[Rivet] Python, Rivet 1.20, CERN AFS, LXBATCH

[Terrance Figy]

Yeah, this is the same error I receive. I'm trying --with-gsl=/afs/cern.ch/sw/lcg/external/GSL/1.14/x86_64-slc5-gcc44-opt .

On Mar 26, 2010, at 4:37 PM, Andy Buckley wrote:

> Ok, the Rivet build fails for me because it can't find the GNU
> Scientific Library (GSL) for that compiler setup. Except that it *does*
> find it, it just seems to fail to use the results of the gsl-config
> script... looks like I need to revisit the bit of configure script that
> handles the GSL searching!
> 
> Is this the same error as you found, Terrance?
> 
> Andy
> 
> 
> On 26/03/10 15:11, Andy Buckley wrote:
>> Okay, I'm trying that now... the bootstrap script is what I would have
>> recommended (Ben, note that rivet-bootstrap takes a --devmode argument,
>> which uses the SVN checkout: might be easier for you in future) anyway,
>> and since this is using gcc4.4 I think it's necessary to --ignore-lcgext
>> because they don't have builds for that setup, IIRC. So it *should*
>> work... let's see!
>> 
>> Andy
>> 
>> 
>> On 26/03/10 14:49, Terrance Figy wrote:
>>> For downloading I used the bootstrap script. I ignored the lcgext. I also installed boost. 
>>> 
>>> On Mar 26, 2010, at 3:45 PM, Ben Waugh wrote:
>>> 
>>>> On 26/03/10 13:58, Terrance Figy wrote:
>>>>> The setup of script I used is given below:
>>>>> 
>>>>>> . /afs/cern.ch/sw/lcg/hepsoft/0.1/x86_64-slc5-gcc44-opt/setup.sh
>>>>> 
>>>>> This is the setup script that has been recently created for us in the TH group. We're making a push to change the computing practices with the TH group.
>>>>> So, any help will be greatly appreciated.
>>>> 
>>>> But how are you downloading Rivet and compiling it? There are three methods that I can think of:
>>>> (1) the bootstrap script: see http://projects.hepforge.org/rivet/trac/wiki/GettingStarted
>>>> (2) download the tarball and build it: see http://projects.hepforge.org/rivet/trac/wiki/ManualInstallation
>>>> (3) check out the code from Subversion and build it
>>>> 
>>>> Those are probably in what should be order of preference, but because of problems finding compatible versions of Python, ROOT etc., and problems with building the PDF manual from the tarball, I went for option (3) at our site.
>>>> 
>>>> With luck, Andy can tell you what you *should* do. In case it is any help, I have attached a log of what I did.
>>>> 
>>>> Cheers,
>>>> Ben
>>>> 
>>>>> 
>>>>> Cheers,
>>>>> Terrance
>>>>> 
>>>>> 
>>>>> On Mar 26, 2010, at 2:47 PM, Ben Waugh wrote:
>>>>> 
>>>>>> Hi Andy, Terrance,
>>>>>> 
>>>>>> Looks like I didn't actually reply to Terrance, just to the Rivet list. Terrance, can you tell us what setup script you are using as Andy suggests? If you are using the code from Svn and want a quick workaround you can still try installing a new Swig as I did.
>>>>>> 
>>>>>> On 26/03/10 13:00, Andy Buckley wrote:
>>>>>>> On 25/03/10 21:49, Ben Waugh wrote:
>>>>>>>> Dear Terrance,
>>>>>>>> 
>>>>>>>> I had the same problem ("invalid conversion" in one of the source files
>>>>>>>> generated by Swig) compiling Rivet on SLC5. There is a bug in the
>>>>>>>> version of swig supplied with the OS, so I had to install my own version
>>>>>>>> (swig-1.3.40) locally and place it in my $PATH before the standard bin
>>>>>>>> directories.
>>>>>>> 
>>>>>>> That seems odd to me: I *think* we are bundling those generated source
>>>>>>> files, so SWIG shouldn't need to be present on the Rivet build system
>>>>>>> unless you're building from SVN.
>>>>>> 
>>>>>> In my case, I am building from Svn. Yesterday the download tarball appeared to be missing a file spiresbib.py that I guess you have since added, but I still get another error when making the PDF manual from the tarball:
>>>>>> No rule to make target `refs.bib', needed by `rivet-manual.pdf'
>>>>>> 
>>>>>> Cheers,
>>>>>> Ben
>>>>>> 
>>>>>>> 
>>>>>>> Terrance, can you point us at / provide that setup script, so we can
>>>>>>> test it ourselves?
>>>>>>> 
>>>>>>> Thanks,
>>>>>>> Andy
>>>>>>> 
>>>>>> 
>>>>>> --
>>>>>> Dr Ben Waugh                                   Tel. +44 (0)20 7679 7223
>>>>>> Dept of Physics and Astronomy                  Internal: 37223
>>>>>> University College London
>>>>>> London WC1E 6BT
>>>>> 
>>>> 
>>>> -- 
>>>> Dr Ben Waugh                                   Tel. +44 (0)20 7679 7223
>>>> Dept of Physics and Astronomy                  Internal: 37223
>>>> University College London
>>>> London WC1E 6BT
>>>> <rivet-build.log>
>>> 
>>> 
>> 
>> 
> 
> 
> -- 
> Dr Andy Buckley
> SUPA Advanced Research Fellow
> Particle Physics Experiment Group, University of Edinburgh