help-glpk
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GLPK library for Dev-C++


From: Andrew Makhorin
Subject: Re: GLPK library for Dev-C++
Date: Mon, 04 Jul 2022 18:16:08 +0300

Hi Christian,

Thank you for your interest in glpk.


On Thu, 2022-06-30 at 11:54 +0200, Christian Prins wrote:
> Dear Andrei,
> 
> I am a professor of operations research and industrial engineering at 
> the University of Technology of Troyes, France.
> 
> I love GLPK and use it for instance an LP solver in Julia. But I
> would 
> like to use it with the refactored version of Dev-C++, supervised by 
> Embarcadero the publisher of Delphi. I use this new Dev-C++ for a
> C/C++ 
> course because the IDE is much smaller than Visual Studio Code and it
> is 
> based on TDM-GCC, the Windows version of the Linux compiler GCC. Using
>
> maximum optimization, I found that TDM-GCC is 20% faster than Visual 
> C/C++ of Microsoft.
> https://www.embarcadero.com/free-tools/dev-cpp?aldSet=en-GB
> 
> As usual in C/C++, the problem is the makefile which can be found for 
> Windows, but only for Visual C/C++. Please do you have by chance a 
> Dev-C++ version of the GLPK library?

Currently no makefile for Dev-C/C++ is provided in the official glpk
distribution.  However, it is quite easy to compile glpk with any modern
C compiler.  To do that you need to provide access to all headers in
subdirectory glpk/src and in all its subdirectories, compile all source
files in those directories, and then run the librarian to build the
object library.  You may find the complete list of include directories
and source files needed to build the library in glpk-5.0/setup.job (it
is a plain text file).


Best regards,

Andrew Makhorin


> 
> Another way to solve the problem wuld be to make a "header only 
> library", by putting all the code in a ".h" file. This is what the 
> designers of the BOOST graph library have chosen.
> I had a look to the code of GLP but the task is not so easy : there
> are 
> many files that cannot be simply concatenated in a single larger file.
> 
> I thank you also if you have any advice or recommendations for this
> problem.
> 
> All the best,
> 
> Christian PRINS
> Professor of Operations Research
> Troyes University of Technology
> France
> 



reply via email to

[Prev in Thread] Current Thread [Next in Thread]