| GDL - a little uppdate... [message #43729 is a reply to message #43565] |
Fri, 22 April 2005 18:55   |
Y.T.
Messages: 25
Registered: December 2004
|
Junior Member |
|
|
m_schellens@hotmail.com wrote:
> GDL - GNU Data Language, the free IDL clone
>
>
> Now with z-buffer device, new subroutines, many improvements,
> important bug fixes...
>
> GDL now understands main programs and the .RUN command :-)
>
[ etc etc ]
So I had a look at that, and I am a lot less impressed than I was when
I first heard that there is such a thing as a free IDL clone.
As far as I can figure out, GDL uses the 'readline' library to receive
input from the user, the 'gsl' library to perform the math and 'plplot'
to output the result -- in other words, it is really just a parser that
connects some other pieces.
Now by itself that isn't a problem, but I do have to wonder about the
choice -- why gsl? why plplot?
Isn't gnuplot much, much, much more popular? I'm not aware of any
current linux distro that doesn't have it -- while the latest plplot
rpm for redhat that I can find is for 7.3, and there don't seem to be
any newer .debs or slack-tgzs either.
And why gsl? That ain't exactly mainstream either (better than plplot,
though).
The more I think about it, the more obvious it appears to me to write
an IDL clone in python, which is popular and widespread, has it's own
array-math routines (which compete with IDL's in speed and versatility)
and through things like python/Tk or python/gtk even has a native link
into the OSes windowing environment.
Maybe I'm naive about something here. Maybe I'm missing something.
But after three days of failed attempts to get GSL and plplot to
compile properly ("[error 1]", why thank you for that valuable
information) on two otherwise perfectly vanilla machines (one RH9 and
one Slack10) I'm just a little puzzled and maybe just a tad frustrated.
Well, OK, maybe quite a bit frustrated. ;)
Why would someone make a current, live, under-development-right-now
project dependent on things that haven't been included in a distro in
three years? That require five year old software tools to be built?
Well, I haven't given up yet and I'm still trying -- but this obviously
requires a pretty hefty pot of coffee before it'll run anywhere...
I'll let y'all know if/when I ever manage to get this working.
cordially
Y.T.
--
Remove YourClothes before you email me.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
