[Dip] Branch-and-Cut-and-Price

Sun Jan 17 10:46:05 EST 2016

Hi Mahdi,

The message about i386 is referring to the "make" command. I'm surprised
that it is a 32-bit program, but apparently, this is the case for all iMacs
and shouldn't make any difference. I get the same result on my machine and
things work fine. Do you have Mac ports installed? I recall that I had some
similar problem like this a long while back and that it had to do with a
conflicting version of a system library that was installed by Mac ports.

In any case, my guess is that this has nothing to do with Dip and is just
due to an incompatible library installed somewhere on your machine. This
kind of thing is very difficult to debug if you're not very familiar with
all the tool chains involved. I personally haven't upgraded to OS X 10.11
yes, but I will do that soon and can also try to set up an OS X 10.11 build
on Travis CI to make sure there are no issues. Are you able to build other
things?

If you can post the output of make somewhere (maybe put up a gist on
github) or send it to me off-line, as well as the config.log file in the
Dip subdirectory, I'll see if I can suggest anything.

Cheers,

Ted

On Sun, Jan 17, 2016 at 2:17 AM, m noorizadegan <m.noorizadegan at gmail.com>
wrote:

> Hi Ted,
>
> The OS X version is 10.11.2.
> The compiler details are:
> Apple LLVM version 7.0.2 (clang-700.1.81)
> Target: x86_64-apple-darwin15.2.0.
> Thread model= posix.
> The iMac itself is MK452.
>
> I have also Xcode 7.2 installed
>
> I follow the instruction on the dip page as follows:
>
> cd
> svn checkout https://projects.coin-or.org/svn/Dip/releases/0.92.0 Dip-0.92.0
> cd Dip-0.92.0
> ./configure
> make
> make install
>
> In “make” step I get the following errors:
>
> ld: symbol(s) not found for architecture x86_64
> clang: error: linker command failed with exit code 1 (use -v to see
> invocation)
> make[4]: *** [libDecomp.la] Error 1
> make[3]: *** [all-recursive] Error 1
> make[2]: *** [all] Error 2
> make[1]: *** [all-recursive] Error 1
> make: *** [all-recursive] Error 1
> mahdis-iMac:Dip-0.92.0 MahdiNoorizadegan$ make -v
> GNU Make 3.81
> Copyright (C) 2006  Free Software Foundation, Inc.
> This is free software; see the source for copying conditions.
> There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
> PARTICULAR PURPOSE.
>
> This program built for* i386-apple-darwin11.3.0*
>
> As the above line suggests, it seems there might be incompatibility issue.
>
> Also I do apologise for the typo mistake (difficult), I meant default too.
>
> Please do let me know if more information are required.
>
> Best,
> Mahdi
>
> On 16 Jan 2016, at 6:29 pm, Ted Ralphs <ted at lehigh.edu> wrote:
>
> Hi Mahdi,
>
> DIP isn't built for any particular architecture. It should build on both
> 32-bit and 64-bit architectures. My iMac is 64-bit and it builds fine
> there. If you can post the OS X version you're using, the compiler and
> version, and the exact sequence of commands, etc., we can try to get it
> sorted out.
>
> Cheers,
>
> Ted
>
> On Sat, Jan 16, 2016 at 8:34 AM, m noorizadegan <m.noorizadegan at gmail.com>
> wrote:
>
>> Hi Ted,
>>
>> I also noticed that dip is built for i386-apple-darwin11.3.0.
>> On my os x (or new clang++) it is x86_64-apple-darwin15.2.0.
>> Is it possible that the problem is due to this?
>>
>> Best,
>> Mahdi
>>
>> On 14 Jan 2016, at 6:32 pm, Ted Ralphs <ted at lehigh.edu> wrote:
>>
>> Hi Mahdi,
>>
>> That certainly seems to be one of the possible explanations. If you just
>> build with default options (no arguments to configure) on a modern version
>> of OS x, it should use clang and you should have no problems. If you need
>> gcc for some reason, then I would suggest using the version provided by
>> homebrew.
>>
>> Cheers,
>>
>> Ted
>>
>> On Thu, Jan 14, 2016 at 7:02 AM, mahdi noorizadegan <
>> m.noorizadegan at gmail.com> wrote:
>>
>>> Hi Ted,
>>>
>>> Thanks for your comment.
>>> I searched on the internet, seems it might be due to incomplete
>>> installation of GCC or missing/outdated some libraries  on OS X
>>>
>>> I try these things and will post it on the mailing list.
>>>
>>> Thanks again,
>>> Cheers,
>>> Mahdi
>>>
>>> On 13 January 2016 at 19:50, Ted Ralphs <ted at lehigh.edu> wrote:
>>>
>>>> Hi Mahdi,
>>>>
>>>> Hmm, it's hard to know what's wrong without a bit more information. I
>>>> build Dip on OS X on a regular basis and it should work out of the box.
>>>> I've seen that error before, but it's pretty generic and can be any number
>>>> of things. Would you mind posting some more details to the mailing list?
>>>> It's always good to have these things archived in case others have the same
>>>> problem. Details would include such things as the exact sequence of command
>>>> you executed (what arguments you gave to configure), clang version, OS X
>>>> version, etc. Eventually, I may ask you for the config.log file, but let's
>>>> get the basics down first.
>>>>
>>>> Cheers,
>>>>
>>>> Ted
>>>>
>>>> On Wed, Jan 13, 2016 at 8:02 AM, m noorizadegan <
>>>> m.noorizadegan at gmail.com> wrote:
>>>>
>>>>> Hi Ted,
>>>>>
>>>>> Thanks a lot.
>>>>> By the description, it seems to work for me.
>>>>> I will install DIP and see how it goes although I have encountered
>>>>> some errors on my machine (iMac) following the installation guild line.
>>>>>  Any suggestion to sort it out would be appreciated!
>>>>>  The screenshot is attached.
>>>>>
>>>>> Cheers,
>>>>>
>>>>> Mahdi<Error.png>
>>>>>
>>>>>
>>>>> On 12 Jan 2016, at 8:14 am, Ted Ralphs <ted at lehigh.edu> wrote:
>>>>>
>>>>> Hi Mahdi,
>>>>>
>>>>> Have you looked at DIP (https://projects.coin-or.org/Dip)? It is much
>>>>> easier to work, can be called through a modeling language, and provides
>>>>> many of the same capabilities as BCP with a much lower burden on the user.
>>>>> Let me know if that looks like it might suffice. I'm happy to provide more
>>>>> details and point you towards materials that would help you get started.
>>>>>
>>>>> Cheers,
>>>>>
>>>>> Tede
>>>>>
>>>>> On Sun, Jan 10, 2016 at 8:49 AM, mahdi noorizadegan <
>>>>> mahdi.noorizadegan.09 at mail.wbs.ac.uk> wrote:
>>>>>
>>>>>> Dear Ted,
>>>>>>
>>>>>> I have been working on branch-and-cut-and-price algorithms for some
>>>>>> applications.
>>>>>> I have used SCIP so far. But as you may know there are problems in
>>>>>> its branch-and-bound tree
>>>>>> and also its general performance.
>>>>>>
>>>>>> As noticed Coin-bcp is not maintained anymore.
>>>>>> Implementing an efficient branch-and-bound method that can handle bcp
>>>>>> algorithm is
>>>>>> not easy too.
>>>>>>
>>>>>> I was wondering if you have any advice?
>>>>>>
>>>>>> Looking forward to hearing from you,
>>>>>> Best,
>>>>>> Mahdi
>>>>>>
>>>>>> p.s., Following your last response in the bcp mailing list, I am
>>>>>> sending this email.
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Dr. Ted Ralphs
>>>>> Professor, Lehigh University
>>>>> (610) 628-1280
>>>>> ted 'at' lehigh 'dot' edu
>>>>> coral.ie.lehigh.edu/~ted
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Dr. Ted Ralphs
>>>> Professor, Lehigh University
>>>> (610) 628-1280
>>>> ted 'at' lehigh 'dot' edu
>>>> coral.ie.lehigh.edu/~ted
>>>>
>>>
>>>
>>
>>
>> --
>> Dr. Ted Ralphs
>> Professor, Lehigh University
>> (610) 628-1280
>> ted 'at' lehigh 'dot' edu
>> coral.ie.lehigh.edu/~ted
>>
>>
>>
>
>
> --
> Dr. Ted Ralphs
> Professor, Lehigh University
> (610) 628-1280
> ted 'at' lehigh 'dot' edu
> coral.ie.lehigh.edu/~ted
>
>
>

-- 
Dr. Ted Ralphs
Professor, Lehigh University
(610) 628-1280
ted 'at' lehigh 'dot' edu
coral.ie.lehigh.edu/~ted
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/dip/attachments/20160117/61c3ac8f/attachment.html>