mailing list
>>>>> bt-devel at crosswire.org
>>>>> http://www.crosswire.org/mailman/listinfo/bt-devel
>>>
>>> _______________________________________________
>>> bt-devel mailing list
>>> bt-devel at crosswire.org
>>> http://www.crosswire.org/mailman/listinfo/bt-devel
>>
>> _______________________________________________
>> bt-devel mailing list
>> bt-devel at crosswire.org
>> http://www.crosswire.org/mailman/listinfo/bt-devel
>
>
> _______________________________________________
> bt-devel mailing list
> bt-devel at crosswire.org
> http://www.crosswire.org/mailman/listinfo/bt-devel