Bug#534281: nosql: uninstallable on ia64, alpha, GNU/kFreeBSD

2009-06-29 Thread Cyril Brulebois
Petr Salinger  (23/06/2009):
> Package: nosql
> Severity: serious
> Version: 4.0.14-4
> User: glibc-bsd-de...@lists.alioth.debian.org
> Usertags: kfreebsd
>
> Hi,
>
> current version have hardcoded dependency on libc
> as Depends: libc6 (>= 2.2.3-1).
> This cannot be satisfied on some architectures.
> It have been introduced in
>
>  nosql  (4.0.14-2) unstable; urgency=low
>* add dependency to libc (warning from lintian)
>
> Please at least drop this dependency or generate
> correct one during build using i.e. dh_shlibdeps.

Hi,

cc'ing -qa@ because I won't be fixing the whole package, and it really
needs it. I think it'd be a nice exercise for a wannabe packager to make
it use debhelper instead of that broken homebrew stuff.

Hint: one needs to get rid of debian/substvars, at the very least.

Another idea might be to get rid of the package itself, but I won't be
judging that here.

Thanks for your time.

Mraw,
KiBi.


signature.asc
Description: Digital signature


Bug#533361: [grahn: Henning's patch tested, and kind of working]

2009-06-29 Thread Henning Makholm
Scripsit Jörgen Grahn

> Just in case you too do not get mail notifications whenever a Debian
> bug gets updated.  I wrote this:

Oops, thanks for the heads-up. I meant to subscribe to the PTS for
xcftools, but never got it done after I resigned as a Debian dev a
couple of months ago.

> Henning, sorry for not replying earlier -- I somehow believed the BTS
> would Cc: me on everything, and didn't check the web interface until
> today.

I thought that too -- otherwise I'd have Cc'ed you directly.

> However, the resulting image still looks strange.
...
> What L2.png actually contains is the 10x10 canvas-visible part and
> everything to the south and east of that. The rest is clipped, i.e.
> filled -b white.

Yes, this is a bug. There were a few more places where I had stored
canvas-based coordinates in unsigned variables. Embarassing.

The attached patch ought to fix fixed all known cases. It is based on
the clean 1.0.4 source.

I don't think any of the newly fixed places are security
relevant. They just cause parts of the source image to be ignored, or
in certain cases wrong pixels to be read. So if someone does a
security upload to lenny, the previous patch is probably the least
invasive one to use.

I plan to release 1.0.5 with the fixes Thursday or Friday -- other
stuff gets in the way until then.

> Also, for me personally, this doesn't matter because now that xcf2pnm
> no longer crashes, I can see that I do not want the -C effect. I want
> my layers clipped to the canvas, just like xcf2pnm does by default.

I will try to think of improvements to the wording of the manpage such
that others will not be similarly confused. Suggestions welcome.

-- 
Henning Makholm "Al lykken er i ét ord: Overvægtig!"
? 533361b.patch
? autotools-stamp
? configure-stamp
? foo.png
? test-stamp
Index: flatten.c
===
RCS file: /home/makcvs/repository/source/xcftools/flatten.c,v
retrieving revision 1.27
diff -u -r1.27 flatten.c
--- flatten.c	22 Feb 2006 00:01:04 -	1.27
+++ flatten.c	29 Jun 2009 22:10:22 -
@@ -619,14 +619,14 @@
   fillTile(&toptile,0);
 
   for( where.t = spec->dim.c.t; where.t < spec->dim.c.b; where.t=where.b ) {
-where.b = (where.t+TILE_HEIGHT) - where.t % TILE_HEIGHT ;
+where.b = TILE_TOP(where.t)+TILE_HEIGHT ;
 if( where.b > spec->dim.c.b ) where.b = spec->dim.c.b ;
 nrows = where.b - where.t ;
 for( y = 0; y < nrows ; y++ )
   rows[y] = xcfmalloc(4*(spec->dim.c.r-spec->dim.c.l));
 
 for( where.l = spec->dim.c.l; where.l < spec->dim.c.r; where.l=where.r ) {
-  where.r = (where.l+TILE_WIDTH) - where.l % TILE_WIDTH ;
+  where.r = TILE_LEFT(where.l)+TILE_WIDTH ;
   if( where.r > spec->dim.c.r ) where.r = spec->dim.c.r ;
   ncols = where.r - where.l ;
 
Index: pixels.c
===
RCS file: /home/makcvs/repository/source/xcftools/pixels.c,v
retrieving revision 1.14
diff -u -r1.14 pixels.c
--- pixels.c	13 Feb 2006 03:04:03 -	1.14
+++ pixels.c	29 Jun 2009 22:10:22 -
@@ -361,8 +361,8 @@
   if( isSubrect(want,dim->c) &&
   (want.l - dim->c.l) % TILE_WIDTH == 0 &&
   (want.t - dim->c.t) % TILE_HEIGHT == 0 ) {
-unsigned tx = (want.l - dim->c.l) / TILE_WIDTH ;
-unsigned ty = (want.t - dim->c.t) / TILE_WIDTH ;
+int tx = TILE_NUM(want.l - dim->c.l);
+int ty = TILE_NUM(want.t - dim->c.t);
 if( want.r == TILEXn(*dim,tx+1) && want.b == TILEYn(*dim,ty+1) ) {
   /* The common case? An entire single tile from the layer */
   copyTilePixels(tile,tiles->tileptrs[tx + ty*dim->tilesx],tiles->params);
@@ -375,8 +375,10 @@
 unsigned width = want.r-want.l ;
 rgba *pixvert = tile->pixels ;
 rgba *pixhoriz ;
-unsigned y, ty, l0, l1, lstart, lnum ;
-unsigned x, tx, c0, c1, cstart, cnum ;
+int y, ty, l0, l1 ;
+int x, tx, c0, c1 ;
+unsigned lstart, lnum ;
+unsigned cstart, cnum ;
 
 if( !isSubrect(want,dim->c) ) {
   if( want.l < dim->c.l ) pixvert += (dim->c.l - want.l),
@@ -394,7 +396,7 @@
 fprintf(stderr,"jig0 (%d-%d),(%d-%d)\n",left,right,top,bottom);
 #endif
 
-for( y=want.t, ty=(want.t-dim->c.t)/TILE_HEIGHT, l0=TILEYn(*dim,ty);
+for( y=want.t, ty=TILE_NUM(want.t-dim->c.t), l0=TILEYn(*dim,ty);
  y want.b ? want.b : l1) - y ;
   
   pixhoriz = pixvert ;
-  for( x=want.l, tx=(want.l-dim->c.l)/TILE_WIDTH, c0=TILEXn(*dim,tx);
+  for( x=want.l, tx=TILE_NUM(want.l-dim->c.l), c0=TILEXn(*dim,tx);
x> TILE_SHIFT)
 
 struct tileDimensions {
   struct rect c ;


Bug#535120: hk-classes FTBFS with python2.6

2009-06-29 Thread Fabrice Coutadeur
Package: hk-classes
Version: 0.8.3-5
Severity: serious
Tags: patch
Justification: no longer builds from source
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu karmic ubuntu-patch

In Ubuntu, python2.6 is the default python version, and this package has
to be adapted to build properly. Here is the patch for that

*** /tmp/tmpO_WB5c
In Ubuntu, we've applied the attached patch to achieve the following:

  * debian/patches/python2.6.patch: patch python/Makefile.in to add parameter
--install-layout=deb to setup.py install call (LP: #392390)

We thought you might be interested in doing the same. 


-- System Information:
Debian Release: squeeze/sid
  APT prefers karmic-updates
  APT policy: (500, 'karmic-updates'), (500, 'karmic-security'), (500, 'karmic')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.28-13-generic (SMP w/2 CPU cores)
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
Shell: /bin/sh linked to /bin/dash
--- hk-classes-0.8.3.orig/debian/patches/python2.6.patch
+++ hk-classes-0.8.3/debian/patches/python2.6.patch
@@ -0,0 +1,30 @@
+diff -Nur -x '*.orig' -x '*~' hk-classes-0.8.3/python/Makefile.am 
hk-classes-0.8.3.new/python/Makefile.am
+--- hk-classes-0.8.3/python/Makefile.am2006-05-23 18:02:11.0 
+
 hk-classes-0.8.3.new/python/Makefile.am2009-06-26 04:07:47.0 
+
+@@ -10,9 +10,9 @@
+ 
+ install-data-hook:
+   if test -z "$(DESTDIR)"; then \
+- python setup.py install ; \
++ python setup.py install --install-layout=deb ; \
+   else \
+- python setup.py install --root $(DESTDIR) ; \
++ python setup.py install --install-layout=deb --root $(DESTDIR) ; \
+   fi
+   /bin/rm -f -R ./build/*
+ 
+diff -Nur -x '*.orig' -x '*~' hk-classes-0.8.3/python/Makefile.in 
hk-classes-0.8.3.new/python/Makefile.in
+--- hk-classes-0.8.3/python/Makefile.in2006-12-17 14:27:52.0 
+
 hk-classes-0.8.3.new/python/Makefile.in2009-06-26 04:07:00.0 
+
+@@ -371,9 +371,9 @@
+ 
+ install-data-hook:
+   if test -z "$(DESTDIR)"; then \
+- python setup.py install ; \
++ python setup.py install --install-layout=deb ; \
+   else \
+- python setup.py install --root $(DESTDIR) ; \
++ python setup.py install --install-layout=deb --root $(DESTDIR) ; \
+   fi
+   /bin/rm -f -R ./build/*
+ 


Re: MMA package on debian

2009-06-29 Thread Jan Hauke Rahm
On Mon, Jun 29, 2009 at 09:26:09AM -0700, Bob van der Poel wrote:
>I'm the  author of the MMA package and the version on the servers is very
>old. A newer one is on my website at http://www.mellowood.ca/mma which
>includes  a .deb.
> 
>The original maintainer of the package has abandoned doing the .deb MMA 
>thing.
> 
> So, how do I get this replaced on the servers in a easy to do fashion.

You might be pleased to hear that there is already someone working on it
as you can see here:

http://mentors.debian.net/cgi-bin/sponsor-pkglist?action=details;package=mma

Kind regards,
Hauke


signature.asc
Description: Digital signature


Рекламa в интернете

2009-06-29 Thread Артем Веб-Мастер
Доброе время суток!

Меня зовут Артем, я представляю рекламное интернет-агентство.

Наша компания, на протяжении многих лет занимается продвижением товаров и 
услуги в сети.

Предлагаем различные методы:

- Рекламные маcсовые paccылки - от 2 000 рублей.
- Хостинг от 150 рублей в месяц.
- Регистрация доменных имен - от 190 рублей.
- Разработка сайтов - от 6 900 рублей.
- Комплекное продвижение в сети интернет - от 3 900 рублей.

Принимаем любые методы оплаты.
От Нашей рекламы всегда есть результат.

Не стесняйтесь, обрайщайтесь по любым вопросам:
Телефон: (495)585-6209
ICQ: 38 376 71 14


-- 
To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



MMA package on debian

2009-06-29 Thread Bob van der Poel
I'm the  author of the MMA package and the version on the servers is very
old. A newer one is on my website at http://www.mellowood.ca/mma which
includes  a .deb.

The original maintainer of the package has abandoned doing the .deb MMA
thing.

 So, how do I get this replaced on the servers in a easy to do fashion.

-- 
 Listen to my CD at http://www.mellowood.ca/music/cedars 
Bob van der Poel ** Wynndel, British Columbia, CANADA **
EMAIL: b...@mellowood.ca
WWW:   http://www.mellowood.ca