libreoffice-3.4.5.2 failing to build

Wed Apr 4 15:01:55 UTC 2012

On 04/01/12 15:55, Da Rock wrote:
> On 03/30/12 17:03, Da Rock wrote:
>> On 03/30/12 16:22, Baptiste Daroussin wrote:
>>> On Thu, Mar 29, 2012 at 02:11:33PM +1000, Da Rock wrote:
>>>> On 03/27/12 21:42, Da Rock wrote:
>>>>> I've been hunting down some info on why this is failing, but all I
>>>>> keep running into is the previous versions failings... :(
>>>>>
>>>>> I'm building libreoffice on 9.0-Release amd64, and I'm getting build
>>>>> errors qa testing(?) sd. I followed the instructions, but it still
>>>>> failed. It spits out this:
>>>>>
>>>>> scripting deliver
>>>>> deliver -- version: 275594
>>>>> module 'scripting' delivered successfully. 0 files copied, 28 files
>>>>> unchanged
>>>>> terminate called after throwing an instance of
>>>>> 'com::sun::star::uno::RuntimeException'
>>>>> /usr/local/bin/bash: line 1: 2019 Abort trap: 6 (core dumped)
>>>>> [ a whole lot of code (I'm copying by hand, so it is a bit
>>>>> overwhelming) - involving clone/impress/sd/unxfbsd.pro, solver, and
>>>>> qa_unit ]
>>>>> dmake: Error code 134, while making 'test'
>>>>> gb_LinkTarget_add_library_objects,CppunitTest/libtest_sc_ucalc.so,sc
>>>>> gb_LinkTarget_add_linktarget_objects,CppunitTest/libtest_sc_ucalc.so,Library/libscfb.so 
>>>>>
>>>>>
>>>>> [ build all ] top level modules: sc
>>>>> [ build all ] loaded modules: sc
>>>>> [ build cut ] sc_ucalc
>>>>> Abort trap (core dumped)
>>>>> terminate called after throwing an instance of
>>>>> 'com::sun::star::uno::RuntimeException'
>>>>> gmake[1]: ***
>>>>> [/usr/ports/editors/libreoffice/work/libreoffice-bootstrap-3.4.5.2/solver/340/unxfbsd.pro/workdir/CppunitTest/sc_ucalc.test] 
>>>>>
>>>>> Error 1
>>>>> dmake: Error code 2, while making 'all'
>>>>> [ build all ] top level modules: sw
>>>>> [ build all ] loaded modules: sw
>>>>> [ build all ] loaded modules: sw
>>>>> [ build log ] sw
>>>>> sw deliver
>>>>> deliver -- version: 275594
>>>>> Module 'sw' delivered successfully. 0 files copied, 1 files unchanged
>>>>>
>>>>> -----------------------------------------------------
>>>>> [ standard error message with webpage link to debug page ]
>>>>>
>>>>> internal build errors:
>>>>>
>>>>> ERROR: error 65280 occurred while making
>>>>> /usr/ports/editors/libreoffice/work/libreoffice-bootstrap-3.4.5.2/sd/qa/unit 
>>>>>
>>>>> ERROR: error 65280 occurred while making
>>>>> /usr/ports/editors/libreoffice/work/libreoffice-bootstrap-3.4.5.2/sc/prj 
>>>>>
>>>>>
>>>>> [ instructions to build sd part of the build to isolate the issue ]
>>>>>
>>>>> As I mentioned I followed the directions and still hit the same error
>>>>> filed above (if more of the data is needed I guess I'll have to 
>>>>> find a
>>>>> way to post it...).
>>>>>
>>>>> I'm on a deadline for this one, so I have to get this to work. How
>>>>> should I further debug this to fix it? Or failing that (as an extreme
>>>>> last resort), will an older package already built work IF it is built
>>>>> for 8.2?
>>>> After a portsnap update I still get core dumps in the
>>>> calc/sc/cppunittester and the impress/sd/qa/unit/cppunittester.
>>>>
>>>> I've tried turning off make jobs, and I also checked the core files
>>>> (when I finally found them). The only clue I have is this:
>>>>
>>>> #0 0x00000008016bca7c in ??
>>>>
>>>> This is in _both_ core files, and although its not very helpful it 
>>>> says
>>>> its stopping in exactly the same place (duh).
>>>>
>>>> Seriously guys, where do I go from here? Googling isn't exactly 
>>>> offering
>>>> up much - my searches can be as much as 7 results (usually my own 
>>>> stuff
>>>> or duplicates of the same).
>>>>
>>>> Or, as my last resort, can I use the 8.2 package instead?
>>>>
>>>> Need some serious pointers, pleeease... :)
>>> is you 9.0-RELEASE a freshly built system? or installed installed from
>>> installers or upgraded?
>> Completely clean install last week, and ports tree refreshed since on 
>> wed.
>>
>> I've been on the libreoffice-dev lists and the only advice is to 
>> disable the tests - but if the test is failing it must be for a 
>> reason don't you think? So why?
>>
>> Thanks for these couple of responses, still need help though :)
>
>  I have news from the eastern front :)
>
> I've hotwired the makefile to use clang, and I have some errors 
> cropping up close to where I was having issues using gcc:
>
> In file included from 
> /usr/ports/editors/libreoffice/work/libreoffice-bootstrap-3.4.5.2/soltools/giparser/gi_list.cxx:33:
> In file included from ../inc/gi_list.hxx:33:
> ../inc/st_list.hxx:298:10: error: use of undeclared identfier 'push_back'
>                 push_back( new XY(*(*it)) );
>                 ^
>                 this->
> ... more notes - usually related to this issue...
> ../inc/st_list.hxx:311:5: error: use of undeclared identifier 'checkSize'
>                 checkSize(DynamicList<XY>::len-2);
>                 ^
>                 this->
> ... even more notes the last error...
>
> And then the usual fail routine.
>
> Any pointers on what is going on? What can I switch to improve the 
> situation? The identifiers are there, but it doesn't seem to see that. 
> Is this a C89/99 thing maybe?
>
> I love clang - it shows up all the little nuances More than likely the 
> root of my original issue...

Interesting outcome on this one. I'm not 100% sure who to send this to, 
but I post here for interest sake.

I built this port on another 9.0 amd64 system and it built fine. The 
only difference between them is an Atom processor. Apparently some 
things can not be built on an Atom processor and there seem to be minute 
differences even though they're supposed to be x86 compatible.

I ran make package and transferred it to the system using usb and pkg_add.

Something similar appears to be happening with samba, but that is still 
in progress.

HTH someone else in the future, and the developers/porters behind 
libreoffice.