Re: Need a clue about compile failure in 2.1.13

[Thomas Walter]

>>>>> "Scott" == O Scott Sands <address@hidden> writes:

[snip]

    Scott> ....back to plplot ...

    Scott> oss

Hello
I got it 'plplot-990122' together with 'octave-2.0.13.96' without
TK/TCL, only the Xwin driver.
The only piece which does not work is the plotting of the symbols.  It
does not find the fonts, I guess.

For those who may be interested in compiling for Linux:

plplot-990122:

configure --with-double --with-opt --disable-mskermit --disable-conex
--disable-vlt --disable-versaterm --disable-dg300 --disable-imp
--disable-tcl --disable-tk --disable-f77

One has to remove 'stupidtk.[co]' and -lMatrix in 'cf/lib_sh_linux.in'
in the original sources of plplot.

Then use 'plplot_octave' without applying any patches.

call
  env SH_LDFLAGS='-shared -Wl,-rpath,/usr/X11R6/lib -Wl,-rpath,/usr/local/lib' 
mkoctfile plplot_octave.cc -L/usr/local/lib -lplplotdX -lX11 -lm

The problems:
  1) plplot installs in 'usr/local/plplot'
     solution:  create links or something similar
  2) mkoctfile complains about not having a rule to work with
          octave_value (int &)
    work around: convert each int-argument in plplot_octave.cc to
    'double' with a cast.

Excuse me if it does not really belong to this mailing list, but I
read so much about plplot in this list.

Regards
Thomas


-- 

"Beweiss" fuer die These: 'Alle ungeraden Zahlen sind prim.'
3, 5, 7, 11, alle sind ungerade und prim.
9 ist ist ein irrelevanter Messfehler.
==> Bingo!  8-))))))

----------------------------------------------
Dipl. Phys. Thomas Walter
Inst. f. Physiklische Chemie II
Egerlandstr. 3                          Tel.: ++9131-85 27326 / 27330
91058 Erlangen, Germany                 email: address@hidden