[Pkg-octave-devel] octave-forge_2004.11.16-4 accepted

[Dirk Eddelbuettel]

On 23 February 2005 at 16:43, Rafael Laboissiere wrote:
| octave-forge_2004.11.16-4 is already in unstable.  This version of the
| package is compliant with our Guidelines, such that all *.oct are installed
| in api-specific directories instead of version-specific directories. Also,
| the package depends on octave2.1 >= 2.1.65, without superior limit.  This
| means that the package will not become automatically uninstallable anymore
| when a new upstream version of Octave enters unstable.  However, we must
| rebuild octave-forge each time time the api-version number is incremented.

I think that is the best compromise. I used the use these "half-open
intervals" for a long time. More recently, we experimented with simulatenous
<< octave2.1_$N++ --- and I think that was a failure.

The new API string methods offers us a chance to not having to rebuild,
depending on Octave changes. That is a good thing, and like the numerous
other changes implemented by the new Debian Octave, shows that I probably
hogged Octave maintenance for a little too long :)

Thanks to all of your for all the new work. It is really good to Octave
vibrant upstream and in Debian.

| An aside note: buildd of octave2.1 on m68k failed after 20h41min with the
| usual error message:
| 
| /usr/bin/g++ -c  -fPIC -I. -I.. -I../liboctave -I../src -I../libcruft/misc  -I../glob -I../glob -DHAVE_CONFIG_H  -Wall -W -Wshadow -O1 -g0 ls-mat5.cc -o pic/ls-mat5.o
| /tmp/cchqGuhc.s: Assembler messages:
| /tmp/cchqGuhc.s:20898: Error: operands mismatch -- statement `move.b %a4,%d1' ignored

I see the -01 -g0 --- for R I actually use -O0 -g0 on m68k. I may be worth a
try with -O0 -g0 on m68k.

| Buildd of octave-forge will obviously fail on m68k.

I won't comment on that any more given the long thread I spawned on
debian-devel.  

Dirk

-- 
Better to have an approximate answer to the right question than a precise 
answer to the wrong question.  --  John Tukey as quoted by John Chambers