> Now, we need G95, but this package is not availlable with synaptic. Ubuntu
> has gfortran but I do not know if it is possible to compile MAP-65 with it
> or not (Any idea ?). I do not remember very well, but I think that we had
> some issues about using gfortran.
>
> G95 sources are availlable on their website<http://www.g95.org/>, sources
> are available and even some debian package. Perhaps I could use these...
MAP65 can be compiled equally well with g95 or gfortran.
-- Joe, K1JT
F1EBK wrote:
> So after installing Gfortran, I get the latest sources using the command
> in a terminal :
>
> svn co http://svn.berlios.de/svnroot/repos/wsjt/branches/map65
You should no longer use the Belios repository, it is being phased out.
The commend to check out MAP65 source from the new WSJT repository at
SourceForge is:
svn co https://svn.code.sf.net/p/wsjt/code/branches/map65
You will need to have registered at SourceForge so that you have a
SourceForge username.
To compile MAP65:
$ ./configure
$ make
To run MAP65:
$ python map65.py
-- 73, Joe, K1JT
F1EBK wrote:
> Good news, today, I downloaded the latest version of MAP65 after several
> days without any try. This is revision 22.
>
> Compiling as usual, I can start map65. At this time, I cannot get data from
> the network, but I have several files recorded during the ARRL EME. During
> the decode process I have these errors :
>
> sh: kvasd_g95: not found
> Error in KV decoder, or no KV decoder present.
> Return code: 32512. Will use BM algorithm.
> sh: kvasd_g95: not found
Obviously you must find out why kvasd_g95 was now found during your
execution of
$ python map65.py
Program kvasd_g95 is a compiled version of the Koetter-Vardy algebraic
soft-decision Reed Solomon decoder. It must be present and executable
in the directory from which you run MAP65. It is part of the svn
distribution of code for MAP65.
-- Joe, K1JT
> With version 26, I cannot compile as I have this error :
> ...
MAP65 v1.2 r26 is not a released version.
At any particular time you might find that the HEAD of the SVN
repository will not compile on your system.
Please remember that the repository is for development purposes! When
there is a good reason to have the latest code running on your system, I
will make some announcement.
-- 73, Joe, K1JT
There is no need to do what you have suggested.
The whole point of a repository system like SVN is that at any time you
can obtain any past revision of the source code. For example, to obtain
r22 of the source code for MAP65, execute the command
svn co -r22 https://map65.svn.sourceforge.net/svnroot/map65
-- Joe, K1JT
Debian 6.03
Kernel Linux 2.6.32-5-686
Necessary Packages:
libfftw3-3 3.2.2-1
libfftw3-dev 3.2.2-1
subversion 1.6.12dfsg-6
python-imaging-tk 1.1.7-2
python-numpy 1:1.4.1-5
python-pmw 1.3.2-5
python2.6 2.6.6-8+b1
python2.6-dev 2.6.6-8+b1
libportaudio2 19+svn20071022-3.2
libsamplerate0-dev 0.1.7-3
gfortran
4:4.4.5-1
portaudio19-dev 19+svn20071022-3.2
FEDORA 16
Kernel Linux 3.1.2-1.fc16.i686.PA
Necessary Packages:
subversion-1.6.17-5.fc16.i686
python-devel-2.7.2-5.2.fc16.i686
python-imaging-tk-1.1.7-4.fc16.i686
python-pmw-1.3.2-11.fc15.noarch
python-2.7.2-5.2.fc16.i686
fftw-devel-3.3-3.fc16.i686
fftw-3.3-3.fc16.i686
portaudio-19-12.fc16.i686
portaudio-devel-19-12.fc16.i686
numpy-1.6.0-2.1.fc16.i686
libsamplerate-devel-0.1.8-1.fc16.i686
gcc-gfortran-4.6.2-1.fc16.i686
UBUNTU 11.10
Kernel Linux 3.0.0-13-generic
Necessary Packages:
subversion 1.6.12dfsg-4ubuntu5
python2.7 2.7.2-5ubuntu1
python2.7-dev 2.7.2-5ubuntu1
python-pmw 1.3.2-6
python-imaging-tk 1.1.7-3ubuntu1
python-numpy 1:1.5.1-2ubuntu2
libfftw3-3 3.2.2-1ubuntu2
libfftw3-dev 3.2.2-1ubuntu2
libsamplerate0-dev 0.1.7-3ubuntu1
portaudio19-dev 19+svn20110326-2
libportaudio2 19+svn20110326-2
gfortran 4:4.6.1-2ubuntu5
Normally with these packages we will be able to compile Map65 (./
configure and Make).
But using the new version of gfortran it generate an error on
decode1.f90 file line 79 (sequential read or write not allowed after
EOF marker,possibility use REWIND or BACKSPACE) when running Map65
i have with Sebastien modified the program by using REWIND 21 or
BACKSPACE 21 on line 79 before WRITE (21,1001).
After a make clean , ./configure,make and python map65.py the program
run correctly.
But when i try to read tf2 file for testing, Map65 crash with same
error in other file.
It seem that it's necessary to modify the code in many other files to
be sure that Map65 will be able to run correctly.
I send my request to Joe who is able to solve our problem quickly.
Thanks for reading
73
Raymond, F6JIG
On Nov 27, 8:49 pm, Patrick DUPRE <fc1...@gmail.com> wrote:
> Hi All,
>
> I managed to compile and run MAP65 on a Fedora-14 distribution. My main
> problem with Fedora is that the package name looks like ... different ! So
> for any problem, I had to find the right package to install. As I did many
> attempts, I may have some missing information, but if you find any problem
> running MAP65 under Fedora-14, let me know, I'll be glad to help.
>
> First of all, I am not using Fedora-15 or the brand new Fedora-16, as their
> fortran compiler do not accept some programming techniques used in MAP65.
> So, It's easier to use Fedora-14 for a first go on MAP65 under Fedora ...
>
> Second, I think that it would be better to use 32bits distribution as we
> have an issue with the KVASD module on 64bits systems. This will be fixed
> in the future, but today, MAP65 cannot be used on 64 bits systems.
>
> I use *yumex* as a graphical interface to add the requested packages. I
> think you can use yum to do the same in text mode, the dependencies should
> be the same. As usual, I started for a brand new installation of Fedora-14.
> Here is the list of the requested packages :
>
> subversion
> python
> numpy
> python-imaging-tk
> python-pmw
> portaudio
> portaudio-devel
> libsamplerate-devel
> gcc-gfortran
> fftw
> fftw-devel
>
> Then, you need the latest released sources (Today, this is r.22), so we get
> them with the command :
>
> svn co -r22https://map65.svn.sourceforge.net/svnroot/map65
I've been working with Raymond F6JIG and Patrick F1EBK on map65 compilation.
I'll follow up today on this fortran runtime problem on the wsjt-devel
mailing list (or is it not used anymore?)
Regards
Sebastien, F4GRX