User:Andy1978: Difference between revisions

From Octave
Jump to navigation Jump to search
 
(40 intermediate revisions by 3 users not shown)
Line 1: Line 1:


== ToDo ==
== ToDo ==


=== Core ===
=== Core ===
* fix imread bug with opaque images, see https://savannah.gnu.org/bugs/?func=detailitem&item_id=36820
* <strike>fix imread bug with opaque images, see https://savannah.gnu.org/bugs/?func=detailitem&item_id=36820</strike> Thanks Carne
* remove removed but still referenced function in "seealso" blocks
* <strike>FFTW multithreaded</strike> hg.savannah.gnu.org/hgweb/octave/rev/bde7731b2b83
* <strike>FFTW multithreaded https://savannah.gnu.org/patch/?7922</strike>
* <strike>remove removed but still referenced function in "seealso" blocks</strike>
 
* Add swisdom (single prec.) to /libinterp/dldfcn/fftw.cc documentation
* linguist-qt4: octave-src/libgui$ "lupdate -recursive -no-obsolete -locations relative src qterminal graphics -ts languages/*ts"
 
=== GUI ===
<strike>When opening and closing the GUI HDF5: infinite loop closing library</strike> [http://hg.savannah.gnu.org/hgweb/octave/rev/9c03b071fd7b]
 
No icons in qt4 GUI?
  gconftool-2 --type boolean --set /desktop/gnome/interface/buttons_have_icons true
  gconftool-2 --type boolean --set /desktop/gnome/interface/menus_have_icons true


=== Forge ===
=== Forge ===
==== signal or fltk plotting? ====
  pkg load signal
  t=0:1/4e4:12;
  y=chirp(t,100,5,18000);
  graphics_toolkit fltk
  specgram(y)
-> r300: Implementation error: Render targets are too big in r300_set_framebuffer_state, refusing to bind framebuffer state!
==== video ====
<strike>fix build, see http://octave.1599824.n4.nabble.com/quot-package-video-is-empty-quot-td4656346.html</strike>
==== signal ====
* findpeaks.m, print error if data is not a column vector (error: vertical dimensions mismatch (1x10000 vs 1x1)


==== image ====
==== image ====


* implement [http://en.wikipedia.org/wiki/SURF SURF], integralImage is cumsum(cumsum(a,1),2)
* implement [http://en.wikipedia.org/wiki/SURF SURF], integralImage is cumsum(cumsum(a,1),2)
* implement impixelinfo http://stackoverflow.com/questions/23706937/displaying-pixel-values-of-an-image-in-octave
* implement getpixelposition, axes2pix, pix2map, map2pix


===== corner/cornermetric, harris =====
===== corner/cornermetric, harris =====


First post on mailing list in 12.01.2013 http://octave.1599824.n4.nabble.com/corner-cornermetric-equivalent-in-octave-td4648802.html
First post on mailing list in 12.01.2013 http://octave.1599824.n4.nabble.com/corner-cornermetric-equivalent-in-octave-td4648802.html
== brainstorming ==
* something like autobuilder for octave would be cool. Jordi mentioned http://hydra.nixos.org/jobset/gnu/octave-default
* perhaps http://www.phoronix-test-suite.com/?


== Snippets ==
== Snippets ==
Line 34: Line 56:
  mtimes(num2cell(m,[1,2]){:})
  mtimes(num2cell(m,[1,2]){:})
  mtimes(s{:})
  mtimes(s{:})
=== commands from which I regularly doesn't find the name ===
* octave_config_info ()
* merge (C's (a)? b: c); syntax
=== libgraphicsmagick++3 quantum-depth ===
in debian wheezy
sudo apt-get build-dep libgraphicsmagick++3
apt-get source libgraphicsmagick++3
edit debian/rules
--with-quantum-depth=16 \
works without problems
--with-quantum-depth=32 \
shows one error:
UYVY format ...
mean-error=0.0209018218321559, maximum-error=0.816509137487552
not ok 77
Failed 1/77 subtests
debuild -i -us -uc -b
dpkg -i ....


== Tracking octave bugs with hg bisect==
== Tracking octave bugs with hg bisect==
I had a strange problem when loading gzip compressed ascii files in octave. It failed dependent on the integer values. After some stripping I made a minimalistic test script (min_testcase_fails.m) which always fails in a current dev (88616c872933):
I had a [http://savannah.gnu.org/bugs/?39652 strange problem] when loading gzip compressed ascii files in octave. It failed dependent on the integer values. After some stripping I made a minimalistic test script (min_testcase_fails.m) which always fails in a current dev (88616c872933):
  fn="file2.txt"
  fn="file2.txt"
  fid = fopen (fn,"w");
  fid = fopen (fn,"w");
Line 50: Line 96:
  system(cmd);
  system(cmd);
  c=load("file2.txt.gz")  #this fails in newer versions
  c=load("file2.txt.gz")  #this fails in newer versions
When run:
$ octave -q min_testcase_fails.m
fn = file2.txt
cmd = gzip file2.txt -c > file2.txt.gz
error: value on right hand side of assignment is undefined
error: called from:
error:  /home/andy/src/octave-bugs/min_testcase_fails.m at line 17, column 2


If line 4 with "663, 15, 154, 978161" is changed to ""663, 16, 154, 978161", load works as expected. '''Strange, isn't it?'''
If line 4 with "663, 15, 154, 978161" is changed to ""663, 16, 154, 978161", load works as expected. '''Strange, isn't it?'''
Line 71: Line 125:
  make -j 7
  make -j 7
After compilation I tried my test min_testcase_fails.m:
After compilation I tried my test min_testcase_fails.m:
./run-octave -q ../min_testcase_fails.m
<pre>~/src/octave-build$ ./run-octave -q ../min_testcase_fails.m
fn = file2.txt
cmd = gzip file2.txt -c > file2.txt.gz
c =


good or bad?
      639      25      160  978160
      687      25      171  978160
      663      31      173  978161
      663      15      154  978161
      655      21      151  978161
</pre>
So this is apparently a good one. Tell it hg bisect!


  cd ../octave-src && hg bisect -g
  cd ../octave-src && hg bisect -g


After this you try make again, bootstrap && configure if this fails, run the testscript, tell hg bisect if its good or bad and repeat this until the revision which introduced the problem is found.
After this you try make again, bootstrap && configure if make fails, rerun the testscript, tell hg bisect if it's good or bad and repeat this until the revision which introduced the problem is found.
Or you can use "hg bisect --command", see next point.
Or you can use "hg bisect --command", see next point.


Line 100: Line 163:
  $BUILDDIR/run-octave -q ~/src/min_testcase_fails.m || exit 1
  $BUILDDIR/run-octave -q ~/src/min_testcase_fails.m || exit 1
  exit 0
  exit 0
After making it executable, let hg bisect use it
~/src/octave-src$ hg bisect -c ../bisect-loadsave.sh
Now sit back and relax, this could last some hours... Finally:
<pre>Änderungssatz 16554:03a28487fa9d: good
Die erste fehlerhafte Revision ist:
Änderung:        16555:04fb96f4bea1
Nutzer:          John W. Eaton <jwe@octave.org>
Datum:          Tue Apr 23 12:57:16 2013 -0400
Zusammenfassung: allow double-click in file browser to load data files</pre>
=== bisect bug 42612 ===
For bug #42612 I created a minimal script which fails: clf_fails_42612.m
subplot (2,1,1);
plot (1:5);
legend ('hello', 'location', 'northwestoutside');
clf
And the script which runs it, bisect_clf.sh:
#!/bin/bash
SRCDIR=~/src/octave-untouched-src
BUILDDIR=~/src/bisect_build
cd $BUILDDIR
## Try a simple build first
if  ! make -j 6
then
  # make failed, run bootstrap and configure
  cd $SRCDIR
  ./bootstrap || exit 127
  cd $BUILDDIR
  $SRCDIR/configure FFLAGS=-g CFLAGS=-g CXXFLAGS=-g --enable-bounds-check --disable-java --disable-gui --disable-docs|| exit 127
  make -j 7|| exit 127
fi
cd $SRCDIR
$BUILDDIR/run-octave -q clf_fails_42612.m
ret=$?
echo $(hg id) $ret >> clf_fails_42612.log
exit $ret
=== show bisect history ===
hg log -r "bisect(good) or bisect(bad)" --template "{rev}:{node|short} {bisect}\n"
== Links ==
* [http://hg.code.sf.net/p/octave/image hgweb on sourceforge]
* [http://mywiki.wooledge.org/BashFAQ/073 bash parameter expansion]

Latest revision as of 20:09, 5 October 2016

ToDo[edit]

Core[edit]

  • Add swisdom (single prec.) to /libinterp/dldfcn/fftw.cc documentation
  • linguist-qt4: octave-src/libgui$ "lupdate -recursive -no-obsolete -locations relative src qterminal graphics -ts languages/*ts"

GUI[edit]

When opening and closing the GUI HDF5: infinite loop closing library [1]

No icons in qt4 GUI?

 gconftool-2 --type boolean --set /desktop/gnome/interface/buttons_have_icons true
 gconftool-2 --type boolean --set /desktop/gnome/interface/menus_have_icons true

Forge[edit]

signal or fltk plotting?[edit]

 pkg load signal
 t=0:1/4e4:12;
 y=chirp(t,100,5,18000);
 graphics_toolkit fltk
 specgram(y)

-> r300: Implementation error: Render targets are too big in r300_set_framebuffer_state, refusing to bind framebuffer state!

video[edit]

fix build, see http://octave.1599824.n4.nabble.com/quot-package-video-is-empty-quot-td4656346.html

signal[edit]

  • findpeaks.m, print error if data is not a column vector (error: vertical dimensions mismatch (1x10000 vs 1x1)

image[edit]

corner/cornermetric, harris[edit]

First post on mailing list in 12.01.2013 http://octave.1599824.n4.nabble.com/corner-cornermetric-equivalent-in-octave-td4648802.html

Snippets[edit]

chain matrix multiplication[edit]

a=[1 2; 3 4];
b=[4 2; 8 1];
c=[2 3; 1 -1];
m=cat(3,a,b,c);
s=cell(3,1);
s{1}=a;s{2}=b; s{3}=c;
a*b*c
s{1}*s{2}*s{3}
mtimes(num2cell(m,[1,2]){:})
mtimes(s{:})

commands from which I regularly doesn't find the name[edit]

  • octave_config_info ()
  • merge (C's (a)? b: c); syntax

libgraphicsmagick++3 quantum-depth[edit]

in debian wheezy

sudo apt-get build-dep libgraphicsmagick++3
apt-get source libgraphicsmagick++3

edit debian/rules

--with-quantum-depth=16 \

works without problems

--with-quantum-depth=32 \

shows one error:

UYVY format ...
mean-error=0.0209018218321559, maximum-error=0.816509137487552
not ok 77
Failed 1/77 subtests 
debuild -i -us -uc -b
dpkg -i ....

Tracking octave bugs with hg bisect[edit]

I had a strange problem when loading gzip compressed ascii files in octave. It failed dependent on the integer values. After some stripping I made a minimalistic test script (min_testcase_fails.m) which always fails in a current dev (88616c872933):

fn="file2.txt"
fid = fopen (fn,"w");
fprintf(fid, "%i %i %i %i\n",639, 25, 160, 978160);
fprintf(fid, "%i %i %i %i\n",687, 25, 171, 978160);
fprintf(fid, "%i %i %i %i\n",663, 31, 173, 978161);
fprintf(fid, "%i %i %i %i\n",663, 15, 154, 978161);
fprintf(fid, "%i %i %i %i\n",655, 21, 151, 978161);
fclose(fid);
## gzip it!
fn_gz=strcat(fn, ".gz");
cmd=cstrcat("gzip ",fn," -c > ",fn_gz)
system(cmd);
c=load("file2.txt.gz")  #this fails in newer versions

When run:

$ octave -q min_testcase_fails.m 
fn = file2.txt
cmd = gzip file2.txt -c > file2.txt.gz
error: value on right hand side of assignment is undefined
error: called from:
error:   /home/andy/src/octave-bugs/min_testcase_fails.m at line 17, column 2

If line 4 with "663, 15, 154, 978161" is changed to ""663, 16, 154, 978161", load works as expected. Strange, isn't it? I also had an old build (dev ab1c6e6d1be6 from Sun Oct 28 21:48:02 2012) and "load" worked in this version.

Running bisect[edit]

I decided to make a fresh clone in a separate directory:

cd ~/src
hg clone http://www.octave.org/hg/octave octave-src
cd octave-src
hg bisect -b 88616c872933
hg bisect -g ab1c6e6d1be6

After this hg bisect uses a binary search strategy to test revisions:

Teste Änderungssatz 16428:f016a5342e19 (1380 Änderungssätze verbleiben, ~10 Tests)

So after checkout I had to compile the source tree (used ccache in the hope of faster checkout-compile cycles).

./bootstrap
cd .. && mkdir octave-build && cd octave-build
export CXX="ccache g++"
export CC="ccache gcc"
../octave-src/configure
make -j 7

After compilation I tried my test min_testcase_fails.m:

~/src/octave-build$ ./run-octave -q ../min_testcase_fails.m
fn = file2.txt
cmd = gzip file2.txt -c > file2.txt.gz
c =

      639       25      160   978160
      687       25      171   978160
      663       31      173   978161
      663       15      154   978161
      655       21      151   978161

So this is apparently a good one. Tell it hg bisect!

cd ../octave-src && hg bisect -g

After this you try make again, bootstrap && configure if make fails, rerun the testscript, tell hg bisect if it's good or bad and repeat this until the revision which introduced the problem is found. Or you can use "hg bisect --command", see next point.

hg bisect --command[edit]

I used Jordis bisect script for bug 32818 and modified it to my needs:

~/src$ cat bisect-loadsave.sh 
#!/bin/bash
SRCDIR=~/src/octave-src
BUILDDIR=~/src/octave-build
cd $SRCDIR
## Try a simple build first
if  ! make -j 7
then
   cd $SRCDIR
   ./bootstrap || exit 127
   #rm -rf $BUILDDIR
   #mkdir $BUILDDIR
   cd $BUILDDIR
   $SRCDIR/configure || exit 127
   make -j 7|| exit 127
fi
$BUILDDIR/run-octave -q ~/src/min_testcase_fails.m || exit 1
exit 0

After making it executable, let hg bisect use it

~/src/octave-src$ hg bisect -c ../bisect-loadsave.sh

Now sit back and relax, this could last some hours... Finally:

Änderungssatz 16554:03a28487fa9d: good
Die erste fehlerhafte Revision ist:
Änderung:        16555:04fb96f4bea1
Nutzer:          John W. Eaton <jwe@octave.org>
Datum:           Tue Apr 23 12:57:16 2013 -0400
Zusammenfassung: allow double-click in file browser to load data files


bisect bug 42612[edit]

For bug #42612 I created a minimal script which fails: clf_fails_42612.m

subplot (2,1,1);
plot (1:5);
legend ('hello', 'location', 'northwestoutside');
clf

And the script which runs it, bisect_clf.sh:

#!/bin/bash
SRCDIR=~/src/octave-untouched-src
BUILDDIR=~/src/bisect_build
cd $BUILDDIR
## Try a simple build first
if  ! make -j 6
then
  # make failed, run bootstrap and configure
  cd $SRCDIR
  ./bootstrap || exit 127
  cd $BUILDDIR
  $SRCDIR/configure FFLAGS=-g CFLAGS=-g CXXFLAGS=-g --enable-bounds-check --disable-java --disable-gui --disable-docs|| exit 127
  make -j 7|| exit 127
fi
cd $SRCDIR
$BUILDDIR/run-octave -q clf_fails_42612.m
ret=$?
echo $(hg id) $ret >> clf_fails_42612.log
exit $ret

show bisect history[edit]

hg log -r "bisect(good) or bisect(bad)" --template "{rev}:{node|short} {bisect}\n"

Links[edit]