For Programmers: Free Programming Magazines  


Home > Archive > Fortran > February 2005 > Source code of large programs wanted









You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

 

Author Source code of large programs wanted
Derek M Jones

2005-02-24, 4:00 pm

All,

My interest is in measuring source code.
You can can find some of my measurements of C
source at www.knosof.co.uk/cbook/usefigtab.pdf
(you might want to skip to table 26; many of the
tools used are available at
www.knosof.co.uk/cbook/srccnt.tgz).

I would like to measure Fortran source and am
looking a selection of large Fortran programs
(i.e., +100K lines per program) for a total of around
3-5 million lines of source. I don't mind what the
code does, but would prefer a wide selection of
different kinds of application. I don't plan to
execute the software, so no data-sets are needed.
I would like others to have access to the source
(so they can also measure it), so publicly available
source is needed. A selection of programs developed
in the 60's, 70's, 80's, and 90's would be great.

Suggestions most welcome (either the name of the
program or a url will do).
Steven G. Kargl

2005-02-24, 4:00 pm

In article <Xns96071F07EADD0derekknosofcouk@194.168.222.124>,
Derek M Jones <derek@NOSPAMknosof.co.uk> writes:
>
> I would like to measure Fortran source and am
> looking a selection of large Fortran programs
> (i.e., +100K lines per program) for a total of around
> 3-5 million lines of source. I don't mind what the
> code does, but would prefer a wide selection of
> different kinds of application. I don't plan to
> execute the software, so no data-sets are needed.
> I would like others to have access to the source
> (so they can also measure it), so publicly available
> source is needed. A selection of programs developed
> in the 60's, 70's, 80's, and 90's would be great.


You can cover a large portion of the 60's, 70's and 80's
by a visit to www.netlib.org. You can also find a nice
selection of more modern Fortran at www.fortran.com.
Other placed to look include Polyhedron and Alan Miller's
web. Both can be found via google.


--
Steve
http://troutmask.apl.washington.edu/~kargl/
beliavsky@aol.com

2005-02-24, 8:58 pm


Derek M Jones wrote:
> All,
>
> My interest is in measuring source code.
> You can can find some of my measurements of C
> source at www.knosof.co.uk/cbook/usefigtab.pdf
> (you might want to skip to table 26; many of the
> tools used are available at
> www.knosof.co.uk/cbook/srccnt.tgz).
>
> I would like to measure Fortran source and am
> looking a selection of large Fortran programs
> (i.e., +100K lines per program) for a total of around
> 3-5 million lines of source. I don't mind what the
> code does, but would prefer a wide selection of
> different kinds of application. I don't plan to
> execute the software, so no data-sets are needed.
> I would like others to have access to the source
> (so they can also measure it), so publicly available
> source is needed. A selection of programs developed
> in the 60's, 70's, 80's, and 90's would be great.
>
> Suggestions most welcome (either the name of the
> program or a url will do).


There are links to Fortran source codes at
http://www.dmoz.org/Computers/Progr...an/Source_Code/
.. I suggest you do your analysis separately for Fortran source codes
using the fixed and free formats, which will depend largely on whether
Fortran 77 or a later version of Fortran was used.

Greg Lindahl

2005-02-24, 8:58 pm

In article <Xns96071F07EADD0derekknosofcouk@194.168.222.124>,
Derek M Jones <derek@NOSPAMknosof.co.uk> wrote:

>I would like to measure Fortran source and am
>looking a selection of large Fortran programs
>(i.e., +100K lines per program) for a total of around
>3-5 million lines of source.


Google is your friend: AIPS MM5 WRF POP AIPS "ASCI Purple Benchmarks"
Ubiksolve Tonto EuroBen NWChem GAMESS AM2 CAM CLM CCSM.

I think those are all Fortran and all available in some form or
another, but I probably goofed once or twice.

-- greg

Derek M Jones

2005-02-25, 3:59 am

Greg,

> Derek M Jones <derek@NOSPAMknosof.co.uk> wrote:
>
>
> Google is your friend: AIPS MM5 WRF POP AIPS "ASCI Purple Benchmarks"
> Ubiksolve Tonto EuroBen NWChem GAMESS AM2 CAM CLM CCSM.


Great suggestions Greg.

Individually most of these keywords appeared on the first
page returned by a Google search and there were some very
large programs to be found :-)

A few of the packages come with specificly written licenses.
People are usually happy to let me measure away, so these
should not be a problem.

> I think those are all Fortran and all available in some form or
> another, but I probably goofed once or twice.


You typed AIPS twice, but whose counting. Some of the
packages appeared to have some associated C (this is often
used for interfacing).

I gave up trying to narrow down the search for CAM and promptly
found it on the first page I clicked on for CLM.
Greg Lindahl

2005-02-25, 3:59 am

In article <Xns96087756AE6Bderekknosofcouk@194.168.222.121>,
Derek M Jones <derek@NOSPAMknosof.co.uk> wrote:

>You typed AIPS twice, but whose counting.


It was the first million+ line program I worked on, so apparently it
is stuck in my mind :-) You'll find my name on the Tektronix 4014
driver. AIPS found 3 bugs in the PathScale pre-1.0 compiler.

> Some of the packages appeared to have some associated C (this is
> often used for interfacing).


Yeah, purity is such a hard thing nowadays. In the good old days (tm),
you could control tape drives more easily in Fortran than in C... and
"we" actually cared about non-Unix systems...

-- greg


Sponsored Links







Also available: Server administration forum archive | Web Design forum archive | Software forum archive | Hardware reviews archive

Copyright 2008 codecomments.com