[Os-project-managers] printModel

[Horand Gassmann]

"R. Kipp Martin" <kmartin at chicagobooth.edu> wrote:

> Hi Guys:
>
>>
>> 1. There will be four different commands:
>>
>> printModel
>> printRow n
>> printObjective n
>> printConstraint n
>>
>> The last three are really just modifications on one theme.
>
> I think we should have just printModel and printModel n. This is the  
> way it is actually implemented in C++.  I think FOUR new commands is  
> a mistake.

I didn`t think you`d go for my original proposal, but can we agree on  
two commands? I think it is better to have

printModel
printRow n

with different names for two very different situations. Especially if  
I have multiple models (which is possible in AMPL, or in a  
decomposition algorithm)

printModel 5

seems to be subject to misunderstandings. Why go there?

Cheers

gus


>> printRow n
>>
>> prints row n, which could be a constraint if n>= 0 or an objective if
>> n<0. I think it is a bit counterintuitive to have to specify
>>
>> printRow -2
>>
>> for the second objective, hence I am thinking of adding the possibility
>> to specify
>>
>> printObjective 1
>>
>> instead. And once I got there, I figured I might as well add
>>
>> printConstraint n
>>
>> for symmetry.
>
> I think it is too complicated.
>>
>> 2. They would be available both in interactive mode and at the command
>> line, but only for files that reside on the local machine.
>>
>> 3. In interactive mode, the command would, of course, be activated
>> immediately. If a .osil file is available (i.e., has been read
>> previously), then we print the osil file or parts of it, if none is
>> available, we look for MPS, nl or GAMS files, in that order. If no file
>> is available at all, an error is generated.
>>
>> 4. In command line mode, the command would be activated just before the
>> service method is called. If no service method is given, the default is
>> solve, so the printModel command is executed just before the solve.
>
> But I assume printModel is not the default, correct. I.e. if the  
> user does not specify a printModel, then nothing should get printed.  
> Correct?
>
> Still not sure I buy into this being necessary in command line mode.
>>
>> I have not yet thought through what would happen if the optimization
>> instance resides on the remote machine. I guess in that case the print
>> command needs to be activated from inside the osol file, right?
>
> No, I would say we don't invoke. This is getting too complicated.
>
> Cheers
>>
>> When you have a chance, can you look this over and see if I am missing
>> something obvious?
>>
>> Cheers
>>
>> gus
>>
>> -------------------------------------------------------
>>
>> Horand I. Gassmann, Professor
>>
>> School of Business Administration, Dalhousie University
>> 6100 University Avenue
>> Halifax, Nova Scotia, Canada , B3H 3J5
>> ph. (902) 494-1844
>> fax (902) 494-1107
>>
>> http://myweb.dal.ca/gassmann/
>>
>
>
> -- 
> Kipp Martin
> Professor of Operations Research
> and Computing Technology
> Booth School of Business
> University of Chicago
> 5807 South Woodlawn Avenue
> Chicago, IL 60637
> 773-702-7456
> kmartin at chicagobooth.edu
> http://www.chicagobooth.edu/faculty/bio.aspx?person_id=12825325568
> http://projects.coin-or.org/OS
>
> Sent without Blackberry, Droid, iPhone, or any other
> wireless device.
> -- 
>
>



-------------------------------------------------------

Horand I. Gassmann, Professor

School of Business Administration, Dalhousie University
6100 University Avenue
Halifax, Nova Scotia, Canada , B3H 3J5
ph. (902) 494-1844
fax (902) 494-1107

http://myweb.dal.ca/gassmann/