Bug#332588: chromium - fails to build

[Ben Hutchings]

It appears that chromium was built using an old version of
libopenal-dev, since it doesn't explicitly build-depend on it.  It
probably should depend on libopenal-dev (>= 0.2005080600-1), that being
the first version in Debian that has a sensible parameter type for
alGetProcAddress.

Ben.

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.


signature.asc
Description: This is a digitally signed message part

Bug#337871: FTBFS (alpha): invalid lvalue in assignment

[Ben Hutchings]

tags 337871 +patch
thanks

This error (use of gcc's old lvalue-cast extension) is in the gdb 6.1
code included in crash, and has been fixed in newer versions of gdb.
Perhaps crash can be built using one of those?  This is the upstream
fix:

===
RCS file: /cvs/src/src/bfd/coff-alpha.c,v
retrieving revision 1.20
retrieving revision 1.21
diff -u -r1.20 -r1.21
--- src/bfd/coff-alpha.c2004/04/30 14:23:39 1.20
+++ src/bfd/coff-alpha.c2004/06/11 14:19:32 1.21
@@ -1455,7 +1455,7 @@
  amt = sizeof (struct ecoff_section_tdata);
  lita_sec_data = ((struct ecoff_section_tdata *)
   bfd_zalloc (input_bfd, amt));
- ecoff_section_data (input_bfd, lita_sec) = lita_sec_data;
+ lita_sec->used_by_bfd = lita_sec_data;
}
 
   if (lita_sec_data->gp != 0)

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.


signature.asc
Description: This is a digitally signed message part

Bug#314671: zynaddsubfx: freezes the system if run as root

[Ben Hutchings]

tags 314671 +patch
thanks

I think this will fix the bug.  However I haven't tested it because I
looked at the surrounding code and decided never to let it run on my
system.

--- src/main.C~ 2005-12-15 05:23:14.0 +
+++ src/main.C  2005-12-15 05:23:14.0 +
@@ -90,7 +90,7 @@
  * Try to get the realtime priority
  */
 void set_realtime(){
-#ifdef OS_LINUX
+#if 0
 sched_param sc;

 sc.sched_priority=50;

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.


signature.asc
Description: This is a digitally signed message part

Bug#337802: sgt-puzzles: please enable window resizing

[Ben Hutchings]

Eric Cooper wrote:
> Please enable window resizing -- the puzzles are too small
> otherwise on a large display like 1600x1200.

I haven't found this to be a problem, but I can see that it would be
useful to make the puzzles resizable and/or to scale them according to
the physical resolution of the screen.  I shall ask Simon why he
disabled resizing in the front-end before making a change though.

By the way, it is possible to override the default sizes by setting
environment variables of the form _TILESIZE, where the
name is changed to upper-case (and doesn't include any "game" suffix
that I've added).

Ben.

-- 
Ben Hutchings
You can't have everything.  Where would you put it?


signature.asc
Description: This is a digitally signed message part

Bug#347309: ITP: libtime-piece-mysql-perl -- Time::Piece::MySQL - Adds MySQL-specific methods to Time::Piece

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>

* Package name: libtime-piece-mysql-perl
  Version : 0.05
  Upstream Author : Marty Pauley <[EMAIL PROTECTED]>
* URL : http://search.cpan.org/~kasei/Time-Piece-MySQL/
* License : dual Artistic/GPL
  Description : Time::Piece::MySQL - Adds MySQL-specific methods to 
Time::Piece

Using this module instead of, or in addition to, Time::Piece adds a
few MySQL-specific date-time methods to Time::Piece objects.

This is a dependency of libclass-dbi-mysql-perl (see #321938).


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345024: sgt-puzzles: Does not clear window margins

[Ben Hutchings]

Package: sgt-puzzles
Version: 6452-1
Severity: minor


If there is a margin between the region that the puzzle is rendered in
and the window border, this margin is never cleared (except as a
result of being obscured and exposed again).  If the user changes the
puzzle's parameters such that it has a different aspect ratio, and if
the window manager prevents the window from being resized to match,
some part of the old window contents may not be cleared.  It appears
that Gtk automatically clears such a margin for windows that are not
intended to be resizable but that are forced by the window manager to
adopt a different size from that they request, so this problem did not
appear in earlier versions.

I have a simple fix for this which will be included in the next
uploaded version, but I don't think this is important enough to deal
with immediately.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#158590: reget support for sftp

[Ben Hutchings]

tags 158590 +patch
thanks

These patches will add a "reget" command to the current sid and sarge
versions of sftp.  They are not thoroughly tested and could do with some
cleaning up, but they appear to work.

The tricky bit, as was explained to me when I raised this issue
originally, is not in resuming a download, but in preventing an
incompletely downloaded file from having gaps.  sftp can send multiple
concurrent requests for several blocks of a file and may receive them
out of order, so it is necessary for it to keep track of any gaps and
truncate the file at the first gap if a download is aborted.  (Actually
it could hold onto isolated blocks in memory, but I'm not sure that the
resulting higher memory consumption would be acceptable.)

I believe I have ensured that the truncation will be done even if a
download is aborted by a signal or fatal error, but I'm not entirely
happy about the way that's done.  It's also difficult to test this as I
believe that it is uncommon for servers to send blocks out of order.

Ben.

-- 
Ben Hutchings
The two most common things in the universe are hydrogen and stupidity.
diff -ur openssh-4.2p1-5/progressmeter.c openssh-4.2p1/progressmeter.c
--- openssh-4.2p1-5/progressmeter.c	2005-06-16 04:18:05.0 +0100
+++ openssh-4.2p1/progressmeter.c	2005-12-29 19:03:10.0 +
@@ -240,7 +240,7 @@
 	start = last_update = time(NULL);
 	file = f;
 	end_pos = filesize;
-	cur_pos = 0;
+	cur_pos = *ctr;
 	counter = ctr;
 	stalled = 0;
 	bytes_per_second = 0;
diff -ur openssh-4.2p1-5/sftp.c openssh-4.2p1/sftp.c
--- openssh-4.2p1-5/sftp.c	2005-08-22 23:06:56.0 +0100
+++ openssh-4.2p1/sftp.c	2005-12-29 19:02:32.0 +
@@ -103,6 +103,7 @@
 #define I_SYMLINK	21
 #define I_VERSION	22
 #define I_PROGRESS	23
+#define I_REGET		24
 
 struct CMD {
 	const char *c;
@@ -119,6 +120,7 @@
 	{ "dir",	I_LS },
 	{ "exit",	I_QUIT },
 	{ "get",	I_GET },
+	{ "reget",	I_REGET },
 	{ "mget",	I_GET },
 	{ "help",	I_HELP },
 	{ "lcd",	I_LCHDIR },
@@ -502,7 +504,8 @@
 }
 
 static int
-process_get(struct sftp_conn *conn, char *src, char *dst, char *pwd, int pflag)
+process_get(struct sftp_conn *conn, char *src, char *dst, char *pwd, int pflag,
+	int rflag)
 {
 	char *abs_src = NULL;
 	char *abs_dst = NULL;
@@ -554,7 +557,8 @@
 			abs_dst = tmp;
 
 		printf("Fetching %s to %s\n", g.gl_pathv[i], abs_dst);
-		if (do_download(conn, g.gl_pathv[i], abs_dst, pflag) == -1)
+		if (do_download(conn, g.gl_pathv[i], abs_dst, pflag, rflag)
+		== -1)
 			err = -1;
 		xfree(abs_dst);
 		abs_dst = NULL;
@@ -897,6 +901,7 @@
 	*path1 = *path2 = NULL;
 	switch (cmdnum) {
 	case I_GET:
+	case I_REGET:
 	case I_PUT:
 		if (parse_getput_flags(&cp, pflag))
 			return(-1);
@@ -1030,7 +1035,9 @@
 		err = -1;
 		break;
 	case I_GET:
-		err = process_get(conn, path1, path2, *pwd, pflag);
+	case I_REGET:
+		err = process_get(conn, path1, path2, *pwd, pflag,
+  cmdnum == I_REGET);
 		break;
 	case I_PUT:
 		err = process_put(conn, path1, path2, *pwd, pflag);
diff -ur openssh-4.2p1-5/sftp-client.c openssh-4.2p1/sftp-client.c
--- openssh-4.2p1-5/sftp-client.c	2005-08-02 08:07:08.0 +0100
+++ openssh-4.2p1/sftp-client.c	2005-12-29 19:02:32.0 +
@@ -735,9 +735,15 @@
 	buffer_free(&msg);
 }
 
+/* Variables used for removing isolated fragments of the downloaded file
+ * in case of a signal or fatal error */
+static volatile int truncate_fd = -1;
+static volatile u_int32_t truncate_len_upper;
+static volatile u_int32_t truncate_len_lower;
+
 int
 do_download(struct sftp_conn *conn, char *remote_path, char *local_path,
-int pflag)
+int pflag, int rflag)
 {
 	Attrib junk, *a;
 	Buffer msg;
@@ -753,10 +759,12 @@
 		u_int64_t offset;
 		TAILQ_ENTRY(request) tq;
 	};
-	TAILQ_HEAD(reqhead, request) requests;
+	TAILQ_HEAD(reqhead, request) requests, fragments;
+	u_int64_t contig_len;
 	struct request *req;
 
 	TAILQ_INIT(&requests);
+	TAILQ_INIT(&fragments);
 
 	a = do_stat(conn, remote_path, 0);
 	if (a == NULL)
@@ -799,7 +807,8 @@
 		return(-1);
 	}
 
-	local_fd = open(local_path, O_WRONLY | O_CREAT | O_TRUNC,
+	local_fd = open(local_path,
+	O_WRONLY | O_CREAT | (rflag ? 0 : O_TRUNC),
 	mode | S_IWRITE);
 	if (local_fd == -1) {
 		error("Couldn't open local file \"%s\" for writing: %s",
@@ -808,11 +817,32 @@
 		xfree(handle);
 		return(-1);
 	}
+	if (rflag) {
+		struct stat stat_buf;
+		if (fstat(local_fd, &stat_buf) == -1) {
+			error("Couldn't stat local file \"%s\": %s",
+			   local_path, strerror(errno));
+			close(local_fd);
+			buffer_free(&msg);
+			xfree(handle);
+			return(-1);
+		}
+		offset = stat_buf.st_size;
+	} else {
+		offset = 0;
+	}
 
 	/* Read from remote and write to local */
-	write_error = read_error = write_errno = num_req = offset = 0;
+	write_error = read_error = 

Bug#158590: reget support for sftp

[Ben Hutchings]

Having had a drink and another look over my previous changes, I saw how
I could and should clean this up.  Here's a patch on top of the previous
one for sarge; it should apply at an offset on the sid version as well.
It also adds the minimal documentation of an entry in the list of
commands that is available by typing "help".

Ben.

-- 
Ben Hutchings
The two most common things in the universe are hydrogen and stupidity.
diff -ur openssh-3.8.1p1-last/sftp.c openssh-3.8.1p1/sftp.c
--- openssh-3.8.1p1-last/sftp.c	2005-12-29 23:54:44.0 +
+++ openssh-3.8.1p1/sftp.c	2005-12-29 23:52:06.0 +
@@ -136,33 +136,34 @@
 help(void)
 {
 	printf("Available commands:\n");
-	printf("cd path   Change remote directory to 'path'\n");
-	printf("lcd path  Change local directory to 'path'\n");
-	printf("chgrp grp pathChange group of file 'path' to 'grp'\n");
-	printf("chmod mode path   Change permissions of file 'path' to 'mode'\n");
-	printf("chown own pathChange owner of file 'path' to 'own'\n");
-	printf("help  Display this help text\n");
-	printf("get remote-path [local-path]  Download file\n");
-	printf("lls [ls-options [path]]   Display local directory listing\n");
-	printf("ln oldpath newpathSymlink remote file\n");
-	printf("lmkdir path   Create local directory\n");
-	printf("lpwd  Print local working directory\n");
-	printf("ls [path] Display remote directory listing\n");
-	printf("lumask umask  Set local umask to 'umask'\n");
-	printf("mkdir pathCreate remote directory\n");
-	printf("progress  Toggle display of progress meter\n");
-	printf("put local-path [remote-path]  Upload file\n");
-	printf("pwd   Display remote working directory\n");
-	printf("exit  Quit sftp\n");
-	printf("quit  Quit sftp\n");
-	printf("rename oldpath newpathRename remote file\n");
-	printf("rmdir pathRemove remote directory\n");
-	printf("rm path   Delete remote file\n");
-	printf("symlink oldpath newpath   Symlink remote file\n");
-	printf("version   Show SFTP version\n");
-	printf("!command  Execute 'command' in local shell\n");
-	printf("! Escape to local shell\n");
-	printf("? Synonym for help\n");
+	printf("cd pathChange remote directory to 'path'\n");
+	printf("lcd path   Change local directory to 'path'\n");
+	printf("chgrp grp path Change group of file 'path' to 'grp'\n");
+	printf("chmod mode pathChange permissions of file 'path' to 'mode'\n");
+	printf("chown own path Change owner of file 'path' to 'own'\n");
+	printf("help   Display this help text\n");
+	printf("get remote-path [local-path]   Download file\n");
+	printf("reget remote-path [local-path] Resume download of file\n");
+	printf("lls [ls-options [path]]Display local directory listing\n");
+	printf("ln oldpath newpath Symlink remote file\n");
+	printf("lmkdir pathCreate local directory\n");
+	printf("lpwd   Print local working directory\n");
+	printf("ls [path]  Display remote directory listing\n");
+	printf("lumask umask   Set local umask to 'umask'\n");
+	printf("mkdir path Create remote directory\n");
+	printf("progress   Toggle display of progress meter\n");
+	printf("put local-path [remote-path]   Upload file\n");
+	printf("pwdDisplay remote working directory\n");
+	printf("exit   Quit sftp\n");
+	printf("quit   Quit sftp\n");
+	printf("rename oldpath newpath Rename remote file\n");
+	printf("rmdir path Remove remote directory\n");
+	printf("rm pathDelete remote file\n");
+	printf("symlink oldpath newpat

Bug#345275: ITP: maypole -- Perl framework for MVC-oriented web applications

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: maypole
  Version : 2.10
  Upstream Author : Simon Cozens <[EMAIL PROTECTED]> and others
* URL : http://maypole.cpan.org/
* License : dual GPL/Artistic
  Description : Perl framework for MVC-oriented web applications

Maypole is a Perl framework for MVC-oriented web applications, similar
to Jakarta's Struts. Maypole is designed to minimize coding requirements
for creating simple web interfaces to databases, while remaining flexible
enough to support enterprise web applications.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345277: ITP: memories -- a web-based photo sharing application

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: memories
  Version : 1.1
  Upstream Author : Simon Cozens <[EMAIL PROTECTED]>
* URL : http://maypole.perl.org/?Applications
* License : to be established
  Description : a web-based photo sharing application

Memories allows multiple users to browse and share photos.  Each
user's photos can be seen in blog format or as albums.  Each photo can
be tagged with names, places, and other descriptive terms, and found
through these tags.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.4.27-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345275: ITP: maypole -- Perl framework for MVC-oriented web applications

[Ben Hutchings]

John Goerzen wrote:
> On Fri, Dec 30, 2005 at 01:16:53AM +0000, Ben Hutchings wrote:
> > * URL : http://maypole.cpan.org/
> 
> That domain doesn't resolve.

Indeed, that should be http://maypole.perl.org/

> > * License : dual GPL/Artistic
> >   Description : Perl framework for MVC-oriented web applications
> 
> That's a MVC-oriented application?

I take it you mean "what" not "that"?  MVC stands for
Model/View/Controller which is a common architectural pattern in GUI
applications.

Ben.

-- 
Ben Hutchings
Lowery's Law:
 If it jams, force it. If it breaks, it needed replacing anyway.


signature.asc
Description: This is a digitally signed message part

Bug#345397: ITP: libtest-mockmodule-perl -- Test::MockModule - Override subroutines in a module for unit testing

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: libtest-mockmodule-perl
  Version : 0.05
  Upstream Author : Simon Flack <[EMAIL PROTECTED]>
* URL : http://search.cpan.org/~simonflk/Test-MockModule/
* License : dual GPL/Artistic
  Description : Test::MockModule - Override subroutines in a module for 
unit testing

Test::MockModule lets you temporarily redefine subroutines in other
packages for the purposes of unit testing.

A Test::MockModule object is set up to mock subroutines for a given
module. The object remembers the original subroutine so it can be
easily restored. This happens automatically when all MockModule
objects for the given module go out of scope, or when you unmock() the
subroutine.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345398: ITP: maypole-plugin-upload -- Maypole::Plugin::Upload - Handle file uploads in Maypole

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: maypole-plugin-upload
  Version : 0.02
  Upstream Author : Marcus Ramberg <[EMAIL PROTECTED]>
* URL : http://search.cpan.org/~mramberg/Maypole-Plugin-Upload/
* License : dual GPL/Artistic
  Description : Maypole::Plugin::Upload - Handle file uploads in Maypole

This plugin adds a upload method to your Maypole request object to allow
you to access file uploads in a platform neutral way.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345399: ITP: maypole-authentication-usersessioncookie -- Maypole::Authentication::UserSessionCookie - Track sessions and, optionally, users

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: maypole-authentication-usersessioncookie
  Version : 1.4
  Upstream Author : Simon Cozens <[EMAIL PROTECTED]>
* URL : 
http://search.cpan.org/~simon/Maypole-Authentication-UserSessionCookie/
* License : to be established
  Description : Maypole::Authentication::UserSessionCookie - Track sessions 
and, optionally, users

This module allows Maypole applications to have the concept of a user,
and to track that user using cookies and sessions.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345403: ITP: libhtml-tagcloud-perl -- HTML::TagCloud - Generate An HTML Tag Cloud

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: libhtml-tagcloud-perl
  Version : 0.32
  Upstream Author : Leon Brocard <[EMAIL PROTECTED]>
* URL : http://search.cpan.org/~lbrocard/HTML-TagCloud/
* License : dual GPL/Artistic
  Description : HTML::TagCloud - Generate An HTML Tag Cloud

The HTML::TagCloud module enables you to generate "tag clouds" in
HTML. Tag clouds serve as a textual way to visualize terms and topics
that are used most frequently. The tags are sorted alphabetically and
a larger font is used to indicate more frequent term usage.

Example sites with tag clouds: http://www.43things.com/,
http://www.astray.com/recipes/ and http://www.flickr.com/photos/tags/.

This module provides a simple interface to generating a CSS-based HTML
tag cloud. You simply pass in a set of tags, their URL and their
count. This module outputs stylesheet-based HTML. You may use the
included CSS or use your own.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#321938: ITP: libclass-dbi-mysql-perl -- Class::DBI extension for MySQL

[Ben Hutchings]

I'm trying to package Memories (see #345277), which depends on this
module.  How is this coming along?

Ben.

-- 
Ben Hutchings
Larkinson's Law: All laws are basically false.


signature.asc
Description: This is a digitally signed message part

Bug#345452: ITP: libclass-dbi-plugin-abstractcount-perl -- Class::DBI::Plugin::AbstractCount - get COUNT(*) results with abstract SQL

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: libclass-dbi-plugin-abstractcount-perl
  Version : 0.06
  Upstream Author : Jean-Christophe Zeus <[EMAIL PROTECTED]> et al.
* URL : 
http://search.cpan.org/~jczeus/Class-DBI-Plugin-AbstractCount/
* License : dual GPL/Artistic
  Description : Class::DBI::Plugin::AbstractCount - get COUNT(*) results 
with abstract SQL

This Class::DBI plugin combines the functionality from
Class::DBI::Plugin::CountSearch (counting objects without having to
use an array or an iterator), and Class::DBI::AbstractSearch, which
allows complex where-clauses a la SQL::Abstract.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345453: ITP: libclass-dbi-plugin-pager-perl -- Class::DBI::Plugin::Pager - paged queries for CDBI

[Ben Hutchings]

Package: wnpp
Severity: wishlist
Owner: Ben Hutchings <[EMAIL PROTECTED]>


* Package name: libclass-dbi-plugin-pager-perl
  Version : 0.561
  Upstream Author : David Baird <[EMAIL PROTECTED]>
* URL : http://search.cpan.org/~davebaird/Class-DBI-Plugin-Pager/
* License : dual GPL/Artistic
  Description : Class::DBI::Plugin::Pager - paged queries for CDBI

Adds a pager method to your class that can query using SQL::Abstract
where clauses, and limit the number of rows returned to a specific
subset.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#321938: ITP: libclass-dbi-mysql-perl -- Class::DBI extension for MySQL

[Ben Hutchings]

I wrote: 
> I'm trying to package Memories (see #345277), which depends on this
> module.  How is this coming along?

I have now created interim packages for Class::DBI::mysql and for
Time::Piece::MySQL, which it uses (though I think the former should
Recommend, not Depend on the latter).  If you would like to use these as
a basis for your own packaging, or if you would be willing to sponsor
uploads of my packages, please let me know.

Ben.

-- 
Ben Hutchings
The generation of random numbers is too important to be left to chance.
- Robert Coveyou


signature.asc
Description: This is a digitally signed message part

Bug#345596: freeloader: fails to start on sarge

[Ben Hutchings]

Package: freeloader
Version: 0.3-3
Severity: grave
Tags: patch
Justification: renders package unusable

Running "freeloader" results in the following error message:

Traceback (most recent call last):
  File "/usr/bin/freeloader", line 42, in ?
import gnomevfs
ImportError: No module named gnomevfs

I guessed that the gnome.vfs module in the stable version of
python2.3-gnome2 might be an earlier version of gnomevfs, so I tried
replacing this line with:

try:
import gnomevfs
except ImportError:
import gnome.vfs as gnomevfs

After this change, it was possible to start the program, open dialogs,
and beging downloading files.  However, after a download is finished,
it shows the error message:

Exception in thread Thread-2:Traceback (most recent call last):
  File "/usr/lib/python2.3/threading.py", line 442, in __bootstrap
self.run()
  File "/usr/share/freeloader/webdl.py", line 111, in run
self.completed()
  File "/usr/share/freeloader/genericdl.py", line 162, in completed
iname = gnome.ui.icon_lookup(theme, fact,
AttributeError: 'module' object has no attribute 'icon_lookup'

and then becomes unresponsive; I had to send it a SIGTERM to kill it.
Looking at the source, I can see why - it fails to release the Gdk
mutex if an exception occurs on this line!  To avoid this sort of
problem, each use of gtk.gdk.threads_{enter,leave} in the program
should be replaced with something like "gdk_lock = GdkLock()" or "del
gdk_lock" respectively, where GdkLock is defined as something like:

class GdkLock:
def __init__(self):
gtk.gdk.threads_enter()
def __del__(self):
gtk.gdk.threads_leave()

With this second change, the program is barely functional, but
continues to print error messages that suggest to me it requires
a more recent version of python-gnome2 than that in sarge.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.4.27-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages freeloader depends on:
ii  bittorrent   3.4.2-3sarge0.1 Scatter-gather network file transf
ii  librsvg2-2   2.8.1-3 SAX-based renderer library for SVG
ii  python   2.3.5-2 An interactive high-level object-o
ii  python-glade22.6.1-2 GTK+ bindings: Glade support
ii  python-gnome22.6.1-1 Python bindings for the GNOME desk
ii  python-gtk2  2.6.1-2 Python bindings for the GTK+ widge

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345597: freeloader: manual page is unfinished

[Ben Hutchings]

Package: freeloader
Version: 0.3-3
Severity: wishlist

The manual page for freeloader says it has a command-line interface,
but does not document it.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.4.27-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages freeloader depends on:
ii  bittorrent   3.4.2-3sarge0.1 Scatter-gather network file transf
ii  librsvg2-2   2.8.1-3 SAX-based renderer library for SVG
ii  python   2.3.5-2 An interactive high-level object-o
ii  python-glade22.6.1-2 GTK+ bindings: Glade support
ii  python-gnome22.6.1-1 Python bindings for the GNOME desk
ii  python-gtk2  2.6.1-2 Python bindings for the GTK+ widge

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#345231: freeloader: unhandled exception when exiting with python2.3

[Ben Hutchings]

socket.SHUT_RDWR is also used in the main module of freeloader, but this
is masked by an "except:" block (this is a great way to create confusing
bugs).

Ben.

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.


signature.asc
Description: This is a digitally signed message part

Bug#345596: freeloader: fails to start on sarge

[Ben Hutchings]

You wrote:
> Le lundi 02 janvier 2006 à 04:10 +0000, Ben Hutchings a écrit :
> > Package: freeloader
> > Version: 0.3-3
> > Severity: grave
> > Tags: patch
> > Justification: renders package unusable
> I don't think the fact that a package from the unstable distribution
> cannot be used on Sarge is not a justification for a grave status.

Yes, you're right - I was confused by the fact that I was able to
install it on sarge without any need to tell apt explicitly to download
unstable packages.

> [...]
> 
> > With this second change, the program is barely functional, but
> > continues to print error messages that suggest to me it requires
> > a more recent version of python-gnome2 than that in sarge.
> I agree, thus I changed dependencies so that python-gnome2,
> python-glade2 and python-gtk2 are at least 2.10 (current testing and
> unstable versions).
> 
> The problem will be fixed in the next upload.

I've actually succeeded in making it build and run in sarge with some
quite small changes, so it might be worth leaving the dependencies as
they are.  I'm attaching a patch that addresses all the issues I
originally mentioned.  This also puts the changes you made to configure
(in patch 01_disable_modules_check.dpatch) into configure.ac along with
the change needed to accept gnome.vfs in place of gnomevfs, as I believe
it is considered bad practice to patch generated files such as configure
scripts.

Most of these changes could probably go upstream.

Ben.

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.


signature.asc
Description: This is a digitally signed message part

Bug#345596: freeloader: fails to start on sarge

[Ben Hutchings]

Julien Valroff wrote:
> Le lundi 02 janvier 2006 à 18:19 +0000, Ben Hutchings a écrit :
> > 
> > I've actually succeeded in making it build and run in sarge with some
> > quite small changes, so it might be worth leaving the dependencies as
> > they are.  I'm attaching a patch that addresses all the issues I
> > originally mentioned.  This also puts the changes you made to
> > configure
> > (in patch 01_disable_modules_check.dpatch) into configure.ac along
> > with
> > the change needed to accept gnome.vfs in place of gnomevfs, as I
> > believe
> > it is considered bad practice to patch generated files such as
> > configure
> > scripts.
> 
> May you please add the attachment?

That would be a good idea!

Ben.

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.
diff -urN freeloader-0.3/configure.ac freeloader-0.3-mine/configure.ac
--- freeloader-0.3/configure.ac
+++ freeloader-0.3-mine/configure.ac
@@ -9,9 +9,15 @@
 
 AX_PYTHON
 AC_PYTHON_MODULE(pygtk, 1)
-AC_PYTHON_MODULE(gtk, 1)
-AC_PYTHON_MODULE(gtk.glade, 1)
-AC_PYTHON_MODULE(gtk.gdk, 1)
+# HACK
+# importing these module fails if no DISPLAY is set
+# which means it fails with pbuilder build and since we do have
+# this module present (due to dependencies) I am going to ignore this
+# error, see: http://oss.codepoet.no/bugzilla/show_bug.cgi?id=155
+#AC_PYTHON_MODULE(gtk, 1)
+#AC_PYTHON_MODULE(gtk.glade, 1)
+#AC_PYTHON_MODULE(gtk.gdk, 1)
+# HACK
 AC_PYTHON_MODULE(gnome, 1)
 AC_PYTHON_MODULE(gnome.ui, 1)
 AC_PYTHON_MODULE(gobject, 1)
@@ -22,7 +28,10 @@
 AC_PYTHON_MODULE(BitTorrent, 1)
 AC_PYTHON_MODULE(BitTorrent.bencode, 1)
 AC_PYTHON_MODULE(BitTorrent.download, 1)
-AC_PYTHON_MODULE(gnomevfs, 1)
+AC_PYTHON_MODULE(gnomevfs)
+if test $HAVE_PYMOD_GNOMEVFS = no; then
+AC_PYTHON_MODULE(gnome.vfs, 1)
+fi
 AC_PYTHON_MODULE(gconf, 1)
 AC_PYTHON_MODULE(tempfile, 1)
 AC_PYTHON_MODULE(urllib, 1)
diff -urN freeloader-0.3/src/Makefile.am freeloader-0.3-mine/src/Makefile.am
--- freeloader-0.3/src/Makefile.am
+++ freeloader-0.3-mine/src/Makefile.am
@@ -3,7 +3,7 @@
 bin_SCRIPTS = freeloader
 
 localdistdir = $(datadir)/@PACKAGE_NAME@
-localdist_SCRIPTS = genericdl.py webdl.py btdl.py sockdae.py
+localdist_SCRIPTS = genericdl.py webdl.py btdl.py sockdae.py gdklock.py
 
 EXTRA_DIST = $(bin_SCRIPTS) $(localdist_SCRIPTS)
 
diff -urN freeloader-0.3/src/btdl.py freeloader-0.3-mine/src/btdl.py
--- freeloader-0.3/src/btdl.py	2005-04-07 03:28:06.0 +0100
+++ freeloader-0.3-mine/src/btdl.py	2006-01-02 04:13:36.0 +
@@ -22,6 +22,7 @@
 import BitTorrent
 import BitTorrent.download
 import genericdl
+from gdklock import GdkLock
 import urllib
 import threading
 import os
@@ -96,17 +97,15 @@
 		self.update_fields()
 
 	def error(self, errormsg):
-		gtk.gdk.threads_enter()
+		gdk_lock = GdkLock()
 		self.print_log(errormsg, gtk.STOCK_DIALOG_ERROR)
-		gtk.gdk.threads_leave()
 
 	def finished(self):
 		self.downRate = 0
 		self.completed()
-		gtk.gdk.threads_enter()
+		gdk_lock = GdkLock()
 		self.print_log("Completed Torrent: " + self.get_tail(),
 "arrows-up")
-		gtk.gdk.threads_leave()
 
 	def run(self):
 		args = ()
@@ -125,10 +124,10 @@
 
 		args += ("--saveas", self.des_file, self.src_file)
 
-		gtk.gdk.threads_enter()
+		gdk_lock = GdkLock()
 		self.print_log("Starting Torrent: " + self.get_tail(),
 "arrows-down-up")
-		gtk.gdk.threads_leave()
+		del gdk_lock
 
 		try:
 			BitTorrent.download.download(	args,
@@ -140,16 +139,16 @@
 			20)
 		except:
 			#there are non fatal exceptions?
-			gtk.gdk.threads_enter()
+			gdk_lock = GdkLock()
 			self.print_log("Error in Torrent thread",
 	gtk.STOCK_DIALOG_ERROR)
-			gtk.gdk.threads_leave()
+			del gdk_lock
 
 		if self.killflag.isSet():
 			self.downRate = 0
 			self.completed()
-			gtk.gdk.threads_enter()
+			gdk_lock = GdkLock()
 			self.print_log("Stopped Torrent: " + self.get_tail(),
 	gtk.STOCK_DIALOG_INFO)
-			gtk.gdk.threads_leave()
+			del gdk_lock
 
diff -urN freeloader-0.3/src/freeloader.in freeloader-0.3-mine/src/freeloader.in
--- freeloader-0.3/src/freeloader.in	2005-04-07 03:31:52.0 +0100
+++ freeloader-0.3-mine/src/freeloader.in	2006-01-02 04:33:22.0 +
@@ -39,7 +39,10 @@
 import gconf
 import socket
 import random
-import gnomevfs
+try:
+	import gnomevfs
+except ImportError:
+	import gnome.vfs as gnomevfs
 import BitTorrent.bencode
 import string
 try:
@@ -78,6 +81,7 @@
 import btdl
 import genericdl
 import sockdae
+from gdklock import GdkLock
 #import webbtdl
 #import fldservice
 
@@ -673,16 +681,14 @@
 	def sync_ui_timer(self):
 		if self.killflag.isSet():
 			return False
-		gtk.gdk.threads_enter()
+		gdk_lock = GdkLock()
 		if gtk.main_level() == 0:
-			gtk.gdk.threads_leave()
 			return False
 		try:
 			self.sync_ui()
 		except:
-			gtk.gdk.threads_leav

Bug#339407: sgt-puzzles: ftbfs [sparc] make[1]: *** [blackboxgame.6] Broken pipe

[Ben Hutchings]

You wrote:
> sgt-puzzles failed to build on a sparc buildd, but did build on my
> sparc pbuilder.
> 
> 
> make -f Makefile.doc
> make[1]: Entering directory `/build/buildd/sgt-puzzles-6452'
> halibut --winhelp=puzzles.hlp --text=puzzles.txt puzzles.but
> halibut --text=HACKING devel.but
> perl mkmanpages.pl puzzles.but
> Generating fifteen.6
> make[1]: *** [blackboxgame.6] Broken pipe
> make[1]: Leaving directory `/build/buildd/sgt-puzzles-6452'

I noticed that this happened on a previous auto-build attempt on SPARC,
and my sponsor has seen it happen once when building on i386.  I am
investigating this but I'm unconvinced that it is a bug in sgt-puzzles
itself.  If you look at the text of mkmanpages.pl you'll see the
pipeline it's running.

Ben.

-- 
Ben Hutchings
If the facts do not conform to your theory, they must be disposed of.


signature.asc
Description: This is a digitally signed message part

Bug#305434: Rhythm method is broken

[Ben Hutchings]

Package: cycle
Version: 0.3.0-1
Severity: important


The length of the menstrual cycle is extremely variable even for an
individual woman.  The rhythm method is not simply "not reliable" as a
method of contraception, but really counterproductive.  It is
irresponsible to claim to predict fertile or "safe" days based on this
method.

The fertility awareness method, based on measurement of bodily changes
during the cycle, can much more accurately predict the timing of
ovulation and menstruation.  That involves day-to-day measurements and
might be well-served by a program like this.

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.7-1-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#309175: ITP: sgt-puzzles -- Simon Tatham's Portable Puzzle Collection - one-player puzzle games

[Ben Hutchings]

Package: wnpp
Severity: wishlist

* Package name: sgt-puzzles
  Version : 5780
  Upstream Author : Simon Tatham <[EMAIL PROTECTED]>
* URL : http://www.chiark.greenend.org.uk/~sgtatham/puzzles/
* License : MIT/X
  Description : Simon Tatham's Portable Puzzle Collection - one-player 
puzzle games

The Portable Puzzle Collection contains a number of popular puzzle games
implemented in mostly portable code.  They will run on Unix/Linux using
Gtk.  The current collection is Net (based on FreeNet), Cube, Fifteen,
Sixteen, Twiddle (another tile variant), Rectangles (based on Divide by
Squares), Netslide (sliding variant of Net), Pattern (nonograms) and
Solo (based on Sudoku).



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#309612: bash: name completion mangles special characters in case of ambiguity if completion-ignore-case is on

[Ben Hutchings]

Package: bash
Version: 2.05b-26
Severity: normal


If the readline option completion-ignore-case is on and a name
completion attempt fails because there is ambiguity about the next
character, then any special characters in the name are escaped even
though they must already be escaped.  The escaped name is truncated to
the number of characters it represents before undergoing re-escaping.
This does not happen if any further characters can be guessed or if
completion-ignore-case is off.

This bug appears to have been fixed in version 3.

Example:

$ mkdir foo\ bar foo\ baz
$ echo 'set completion-ignore-case on' >inputrc
$ echo 'set show-all-if-ambiguous on' >>inputrc
$ INPUTRC=inputrc bash
$ cd foo  # press tab
foo bar  foo baz
$ cd foo\ ba  # press tab again
foo bar  foo baz
$ cd foo\\\ b

(show-all-if-ambiguous does not need to be set, but it means bash
won't modify the command line in place and hopefully makes the
example clearer.)

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages bash depends on:
ii  base-files  3.1.2Debian base system miscellaneous f
ii  libc6   2.3.2.ds1-21 GNU C Library: Shared libraries an
ii  libncurses5 5.4-4Shared libraries for terminal hand
ii  passwd  1:4.0.3-30.7 Change and administer password and

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#310073: clearsilver: FTBFS: fails to apply patch.

[Ben Hutchings]

tags 310073 patch
thanks

debian/rules has a rather unwise definition of DEB_TAR_SRCDIR that means
the build directory isn't determined consistently.  This seems to fix
it:

--- debian/rules~
+++ debian/rules
@@ -1,6 +1,6 @@
 #!/usr/bin/make -f

-DEB_TAR_SRCDIR=$(shell ls clearsilver-* | sed -e 's/\.tar.*//')
+DEB_TAR_SRCDIR:=$(shell ls clearsilver-* | sed -e 's/\.tar.*//;q')
 DEB_MAKE_CHECK_TARGET=( i386 ia64 )

 include /usr/share/cdbs/1/rules/tarball.mk

-- 
Ben Hutchings
If more than one person is responsible for a bug, no one is at fault.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#293764: gpsman FTBFS due to broken diff

[Ben Hutchings]

Package: gpsman
Version: 5.4.2-2
Severity: important
Justification: fails to build from source


dpkg-source says:

dpkg-source: extracting gpsman in gpsman-5.4.2
1 out of 6 hunks FAILED -- saving rejects to file gmsrc/rgb.tcl.rej
dpkg-source: failure: patch gave error exit status 1


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#318869: sgt-puzzles: FTBFS: 'r2.x' may be used uninitialized in this function

[Ben Hutchings]

This warning is actually a false report from gcc, which unfortunately is
treated as an error due to compiler options chosen upstream.  r1 and r2
will always be initialised by one of the cases in the switch statement
before they are used.  gcc doesn't know that one of these cases will
always match, but it can be convinced by the addition of a default case
that asserts a falsehood.  I have made this change in a new version of
the package but am waiting for a sponsor to upload it.

Note that this affects all architectures when the new version of gcc is
used.

-- 
Ben Hutchings
When you say `I wrote a program that crashed Windows', people just stare ...
and say `Hey, I got those with the system, *for free*'. - Linus Torvalds


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#320195: dvdauthor: Assertion failure in dvdvob.c

[Ben Hutchings]

Package: dvdauthor
Version: 0.6.11-1
Severity: important

I have used dvdauthor successfully a number of times, but on my
latest attempt it dies with the message:

dvdauthor: dvdvob.c:1316: FindVobus: Assertion `p>vi->sectpts[0]' failed.

This is reliably reproducible with the current input files.  Please let
me know what information you need to investigate this.

It may be relevant that I have changed from using only MPEG-2 streams
to a mixture of MPEG-1 and MPEG-2 (but consistent within each titleset).

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages dvdauthor depends on:
ii  libbz2-1.0   1.0.2-7 high-quality block-sorting file co
ii  libc62.3.2.ds1-22GNU C Library: Shared libraries an
ii  libdps1  4.3.0.dfsg.1-14 Display PostScript (DPS) client li
ii  libdvdread3  0.9.4-5 Simple foundation for reading DVDs
ii  libfreetype6 2.1.7-2.4   FreeType 2 font engine, shared lib
ii  libfribidi0  0.10.4-6Free Implementation of the Unicode
ii  libice6  4.3.0.dfsg.1-14 Inter-Client Exchange library
ii  libjasper-1.701-11.701.0-2   The JasPer JPEG-2000 runtime libra
ii  libjpeg626b-10   The Independent JPEG Group's JPEG 
ii  liblcms1 1.13-1  Color management library
ii  libmagick6   6:6.0.6.2-2.4   Image manipulation library
ii  libpng12-0   1.2.8rel-1  PNG library - runtime
ii  libsm6   4.3.0.dfsg.1-14 X Window System Session Management
ii  libtiff4 3.7.2-3 Tag Image File Format (TIFF) libra
ii  libx11-6 4.3.0.dfsg.1-14 X Window System protocol client li
ii  libxext6 4.3.0.dfsg.1-14 X Window System miscellaneous exte
ii  libxml2  2.6.16-7GNOME XML library
ii  libxt6   4.3.0.dfsg.1-14 X Toolkit Intrinsics
ii  xlibs4.3.0.dfsg.1-14 X Keyboard Extension (XKB) configu
ii  zlib1g   1:1.2.2-4   compression library - runtime

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#320333: halibut: Generates erroneous .UC macro in manual pages

[Ben Hutchings]

Package: halibut
Version: 0.9+svn20050412-1
Severity: minor
Tags: patch

When halibut is used to generate a manual page, the resulting file
includes the macro invokation ".UC" which causes the manual page to be
given a left-hand footer of "Third Berkeley Distribution" when
formatted by groff.  This is obviously erroneous.

I have reported this to the upstream author, Simon Tatham, in person
and expect that he will fix it soon.  In any case this trivial should
fix it:

--- bk_man.c~
+++ bk_man.c
@@ -164,8 +164,6 @@
 }
 fputc('\n', fp);

-fprintf(fp, ".UC\n");
-
 had_described_thing = FALSE;
 #define cleanup_described_thing do { \
 if (had_described_thing) \

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages halibut depends on:
ii  libc6   2.3.2.ds1-22 GNU C Library: Shared libraries an

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#320195: dvdauthor: Assertion failure in dvdvob.c

[Ben Hutchings]

It's probably relevant that immediately before this I see the message:
WARN: Video PTS does not line up on a multiple of a field.

It may be relevant that for some titlesets the menus are MPEG-2 VOBs
with resolution 720x756 while the "titles" are MPEG-1 VOBs with
resolution 352x288.  The above warning and assertion failure appear
to be produced while dvdauthor is processing these titlesets, after
it finishes scanning the "titles".

-- 
Ben Hutchings
Hoare's Law of Large Problems:
Inside every large problem is a small problem struggling to get out.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#334640: sgt-puzzles: Newer version depends gnome libraries

[Ben Hutchings]

Djalma Brandão Martins wrote:
> Dependencies of version 6229-1:
> 
> Depends: libatk1.0-0 (>= 1.9.0), libc6 (>= 2.3.5-1), libglib2.0-0 (>=
> 2.8.0), libgtk2.0-0 (>= 2.6.0), libpango1.0-0 (>= 1.8.2),
> libx11-6 | xlibs (>> 4.1.0)
> 
> Dependencies of version 6378-1:
> 
> Depends: libatk1.0-0 (>= 1.9.0), libbonobo2-0 (>= 2.8.0), libc6 (>=
> 2.3.5-1), libgconf2-4 (>= 2.9), libglib2.0-0 (>= 2.8.0), libgnome2-0 (>=
> 2.8.0), libgnomevfs2-0 (>= 2.10.0-0), libgtk2.0-0 (>= 2.6.0), liborbit2
> (>= 1:2.10.0), libpango1.0-0 (>= 1.8.2), libpopt0 (>= 1.7)
> 
> Is it really necessary? I use fluxbox.

The new version uses GNOME Help to provide online help similarly to the
Window version.

Ben.

-- 
Ben Hutchings
Reality is just a crutch for people who can't handle science fiction.


signature.asc
Description: This is a digitally signed message part

Bug#335267: sgt-puzzles: new release available

[Ben Hutchings]

Package: sgt-puzzles
Version: 6378-1
Severity: wishlist

The upstream version (currently 6423) has two new puzzles: Bridges and Tents.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.4.27-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages sgt-puzzles depends on:
ii  libatk1.0-0   1.10.1-2   The ATK accessibility toolkit
ii  libbonobo2-0  2.8.1-2Bonobo CORBA interfaces library
ii  libc6 2.3.5-7GNU C Library: Shared libraries an
ii  libgconf2-4   2.8.1-6GNOME configuration database syste
ii  libglib2.0-0  2.6.4-1The GLib library of C routines
ii  libgnome2-0   2.8.1-2The GNOME 2 library - runtime file
ii  libgnomevfs2-02.8.4-4The GNOME virtual file-system libr
ii  libgtk2.0-0   2.6.4-3The GTK+ graphical user interface 
ii  liborbit2 1:2.12.2-1 libraries for ORBit2 - a CORBA ORB
ii  libpango1.0-0 1.8.2-1Layout and rendering of internatio
ii  libpopt0  1.7-5  lib for parsing cmdline parameters

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#333273: xvfb: Uses incorrect line padding when depth>8

[Ben Hutchings]

Package: xvfb
Version: 4.3.0.dfsg.1-14sarge1
Severity: important

According to the Xvfb manual page, the -fbdir option can be used to
make it expose its framebuffer in X window dump format.  However, if
it is run with a depth of 32, the bytes_per_line member of the
XWDHeader structure is set to 4 * width but each line is actually
padded to 16 * width.  Similarly, if it is run with a depth of 16,
bytes_per_line is set to 2 * width but lines are padded to 4 * width.
This makes the feature quite useless.  It is, however, possible to
generate correct window dumps using xwd -root.

-- System Information:
Debian Release: 3.1
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.4.27-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages xvfb depends on:
ii  libc6  2.3.2.ds1-22  GNU C Library: Shared libraries an
ii  libfreetype6   2.1.7-2.4 FreeType 2 font engine, shared lib
ii  xfree86-common 4.3.0.dfsg.1-14   X Window System (XFree86) infrastr
ii  zlib1g 1:1.2.2-4.sarge.2 compression library - runtime

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#323072: normalize-audio: wrong references in manual pages

[Ben Hutchings]

Package: normalize-audio
Version: 0.7.6-4
Severity: normal

The manual page for normalize-audio refers to it as normalize.
The manual page for normalize-{mp3,ogg} refers to a normalize(1) manual page.

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages normalize-audio depends on:
ii  libaudiofile0   0.2.6-6  Open-source version of SGI's audio
ii  libc6   2.3.2.ds1-22 GNU C Library: Shared libraries an
ii  libmad0 0.15.1b-1.1  MPEG audio decoder library

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#323115: libxine1: Video decoding broken after switching from MPEG-2 to MPEG-1 on DVD

[Ben Hutchings]

Package: libxine1
Version: 1.0.1-1
Severity: normal

I have authored a DVD which uses a mixture of MPEG-1 (352x288 pixels)
and MPEG-2 (704x576 pixels, interlaced) video.  When I play the DVD
with Xine and select a title that uses MPEG-1 from the menus, it is
displayed very strangely: the picture is recognisable, but cells
appear to be decoded with the wrong DCT coefficients.  If I specify
one of these titles on the xine command line, it displays as
expected.  No such problem occurs when using a hardware DVD player.

I think the MPEG-2 decoder is failing to re-check the video stream
type when it sees the program stream header for the MPEG-1 video, but
I have not yet investigated far enough to work out exactly what the
cause is and how it can be fixed.

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages libxine1 depends on:
ii  libasound2 1.0.8-3   ALSA library
ii  libc6  2.3.2.ds1-22  GNU C Library: Shared libraries an
ii  libfreetype6   2.1.7-2.4 FreeType 2 font engine, shared lib
ii  libmodplug01:0.7-4   shared libraries for mod music bas
ii  libogg01.1.2-1   Ogg Bitstream Library
ii  libpng12-0 1.2.8rel-1PNG library - runtime
ii  libspeex1  1.1.6-2   The Speex Speech Codec
ii  libstdc++5 1:3.3.5-13The GNU Standard C++ Library v3
ii  libtheora0 0.0.0.alpha4-1.1  The Theora Video Compression Codec
ii  libvorbis0a1.1.0-1   The Vorbis General Audio Compressi
ii  libxext6   4.3.0.dfsg.1-14   X Window System miscellaneous exte
ii  xlibmesa-gl [libgl1]   4.3.0.dfsg.1-14   Mesa 3D graphics library [XFree86]
ii  xlibmesa-glu [libglu1] 4.3.0.dfsg.1-14   Mesa OpenGL utility library [XFree
ii  xlibs  4.3.0.dfsg.1-14   X Keyboard Extension (XKB) configu
ii  zlib1g 1:1.2.2-4.sarge.2 compression library - runtime

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#323410: dvdauthor: Sometimes generates files that are unacceptable to mkisofs

[Ben Hutchings]

Package: dvdauthor
Version: 0.6.11-1
Severity: important
Tags: patch

After running dvdauthor to create a DVD filesystem tree, I run mkisofs
with the -dvd-video to create a filesystem image from this.  In some
cases this results in the following error messages:

mkisofs: Either VIDEO_TS.IFO or VIDEO_TS.VOB is not of correct size.
mkisofs: Unable to parse DVD-Video structures.
mkisofs: Unable to make a DVD-Video image.

I examined the mkisofs source and found that this message could
actually be triggered by files other than those named, so I edited it
to report the correct names and also the values of the variables used
in the comparison.  I found that VTS_01_0.IFO was the culprit in this
case.

The variables mkisofs tests before emitting this error message, and
their values in this case, are:

vts_ifo->vtsi_mat->vts_last_sector = 3695431
 DVDFileSize(vts_ifo_file) = 316 (sectors in VTS_01_0.IFO)
   DVDFileSize(vts_title_file) = 3694824 (sectors in VTS_01_[1-8].VOB)
DVDFileSize(vts_menu_file) = 232 (sectors in VTS_01_0.vOB)

mkisofs expects that vts_ifo->vtsi_mat->vts_last_sector + 1 >= 2 *
DVDFileSize(vts_ifo_file) + DVDFileSize(vts_title_file) +
DVDFileSize(vts_menu_file), which seems reasonable to me.  In this
case the left hand side is smaller by 256.

To cut a long story short, I found that the WriteIFO function in
dvdifo.c only writes a single byte for many of the sector numbers 
used in the IFO file, so it only works so long as the IFO file takes
up less than 256 sectors.  This is generally true, but this titleset
contains an unusually large amount of video (over 30 hours).

I have changed dvdifo.c to write sector numbers as 4 bytes, on the
assumption that this is the correct length of the fields, which
satisfies mkisofs in at least this case.  However I do not have
documentation of the IFO file format and I have not yet had a chance
to check the resulting image in a DVD player.  A patch follows.

--- dvdauthor-0.6.11/src/dvdifo.c.orig  2005-08-16 14:18:20.893116876 +0100
+++ dvdauthor-0.6.11/src/dvdifo.c   2005-08-16 14:22:03.065885528 +0100
@@ -371,43 +371,43 @@
 write4(buf+128,0x7ff);
 i=1;
 
-buf[0xCB]=i; // VTS_PTT_SRPT
+write4(buf+0xC8,i); // VTS_PTT_SRPT
 i+=Create_PTT_SRPT(0,ws->titles);
 
-buf[0xCF]=i; // VTS_PGCI
+write4(buf+0xCC,i); // VTS_PGCI
 i+=CreatePGC(0,ws,0);
 
 if( jumppad || forcemenus ) {
-buf[0xD3]=i; // VTSM_PGCI
+write4(buf+0xD0,i); // VTSM_PGCI
 i+=CreatePGC(0,ws,1);
 }
 
-buf[0xD7]=i; // VTS_TMAPT
+write4(buf+0xD4,i); // VTS_TMAPT
 i+=numsectTMAPT(ws->titles);
 
 if( jumppad || forcemenus ) {
-buf[0xDB]=i; // VTSM_C_ADT
+write4(buf+0xD8,i); // VTSM_C_ADT
 i+=CreateCallAdr(0,ws->menus->vg);
 
-buf[0xDF]=i; // VTSM_VOBU_ADMAP
+write4(buf+0xDC,i); // VTSM_VOBU_ADMAP
 i+=numsectVOBUAD(ws->menus->vg);
 }
 
-buf[0xE3]=i; // VTS_C_ADT
+write4(buf+0xE0,i); // VTS_C_ADT
 i+=CreateCallAdr(0,ws->titles->vg);
 
-buf[0xE7]=i; // VTS_VOBU_ADMAP
+write4(buf+0xE4,i); // VTS_VOBU_ADMAP
 i+=numsectVOBUAD(ws->titles->vg);
 
-buf[31]=i-1;
+write4(buf+28,i-1);
 if( jumppad || forcemenus ) {
-buf[0xC3]=i;
+write4(buf+0xC0,i);
 i+=getvoblen(ws->menus->vg);
 }
 write4(buf+0xC4,i);
 if( ws->titles->numpgcs )
 i+=getvoblen(ws->titles->vg);
-i+=buf[31];
+i+=read4(buf+28);
 write4(buf+12,i);
 
 if( jumppad || forcemenus )

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages dvdauthor depends on:
ii  libbz2-1.0 1.0.2-7   high-quality block-sorting file co
ii  libc6  2.3.2.ds1-22  GNU C Library: Shared libraries an
ii  libdps14.3.0.dfsg.1-14   Display PostScript (DPS) client li
ii  libdvdread30.9.4-5   Simple foundation for reading DVDs
ii  libfreetype6   2.1.7-2.4 FreeType 2 font engine, shared lib
ii  libfribidi00.10.4-6  Free Implementation of the Unicode
ii  libice64.3.0.dfsg.1-14   Inter-Client Exchange library
ii  libjasper-1.701-1  1.701.0-2 The JasPer JPEG-2000 runtime libra
ii  libjpeg62  6b-10 The Independent JPEG Group's JPEG 
ii  liblcms1   1.13-1Color management library
ii  libmagick6 6:6.0.6.2-2.4 Image manipulation library
ii  libpng12-0 1.2.8rel-1PNG library - runtime
ii  libsm6 4.3.0.dfsg.1-14   X Window System Session Management
ii  libtiff4   3.7.2-3   Tag Image File Format (TIFF) libra
ii  libx11-6   4.3.0.dfsg.1-14   X Window System protocol client li
ii  li

Bug#320195: dvdauthor: Assertion failure in dvdvob.c

[Ben Hutchings]

Marc Leeman wrote:
> > It's probably relevant that immediately before this I see the message:
> > WARN: Video PTS does not line up on a multiple of a field.
> 
> I forwarded it to the developers mailing lists. If a fix is returned,
> I'll provide a patch and most likely, do a service release of the
> package (another fix for a bug is pending in my tree).

I have investigated a little further and found:
- The assertion is actually triggered while processing specific VOBs,
not at the end of the titleset.
- Removing these specific VOBs avoids it.
- Attempting to generate a DVD with one of these VOBs on its own (well,
plus menus) triggers the assertion.

I don't know what's special about these problematic VOBs.  They were
produced using mencoder with identical or at least similar options to
other VOBs that don't cause problems.

The warning "WARN: Video PTS does not line up on a multiple of a field."
doesn't appear for all of these titles, so is probably not relevant
after all.

-- 
Ben Hutchings
Hoare's Law of Large Problems:
Inside every large problem is a small problem struggling to get out.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#321183: dvdauthor: spumux crashes if button coordinates are out of range

[Ben Hutchings]

Package: dvdauthor
Version: 0.6.11-1
Severity: normal
Tags: patch

spumux doesn't validate button coordinates read from an XML file, and
will happily read beyond the bounds of the image bitmaps if they are
out of range.  The following patch adds appropriate validation.

--- dvdauthor-0.6.11/src/subgen-image.c 2005-02-01 02:26:22.0 +
+++ dvdauthor-0.6.11.new/src/subgen-image.c 2005-08-03 23:48:23.718731801 
+0100
@@ -361,12 +361,25 @@
 gs=malloc(ng*sizeof(palgroup));
 memset(gs,0,ng*sizeof(palgroup));
 
+assert(!useimg || s->xd <= s->img.width);
+assert(s->xd <= s->hlt.width);
+assert(s->xd <= s->sel.width);
+
 // fprintf(stderr,"attempt %d groups, %d useimg\n",ng,useimg);
 // find unique colors per button
 for( i=0; inumbuttons; i++ ) {
 button *b=&s->buttons[i];
 palgroup *bp=&bpgs[i];
 
+if( b->r.x0<0 || b->r.x0>b->r.x1 || b->r.x1>=s->xd
+|| b->r.y0<0 || b->r.y0>b->r.y1 || b->r.y1>=s->yd )
+{
+if (debug > -1)
+fprintf(stderr,
+"ERR: Button coordinates out of range: 
(%d,%d)-(%d,%d)\n",
+b->r.x0, b->r.y0, b->r.x1, b->r.y1);
+exit(1);
+}
 for( y=b->r.y0; yr.y1; y++ )
 for( x=b->r.x0; xr.x1; x++ )
 if( !checkcolor(bp,gettricolor(s,y*s->xd+x,useimg)) )

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages dvdauthor depends on:
ii  libbz2-1.0   1.0.2-7 high-quality block-sorting file co
ii  libc62.3.2.ds1-22GNU C Library: Shared libraries an
ii  libdps1  4.3.0.dfsg.1-14 Display PostScript (DPS) client li
ii  libdvdread3  0.9.4-5 Simple foundation for reading DVDs
ii  libfreetype6 2.1.7-2.4   FreeType 2 font engine, shared lib
ii  libfribidi0  0.10.4-6Free Implementation of the Unicode
ii  libice6  4.3.0.dfsg.1-14 Inter-Client Exchange library
ii  libjasper-1.701-11.701.0-2   The JasPer JPEG-2000 runtime libra
ii  libjpeg626b-10   The Independent JPEG Group's JPEG 
ii  liblcms1 1.13-1  Color management library
ii  libmagick6   6:6.0.6.2-2.4   Image manipulation library
ii  libpng12-0   1.2.8rel-1  PNG library - runtime
ii  libsm6   4.3.0.dfsg.1-14 X Window System Session Management
ii  libtiff4 3.7.2-3 Tag Image File Format (TIFF) libra
ii  libx11-6 4.3.0.dfsg.1-14 X Window System protocol client li
ii  libxext6 4.3.0.dfsg.1-14 X Window System miscellaneous exte
ii  libxml2  2.6.16-7GNOME XML library
ii  libxt6   4.3.0.dfsg.1-14 X Toolkit Intrinsics
ii  xlibs4.3.0.dfsg.1-14 X Keyboard Extension (XKB) configu
ii  zlib1g   1:1.2.2-4   compression library - runtime

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#321183: dvdauthor: spumux crashes if button coordinates are out of range

[Ben Hutchings]

The assertions added by the patch really ought to sanity-check s->yd
against image dimensions as well.

Ben.

-- 
Ben Hutchings
Life is what happens to you while you're busy making other plans.
   - John Lennon


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#305490: mozilla-firefox: Opens (frame-)links in new windows instead of mainframe, all instances crash when closing these w.

[Ben Hutchings]

Could this be the same as or related to upstream bug 244482?

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=305490
https://bugzilla.mozilla.org/show_bug.cgi?id=244482

Ben.

-- 
Ben Hutchings
Experience is what causes a person to make new mistakes instead of old ones.


signature.asc
Description: This is a digitally signed message part

Bug#404758: libavcodec.pc version gratuitously broken due to "0d." prefix

[Ben Hutchings]

On Thu, 2006-12-28 at 00:33 +, Darren Salt wrote:
> Package: libavcodec-dev
> Version: 0.cvs20060823-4
> Severity: important
> 
> The version number in /usr/lib/pkgconfig/libavcodec.pc is gratuitously
> broken,


The version number there is the same version that goes in the library's
SONAME, which is arguably an upstream bug.  (SONAME versions represent
ABIs but pkg-config is all about source compatibility.)

I believe the prefix was added to the SONAME because our libavcodec may
not be binary-compatible with another distributions's version, depending
on configuration options.  In addition it may be that upstream bumps the
major number too often.  (However, I'm fairly sure the major number
*does* need to be changed from time to time, so I'm not sure the prefix
should still be "0d.".)

Ben.

-- 
Ben Hutchings
For every complex problem
there is a solution that is simple, neat, and wrong.


signature.asc
Description: This is a digitally signed message part

Bug#404733: dpkg error upgrading iceweasel

[Ben Hutchings]

Well this is weird.  The iceweasel package contains a symlink
/usr/lib/firefox -> iceweasel which I would expect to conflict with the
many plugins that install under the directory /usr/lib/firefox, as seen
here with totem-mozilla.  However, on my etch system the upgrade
succeeded and /usr/lib/firefox remains a directory.

Ben.

-- 
Ben Hutchings
For every complex problem
there is a solution that is simple, neat, and wrong.


signature.asc
Description: This is a digitally signed message part

Bug#404649: iceweasel: firstrun-page not found, still references to mozilla.org

[Ben Hutchings]

This behaviour appears to be controlled by the file firefox-branding.js

Ben.

-- 
Ben Hutchings
For every complex problem
there is a solution that is simple, neat, and wrong.


signature.asc
Description: This is a digitally signed message part

Bug#400302: Customizing toolbars, removing and readding location bar makes menu stop working

[Ben Hutchings]

This is reliably reproducible for me in version 2.0.0.1+dfsg-1, starting
with a fresh profile and no other packages that install files
in /usr/lib/iceweasel.

Opening the customisation window or dragging something and dropping it
where it started isn't enough to trigger this.  You have to make some
kind of real change, e.g. add a separator to the navigation bar.  After
this the menu headings are greyed out; mousing over makes them black but
they can't be pulled down.

Ben.

-- 
Ben Hutchings
For every complex problem
there is a solution that is simple, neat, and wrong.


signature.asc
Description: This is a digitally signed message part

Bug#401167: firefox not expunged from iceweasel 2.0 in help->about iceweasel

[Ben Hutchings]

forcemerge 399791 401167
thanks

This was fixed in version 2.0.0.1+dfsg-1.

Ben.

-- 
Ben Hutchings
For every complex problem
there is a solution that is simple, neat, and wrong.


signature.asc
Description: This is a digitally signed message part

Bug#342105: certain JavaScript sites make firefox consume 100% CPU

[Ben Hutchings]

I've also experienced this when scrolling pages on
http://www.worldchanging.com which doesn't seem like it would involve
any JS.  Scrolling it a page at a time, or dragging the scrollbar
quickly, seems to keep the X server busy for about half a second.  This
is long-enough that a page-down key can auto-repeat!  Eric Wong's
suggestion to use EXA fixes the problem for me on an "ATI Technologies
Inc RV350 [Mobility Radeon 9600 M10]".

I haven't tried using XAA and Firefox/Iceweasel 2.0, but could do if
necessary.  In any case this seems more like an X server (or driver)
bug, even if the browser is particularly prone to provoking it.

Ben.

-- 
Ben Hutchings
For every complex problem
there is a solution that is simple, neat, and wrong.


signature.asc
Description: This is a digitally signed message part

Bug#404733: Conflicts between symbolic links and directories

[Ben Hutchings]

Debian bug #404733 shows a case where installation of a package fails
because it contains a symbolic link with the same path as a directory
included in an installed package.  I think this is the expected
behaviour.  However, installation of this package worked for me in a
similar situation, and presumably has for other users.  Has there been
any deliberate change in dpkg that would explain why this does work in
some cases or versions?

Ben.

-- 
Ben Hutchings
For every action, there is an equal and opposite criticism. - Harrison


signature.asc
Description: This is a digitally signed message part

Bug#404733: Conflicts between symbolic links and directories

[Ben Hutchings]

You wrote:
> I installed a new version of this package from unstable on a testing
> box, and got no problems on its way.

The problem was reported as appearing during an attempt to upgrade to
version 2.0.0.1+dfsg-1, so it cannot be very well fixed by this version!
However, I'm beginning to suspect that the totem-mozilla package may
actually have been at fault, not iceweasel.

Ben.

-- 
Ben Hutchings
A free society is one where it is safe to be unpopular. - Adlai Stevenson


signature.asc
Description: This is a digitally signed message part

Bug#404733: Conflicts between symbolic links and directories

[Ben Hutchings]

On Sun, 2006-12-31 at 02:15 +, Ben Hutchings wrote:
> Debian bug #404733 shows a case where installation of a package fails
> because it contains a symbolic link with the same path as a directory
> included in an installed package.  I think this is the expected
> behaviour.  However, installation of this package worked for me in a
> similar situation, and presumably has for other users.  Has there been
> any deliberate change in dpkg that would explain why this does work in
> some cases or versions?

I've now found the following comment in processarc.c:

   * For each file, we check to see if it already exists.
   * There are several possibilities:
   * + We are trying to install a non-directory ...
[...]
   * -  It is a directory.  In this case it depends on whether we're
   *trying to install a symlink or something else.
[...]
   *   = If we are trying to install a symlink we do nothing - ie,
   * dpkg will never replace a directory tree with a symlink.  This
   * is to avoid embarrassing effects such as replacing a directory
   * tree with a link to a link to the original directory tree.

I can't follow the logic of the following code, so I'll have to take
this on trust.  However, if someone who's worked on dpkg could confirm
that symbolic links whose paths match existing directories are
intentionally ignored, that would be very much appreciated.  (Does it
make any difference whether the symbolic link was in the previously
installed package?)  The comment is present in sarge and the current
version, so it looks like we can rely on this working during an upgrade
from sarge.

The error message in the bug report is a translation of "trying to
overwrite `%.250s', which is also in package %.250s" which appears to
indicate that totem-mozilla installed something other than a directory
(perhaps another symbolic link?) at /usr/lib/firefox.  Is this the only
possible reason for that message?

Ben.

-- 
Ben Hutchings
A free society is one where it is safe to be unpopular. - Adlai Stevenson


signature.asc
Description: This is a digitally signed message part

Bug#402822: CVE-2006-5330: HTTP header injection vulnerabilities

[Ben Hutchings]

Package: flashplugin-nonfree
Version: 9.0.21.78.3
Severity: important
Tags: security

According to the upstream security bulletin
:

"Adobe has provided a Flash Player updates to resolve potential
vulnerabilities in Adobe Flash Player. These vulnerabilities could
allow remote attackers to modify HTTP headers of client requests and
conduct HTTP Request Splitting attacks."

Adobe classifies this as "important", meaning that it could be
exploited to "compromise data security, potentially allowing access to
confidential data, or could compromise processing resources in a
user's computer."

-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.17-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages flashplugin-nonfree depends on:
ii  debconf [debconf-2.0] 1.5.8  Debian configuration management sy
ii  gsfonts-x11   0.20   Make Ghostscript fonts available t
ii  libxext6  1:1.0.1-2  X11 miscellaneous extension librar
ii  libxmu6   1:1.0.2-2  X11 miscellaneous utility library
ii  libxt61:1.0.2-2  X11 toolkit intrinsics library
ii  wget  1.10.2-2   retrieves files from the web

Versions of packages flashplugin-nonfree recommends:
pn  xfs(no description available)

-- debconf information excluded


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#402822: CVE-2006-5330: HTTP header injection vulnerabilities

[Ben Hutchings]

According to Adobe's bulletin, the following versions are vulnerable:

Flash Player 9.0.20.0 and earlier
Flash Professional 8 [prior to 8.0.34.0]
Flash Player 7.0.68.0 and earlier

Therefore, although Adobe recommends upgrading to 9.0.28.0, I believe
only sarge needs to be updated.

The advisory at <http://www.rapid7.com/advisories/R7-0026.jsp> is more
explicit about the vulnerability: a Flash script can specify values for
Content-Type or custom (non-standard) headers in HTTP requests it makes
and these are not restricted from including CR and LF characters.  This
means that the 'header values' can include additional header lines and
even (if the server supports pipelining) entire requests.

The advisory includes a script fragment that demonstrates the exploit.
A complete example file would be helpful in verifying that etch and sid
are not vulnerable.

Ben.

-- 
Ben Hutchings
Computers are not intelligent.  They only think they are.


signature.asc
Description: This is a digitally signed message part

Bug#402836: CVE-2006-6332: Stack buffer overflow vulnerabilities

[Ben Hutchings]

Package: madwifi
Version: 1:0.9.2+r1842.20061207-1
Severity: critical
Tags: security
Justification: root security hole

According to the upstream advisory
:

"A critical security flaw (CVE-2006-6332) has been discovered which can
be exploited from remote and allows arbitrary code injection. The fix
has been committed to trunk in r1842. In addition, we released
v0.9.2.1 (v0.9.2 plus the fix for the issue), which is available for
download from sf.net."

However, r1842 did *not* completely fix the flaw.  One of the length
checks was wrong and was subsequently fixed in SVN r1847.

Ben.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#402822: CVE-2006-5330: HTTP header injection vulnerabilities

[Ben Hutchings]

On Wed, 2006-12-13 at 07:32 +0100, Bart Martens wrote:

> I'm confused now.  Maybe "earlier" does mean "older".  You seem to be
> from the UK, so probably your English is better than mine.

Yes, earlier means older here.

> Is 9.0.21.78 affected or not according to this bulletin?

It is not.  I would prefer to confirm this with a test case though,
since it is between the last-vulnerable and recommended versions.  If
version 9.0.21.78 was only ever released as a beta for Linux it's
possible they could have forgotten to include it in the vulnerable
versions.

Ben.

-- 
Ben Hutchings
Reality is just a crutch for people who can't handle science fiction.


signature.asc
Description: This is a digitally signed message part

Bug#403498: netpbm-free: copyright file incorrectly says ppmtogif is excluded due to patent issues

[Ben Hutchings]

Package: netpbm-free
Version: 2:10.0-10.1
Severity: minor

The copyright file says:

"""
Not distributed due to patent issues:
[ jbigtopnm: (C) 2000  Markus Kuhn <[EMAIL PROTECTED]> ]
[ pnmtojbig: (C) 2000  Markus Kuhn <[EMAIL PROTECTED]> ]
filter to create GIFs - patent restricted. (http://www.unisys.com/unisys/lzw/
[ ppmtogif:  (A) 1989  Jef Poskanzer ]
"""

However ppmtogif has been distributed since version 2:10.0-4 due to
expiry of the LZW patents.

Ben.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#393483: apt-proxy fails on packages with version including '~'

[Ben Hutchings]

On Tue, 2006-10-31 at 19:13 +, Chris Halls wrote:
> Sorry this mail got delayed on my server, resending.
> 
> On Tue, Oct 24, 2006 at 10:03:33AM +0200, Jerome Warnier wrote:
> > This is still a bug (whoever it belongs to, I don't know) with FTP.
> > Perhaps the bug title should say that ("on FTP backend").
> 
> At a guess this bug will affect all non-http backends. I see that this was
>  not being picked up by the unit tests because the test client was sending
>  unescaped characters to the test apt-proxy.

The rsync fetcher appeared to be unaffected - it's passing a URL rather
than a path to the rsync command.

Ben.

-- 
Ben Hutchings -- [EMAIL PROTECTED] shortened to [EMAIL PROTECTED]
If you've signed my GPG key, please send a signature on and to the new uid.
Usenet is essentially a HUGE group of people passing notes in class.
  - Rachel Kadel, `A Quick Guide to Newsgroup Etiquette'


signature.asc
Description: This is a digitally signed message part

Bug#395094: CVE-2006-545[3-5]: Multiple security issues in bugzilla

[Ben Hutchings]

Upstream security advisory: http://www.bugzilla.org/security/2.18.5/

These are fixed in 2.22.1 which would be suitable for sid.

There is no upstream fix for the 2.16 series, as used in sarge.  I am
looking at the upstream fix for the 2.18 series to see whether it is
applicable or easily adaptable to 2.16.

Ben.

-- 
Ben Hutchings -- [EMAIL PROTECTED] shortened to [EMAIL PROTECTED]
If you've signed my GPG key, please send a signature on and to the new uid.
The world is coming to an end.  Please log off.


signature.asc
Description: This is a digitally signed message part

Bug#406054: dvdauthor: newer version available

[Ben Hutchings]

Package: dvdauthor
Version: 0.6.11-5
Severity: wishlist

Version 0.6.12 was just announced by the author.

-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing'), (100, 'unstable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.17-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages dvdauthor depends on:
ii  libbz2-1.0  1.0.3-6  high-quality block-sorting file co
ii  libc6   2.3.6.ds1-8  GNU C Library: Shared libraries
ii  libdvdread3 0.9.7-2  library for reading DVDs
ii  libfreetype62.2.1-5  FreeType 2 font engine, shared lib
ii  libfribidi0 0.10.7-4 Free Implementation of the Unicode
ii  libice6 1:1.0.1-2X11 Inter-Client Exchange library
ii  libjasper-1.701-1   1.701.0-2The JasPer JPEG-2000 runtime libra
ii  libjpeg62   6b-13The Independent JPEG Group's JPEG 
ii  liblcms11.15-1   Color management library
ii  libmagick9  7:6.2.4.5.dfsg1-0.13 Image manipulation library
ii  libpng12-0  1.2.15~beta5-1   PNG library - runtime
ii  libsm6  1:1.0.1-3X11 Session Management library
ii  libtiff43.8.2-6  Tag Image File Format (TIFF) libra
ii  libx11-62:1.0.3-4X11 client-side library
ii  libxext61:1.0.1-2X11 miscellaneous extension librar
ii  libxml2 2.6.27.dfsg-1GNOME XML library
ii  zlib1g  1:1.2.3-13   compression library - runtime

dvdauthor recommends no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#394025: CVE-2006-5444: Remote compromise in chan_skinny

[Ben Hutchings]

Enough stalling.  Here's a debdiff for a sarge security update.  I have
tested that the warning message is certainly triggered by the bogus
length values Adam found.  I *think* the bug may only be exploitable on
64-bit systems, as read() calls seem to fail immediately where the
length would result in wrapping around the virtual address space.  I
don't have a 64-bit system to test on.

I removed the letohl() calls because the function doesn't exist in
Asterisk 1.0.7 and it wouldn't fix the whole problem (and I don't know
quite what would need to be changed to do so).  If any of the so-called
"maintainers" actually care about this package they could perhaps
arrange to disable chan_skinny on big-endian platforms in this security
update, since it can't work properly and the endian bugs might pose
other security problems.

Ben.

diff -u asterisk-1.0.7.dfsg.1/debian/patches/00list 
asterisk-1.0.7.dfsg.1/debian/patches/00list
--- asterisk-1.0.7.dfsg.1/debian/patches/00list
+++ asterisk-1.0.7.dfsg.1/debian/patches/00list
@@ -18,0 +19,1 @@
+99_CVE-2006-5444.dpatch
diff -u asterisk-1.0.7.dfsg.1/debian/changelog 
asterisk-1.0.7.dfsg.1/debian/changelog
--- asterisk-1.0.7.dfsg.1/debian/changelog
+++ asterisk-1.0.7.dfsg.1/debian/changelog
@@ -1,3 +1,12 @@
+asterisk (1:1.0.7.dfsg.1-2sarge4) stable-security; urgency=high
+
+  * Non-maintainer upload
+  * Backported fix for buffer overflow in chan_skinny driver
+induced by an undetected integer underflow
+[debian/patches/99_CVE-2006-5444.dpatch] (Closes: #394025)
+
+ -- Ben Hutchings <[EMAIL PROTECTED]>  Sat, 25 Nov 2006 16:12:26 +
+
 asterisk (1:1.0.7.dfsg.1-2sarge3) stable-security; urgency=high
 
   * Non-maintainer upload by the Security Team
only in patch2:
unchanged:
--- asterisk-1.0.7.dfsg.1.orig/debian/patches/99_CVE-2006-5444.dpatch
+++ asterisk-1.0.7.dfsg.1/debian/patches/99_CVE-2006-5444.dpatch
@@ -0,0 +1,27 @@
+#! /bin/sh /usr/share/dpatch/dpatch-run
+## 99_CVE-2006-5444.dpatch by <[EMAIL PROTECTED]>
+##
+## All lines beginning with `## DP:' are a description of the patch.
+## DP: Fix buffer overflow in chan_skinny driver induced by an undetected
+## DP: integer underflow (CVE-2006-5444).
+## DP: Remove bogus mutex unlock in case it can cause denial of service.
+
[EMAIL PROTECTED]@
+--- asterisk-1.0.7.dfsg.1/channels/chan_skinny.c.orig  2006-10-20 
00:10:49.0 +
 asterisk-1.0.7.dfsg.1/channels/chan_skinny.c   2006-10-20 
00:16:37.0 +
+@@ -2305,10 +2305,14 @@
+   return -1;
+   }
+   dlen = *(int *)s->inbuf;
++  if (dlen < 0) {
++  ast_log(LOG_WARNING, "Skinny Client sent invalid 
data.\n");
++  return -1;
++  }
+   if (dlen+8 > sizeof(s->inbuf))
+   dlen = sizeof(s->inbuf) - 8;
++  *(int *)s->inbuf = dlen;
+   res = read(s->fd, s->inbuf+4, dlen+4);
+-  ast_mutex_unlock(&s->lock);
+   if (res != (dlen+4)) {
+   ast_log(LOG_WARNING, "Skinny Client sent less data than 
expected.\n");
+   return -1;
-- END --

-- 
Ben Hutchings
Tomorrow will be cancelled due to lack of interest.


signature.asc
Description: This is a digitally signed message part

Bug#400483: fmit: crashes at startup

[Ben Hutchings]

Ludovic RESLINGER <[EMAIL PROTECTED]> wrote:
> If fmit works when you add --disable-alsa, it is probably
> because it is an alsa bug.

Of course!  The fact that it selects different capturing code in fmit
couldn't have anything to do with it!

One or more of these obvious errors might be involved:

1. CaptureThread uses volatile bool flags for communication between
threads, but there's no guarantee that a volatile bool is or even can be
updated atomically.
2. CaptureThread::m_packet_size needs to be consistent with
CaptureThread::m_values (I think) but it is written without the lock
held in CaptureThreadImplALSA::capture_loop() and
CaptureThreadImplJACK::jackProcess().
3. CaptureThreadImplALSA::capture_finished() uses delete to free an
array; it must use delete[] instead.
4. CaptureThreadImplALSA::set_params() will continue if it can't set the
format as requested (however the logs posted to this bug report indicate
that this hasn't happened).  The format could use 32-bit samples, which
would overflow the buffer in captureLoop().

Someone who cares about this program could try to fix them.

Ben.

-- 
Ben Hutchings
It is easier to write an incorrect program than to understand a correct one.


signature.asc
Description: This is a digitally signed message part

Bug#393717: wprint: package claims a modified Microsoft font is "public domain"

[Ben Hutchings]

The font is used for a manual test that does not run in the build
process and seems not to be useful in the binary package.  Therefore I
propose to remove it from the upstream tarball and to remove the test
files from the binary package.

I'm making an NMU with a 2-day delay; the diff follows.

Ben.

diff -Nru /tmp/RcEifLxX8h/wprint-2.05/debian/changelog 
/tmp/Gt586blbMA/wprint-2.05+dfsg/debian/changelog
--- /tmp/RcEifLxX8h/wprint-2.05/debian/changelog2006-10-28 
16:29:58.0 +
+++ /tmp/Gt586blbMA/wprint-2.05+dfsg/debian/changelog   2006-10-28 
16:29:58.0 +
@@ -1,3 +1,11 @@
+wprint (2.05+dfsg-0.1) unstable; urgency=low
+
+  * Non-maintainer upload
+  * Removed non-free font COURR_SE.TTF. Closes: #393717
+  * Removed test data from binary package
+
+ -- Ben Hutchings <[EMAIL PROTECTED]>  Sat, 28 Oct 2006 16:12:40 +0100
+
 wprint (2.05-1) unstable; urgency=low
 
   * New upstream release
diff -Nru /tmp/RcEifLxX8h/wprint-2.05/debian/rules 
/tmp/Gt586blbMA/wprint-2.05+dfsg/debian/rules
--- /tmp/RcEifLxX8h/wprint-2.05/debian/rules2006-10-28 16:29:58.0 
+
+++ /tmp/Gt586blbMA/wprint-2.05+dfsg/debian/rules   2006-10-28 
16:29:58.0 +
@@ -57,8 +57,6 @@
# Add here commands to install the package into debian/tmp.
# $(MAKE) install prefix=`pwd`/debian/tmp/usr
$(MAKE) install prefix=/usr DESTDIR=`pwd`/debian/tmp
-   # (FIXME) This should be done in the Makefile
-   chmod 755 `pwd`/debian/tmp/usr/share/wprint/test.sh
 
 
 # Build architecture-independent files here.
Binary files /tmp/RcEifLxX8h/wprint-2.05/doc/test/COURR_SE.TTF and 
/tmp/Gt586blbMA/wprint-2.05+dfsg/doc/test/COURR_SE.TTF differ
diff -Nru /tmp/RcEifLxX8h/wprint-2.05/doc/test/Makefile.am 
/tmp/Gt586blbMA/wprint-2.05+dfsg/doc/test/Makefile.am
--- /tmp/RcEifLxX8h/wprint-2.05/doc/test/Makefile.am2001-08-14 
14:47:15.0 +
+++ /tmp/Gt586blbMA/wprint-2.05+dfsg/doc/test/Makefile.am   2006-10-28 
16:29:58.0 +
@@ -1,2 +1,2 @@
-pkgdata_DATA = README.test COURR_SE.TXT COURR_SE.TTF netscape.ps news.eo.gif 
search.eo.gif test.html test.sh wprint-test.conf
-EXTRA_DIST = $(pkgdata_DATA)
+#pkgdata_DATA = README.test COURR_SE.TTF COURR_SE.TXT netscape.ps news.eo.gif 
search.eo.gif test.html test.sh wprint-test.conf
+#EXTRA_DIST = $(pkgdata_DATA)
diff -Nru /tmp/RcEifLxX8h/wprint-2.05/doc/test/Makefile.in 
/tmp/Gt586blbMA/wprint-2.05+dfsg/doc/test/Makefile.in
--- /tmp/RcEifLxX8h/wprint-2.05/doc/test/Makefile.in2005-05-11 
14:26:11.0 +
+++ /tmp/Gt586blbMA/wprint-2.05+dfsg/doc/test/Makefile.in   2006-10-28 
16:29:58.0 +
@@ -10,6 +10,9 @@
 # even the implied warranty of MERCHANTABILITY or FITNESS FOR A
 # PARTICULAR PURPOSE.
 
+#pkgdata_DATA = README.test COURR_SE.TTF COURR_SE.TXT netscape.ps news.eo.gif 
search.eo.gif test.html test.sh wprint-test.conf
+#EXTRA_DIST = $(pkgdata_DATA)
+
 
 SHELL = @SHELL@
 
@@ -62,13 +65,8 @@
 PACKAGE = @PACKAGE@
 VERSION = @VERSION@
 ftconfig = @ftconfig@
-
-pkgdata_DATA = README.test COURR_SE.TXT COURR_SE.TTF netscape.ps news.eo.gif 
search.eo.gif test.html test.sh wprint-test.conf
-EXTRA_DIST = $(pkgdata_DATA)
 mkinstalldirs = $(SHELL) $(top_srcdir)/mkinstalldirs
 CONFIG_CLEAN_FILES = 
-DATA =  $(pkgdata_DATA)
-
 DIST_COMMON =  Makefile.am Makefile.in
 
 
@@ -85,25 +83,6 @@
cd $(top_builddir) \
  && CONFIG_FILES=$(subdir)/$@ CONFIG_HEADERS= $(SHELL) ./config.status
 
-
-install-pkgdataDATA: $(pkgdata_DATA)
-   @$(NORMAL_INSTALL)
-   $(mkinstalldirs) $(DESTDIR)$(pkgdatadir)
-   @list='$(pkgdata_DATA)'; for p in $$list; do \
- if test -f $(srcdir)/$$p; then \
-   echo " $(INSTALL_DATA) $(srcdir)/$$p $(DESTDIR)$(pkgdatadir)/$$p"; \
-   $(INSTALL_DATA) $(srcdir)/$$p $(DESTDIR)$(pkgdatadir)/$$p; \
- else if test -f $$p; then \
-   echo " $(INSTALL_DATA) $$p $(DESTDIR)$(pkgdatadir)/$$p"; \
-   $(INSTALL_DATA) $$p $(DESTDIR)$(pkgdatadir)/$$p; \
- fi; fi; \
-   done
-
-uninstall-pkgdataDATA:
-   @$(NORMAL_UNINSTALL)
-   list='$(pkgdata_DATA)'; for p in $$list; do \
- rm -f $(DESTDIR)$(pkgdatadir)/$$p; \
-   done
 tags: TAGS
 TAGS:
 
@@ -134,20 +113,19 @@
 install-exec-am:
 install-exec: install-exec-am
 
-install-data-am: install-pkgdataDATA
+install-data-am:
 install-data: install-data-am
 
 install-am: all-am
@$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am
 install: install-am
-uninstall-am: uninstall-pkgdataDATA
+uninstall-am:
 uninstall: uninstall-am
-all-am: Makefile $(DATA)
+all-am: Makefile
 all-redirect: all-am
 install-strip:
$(MAKE) $(AM_MAKEFLAGS) AM_INSTALL_PROGRAM_FLAGS=-s install
 installdirs:
-   $(mkinstalldirs)  $(DESTDIR)$(pkgdatadir)
 
 
 mostlyclean-generic:
@@ -177,12 +155,12 @@
 
 maintainer-clean: maintainer-clean-am
 
-.PHONY: uninstall-pkgdataDATA install-pkgdataDATA tags distdi

Bug#295270: Messages cannot be deleted and index file gets corrupted when there isn't enough disk space

[Ben Hutchings]

Shouldn't this be assigned to evolution-data-server?  Unless I'm very
much mistaken, Evolution does no folder manipulation by itself.

Ben.

-- 
Ben Hutchings -- [EMAIL PROTECTED] shortened to [EMAIL PROTECTED]
If you've signed my GPG key, please send a signature on and to the new uid.
Once a job is fouled up, anything done to improve it makes it worse.


signature.asc
Description: This is a digitally signed message part

Bug#295270: Messages cannot be deleted and index file gets corrupted when there isn't enough disk space

[Ben Hutchings]

reassign 295270 evolution-data-server
thanks

The problem appears to be that camel_mbox_summary_sync_mbox() in
camel/providers/local/camel-mbox-summary.c marks changes in the summary
as having been applied to the folder as soon as they have been applied
to the new mbox file, even though an error later on - such as running
out of disk space - will result in the new mbox file being removed.  It
should process all summary updates after updating the mbox file.  (Also,
possibly it should delete the summary files immediately before replacing
the mbox file.)

Fabian, this code is specific to mbox but there may be similar problems
with summary/folder synchronisation for other access methods.

Ben.

-- 
Ben Hutchings -- [EMAIL PROTECTED] shortened to [EMAIL PROTECTED]
If you've signed my GPG key, please send a signature on and to the new uid.
The world is coming to an end.  Please log off.


signature.asc
Description: This is a digitally signed message part

Bug#393717: wprint: package claims a modified Microsoft font is "public domain"

[Ben Hutchings]

On Sun, Oct 29, 2006 at 07:51:47PM -0300, Eduardo Trápani wrote:
> Yes, that is the situation.  I'll remove it as soon as possible.
> 
> I see you propose some more cleaning up.  It looks good but I'll check 
> it though.  It's been a while since the last update.

The font is required for running a semi-automated test procedure, and
only that.  Removing the font makes it impossible to carry out the test
- not that it seems appropriate to put test procedures in a binary
package anyway - so I removed all the other test data and scripts from
the binary package.

By the way, if you find it necessary to upload a new source package, the
following incantation appears to be the correct way to rebuild the
autotools files:

autoreconf2.13 --include-deps -l /usr/share/automake-1.4 -m 
/usr/share/autoconf2.13

(it took me a while to work this out).

Ben.

-- 
Ben Hutchings
Humour is the best antidote to reality.


signature.asc
Description: Digital signature

Bug#395080: security issues in asterisk

[Ben Hutchings]

package asterisk
retitle 394025 CVE-2006-5444: Remote compromise in chan_skinny
retitle 395080 CVE-2006-5445: Denial of service in chan_sip
thanks

One bug per bug report, please.

Ben.

-- 
Ben Hutchings
Lowery's Law:
 If it jams, force it. If it breaks, it needed replacing anyway.


signature.asc
Description: This is a digitally signed message part

Bug#395080: CVE-2006-5445: Denial of service in chan_sip

[Ben Hutchings]

_inet_ntoa(iabuf, sizeof(iabuf), p->ourip), 
ourport, p->branch);
+   else
+   snprintf(p->via, sizeof(p->via), "SIP/2.0/UDP 
%s:%d;branch=z9hG4bK%08x", ast_inet_ntoa(iabuf, sizeof(iabuf), p->ourip), 
ourport, p->branch);
+   strncpy(p->callid, callid, sizeof(p->callid) - 1);
+
+   build_contact(p);
+
+   __transmit_response(p, msg, req, 0);
+
+   return 0;
+}
+
 /*--- transmit_response: Transmit response, no retransmits */
 static int transmit_response(struct sip_pvt *p, char *msg, struct sip_request 
*req) 
 {
-- END --

-- 
Ben Hutchings
Reality is just a crutch for people who can't handle science fiction.


signature.asc
Description: This is a digitally signed message part

Bug#294621: xwhois: server file is not found, program won't start

[Ben Hutchings]

tag 294621 patch
thanks

An earlier maintainer of this package overrode the default location of
xwhois.servers, defined by the XWHOIS_SERVERS macro, by editing the
confgiure script.  Re-running autoconf erased that change, since
configure.in had the original default.  I think a reasonable way to
override XWHOIS_SERVERS would be to make this change to debian/rules:

--- xwhois-0.4.2/debian/rules~  2005-02-13 01:04:45.0 +
+++ xwhois-0.4.2/debian/rules   2005-02-13 01:04:45.0 +
@@ -29,7 +29,7 @@

 build:
test -f config.cache || ./configure --prefix=/usr
-   $(MAKE)
+   $(MAKE) XWHOIS_SERVERS=/etc/X11/xwhois/servers
touch stamp-build

 clean: debclean

-- 
Ben Hutchings
Theory and practice are closer in theory than in practice.
- John Levine, moderator of comp.compilers


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#297022: cantus3: Saving edited tag crashes if file is not writable

[Ben Hutchings]

Package: cantus3
Version: 3.0.0-1-1
Severity: normal

I found that Cantus 3 crashed when I tried to save an ID3 tag in one
particular file.  It turned out that it was erroneously write-
protected (permissions were 444) and adding write permission allowed
it to work.  Cantus should check for and report errors when opening
the file for writing.

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing'), (600, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.7-1-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages cantus3 depends on:
ii  libatk1.0-0 1.8.0-4  The ATK accessibility toolkit
ii  libc6   2.3.2.ds1-20 GNU C Library: Shared libraries an
ii  libgcc1 1:3.4.3-6GCC support library
ii  libglade2-0 1:2.4.1-2library to load .glade files at ru
ii  libglademm2.0-1c102 2.2.0-1  C++ wrappers for libglade2 (shared
ii  libglib2.0-02.6.1-2  The GLib library of C routines
ii  libgtk2.0-0 2.4.14-2 The GTK+ graphical user interface 
ii  libgtkmm2.0-1c102   2.2.12-1.1   C++ wrappers for GTK+ 2.0 (shared 
ii  libogg0 1.1.0-1  Ogg Bitstream Library
ii  libpango1.0-0   1.6.0-3  Layout and rendering of internatio
ii  libsigc++-1.2-5c102 1.2.5-1  Type-safe Signal Framework for C++
ii  libstdc++5  1:3.3.5-5The GNU Standard C++ Library v3
ii  libvorbis0a 1.0.1-1  The Vorbis General Audio Compressi
ii  libvorbisfile3  1.0.1-1  The Vorbis General Audio Compressi
ii  libxml2 2.6.11-5 GNOME XML library
ii  zlib1g  1:1.2.2-3compression library - runtime

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#411205: openarena: Description is not written in English

[Ben Hutchings]

Package: openarena
Version: 0.6.0-2
Severity: minor

"Ego shooter" is a German (Denglish) phrase, not English.  The proper
English term is "first-person shooter" or "FPS".  (See
http://de.wikipedia.org/wiki/Ego-Shooter)

Ben.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#333273: xvfb: Uses incorrect line padding when depth>8

[Ben Hutchings]

reopen 333273
close 333273 1:1.0.2-8.1

Brice Goglin wrote:
> I think this has been fixed in upstream XFree86 right after 4.3.0 got
> released. So it's probably fixed in Xorg nowadays.

Agreed.  I can see the bug fix in Xorg's git tree way back in 2003 (!)
and have verified this fix in the current version in Debian.  Therefore
I'm closing this with the first version number for xorg-server in
Debian.

Ben.

-- 
Ben Hutchings
It is easier to change the specification to fit the program than vice versa.


signature.asc
Description: This is a digitally signed message part

Bug#408530: libcapi20-3: buffer overflow in "printbuf" called from capi_cmsg2str

[Ben Hutchings]

clone 408530 -1 -2
reassign -1 asterisk-chan-capi
retitle -1 asterisk-chan-capi: Need a mutex for calls to capi_{cmsg,message}2str
reassign -2 linux-2.6
retitle -2 linux-2.6: capi_{cmsg,message}2str not thread-safe; vulnerable to 
buffer overflow
block -1 with 408530
tags -2 upstream
forwarded -2 http://bugzilla.kernel.org/show_bug.cgi?id=8028
thanks

This function and capi_message2str are not thread-safe either; nor can
they be made so without the use of TSS for their buffers.  chan_capi
will need to use a mutex to prevent collision between concurrent uses of
these functions.  I don't know what can be done in the kernel.  The
buffer overflow could conceivably be due to two concurrent calls to
these functions rather than a single message.

Ben.

-- 
Ben Hutchings
It is easier to change the specification to fit the program than vice versa.


signature.asc
Description: This is a digitally signed message part

Bug#408530: libcapi20-3: buffer overflow in "printbuf" called from capi_cmsg2str

[Ben Hutchings]

tags 408530 patch
thanks

Patch for isdnutils:

diff -u isdnutils-3.9.20060704/debian/rules isdnutils-3.9.20060704/debian/rules
--- isdnutils-3.9.20060704/debian/rules
+++ isdnutils-3.9.20060704/debian/rules
@@ -388,6 +388,7 @@
ppp-2.4.4b1 \
vbox-little-endian \
toplevel-make \
+   capi20-msg2str-safety \
 
 ifeq ($(distribution),Ubuntu)
   debian_patches += no-imake
only in patch2:
unchanged:
--- isdnutils-3.9.20060704.orig/debian/patches/capi20-msg2str-safety.dpatch
+++ isdnutils-3.9.20060704/debian/patches/capi20-msg2str-safety.dpatch
@@ -0,0 +1,58 @@
+#! /bin/sh -e
+
+# DP: Prevent buffer overflow in capi20_{cmsg,message}2str.
+# DP: Add warning that they are not thread-safe.
+
+dir=
+if [ $# -eq 3 -a "$2" = '-d' ]; then
+pdir="-d $3"
+dir="$3/"
+elif [ $# -ne 1 ]; then
+echo >&2 "usage: `basename $0`: -patch|-unpatch [-d ]"
+exit 1
+fi
+case "$1" in
+-patch)
+patch $pdir -f --no-backup-if-mismatch -p0 < $0
+;;
+-unpatch)
+patch $pdir -f --no-backup-if-mismatch -R -p0 < $0
+;;
+*)
+   echo >&2 "usage: `basename $0`: -patch|-unpatch [-d ]"
+exit 1
+esac
+exit 0
+
+--- capi20/capiutils.h~2005-03-08 07:26:47.0 +
 capi20/capiutils.h 2007-02-17 20:22:48.0 +
+@@ -308,6 +308,10 @@
+ #define capi20_cmd2strcapi_cmd2str
+ char *capi_cmd2str(_cbyte cmd, _cbyte subcmd);
+ 
++/*
++ * WARNING: The following two functions use a single static buffer and
++ * are not thread-safe.
++ */
+ #define capi20_cmsg2str   capi_cmsg2str
+ char *capi_cmsg2str(_cmsg * cmsg);
+ 
+--- capi20/convert.c~  2005-05-09 09:23:01.0 +0100
 capi20/convert.c   2007-02-17 20:34:17.0 +
+@@ -894,10 +894,14 @@
+ static void bufprint(char *fmt,...)
+ {
+   va_list f;
++  size_t space = buf + sizeof(buf) - p, len;
+   va_start(f, fmt);
+-  vsprintf(p, fmt, f);
++  len = vsnprintf(p, space, fmt, f);
+   va_end(f);
+-  p += strlen(p);
++  if (len < space - 1)
++  p += len;
++  else
++  p += space - 1;
+ }
+ 
+ static void printstructlen(_cbyte * m, unsigned len)
-- END --

I can't test this in place because I don't know how to construct a
message that would overflow the buffer.  However, the following test
program:

-- BEGIN --
static char buf[8192];
static char *p = 0;

#include 
#include 

static void bufprint(char *fmt,...)
{
va_list f;
size_t space = buf + sizeof(buf) - p, len;
va_start(f, fmt);
len = vsnprintf(p, space, fmt, f);
va_end(f);
if (len < space - 1)
p += len;
else
p += space - 1;
}

int main(void)
{
int i;
p = buf;
p[0] = 0;
for (i = 0; i != 10; ++i)
{
bufprint("%4096s", "foo");
bufprint("%4096s", "bar?");
}
puts(buf);
}
-- END --

shows that output is truncated after the last character that will fit in
the buffer ("r" in this case) as intended.

Ben.

-- 
Ben Hutchings
It is easier to change the specification to fit the program than vice versa.


signature.asc
Description: This is a digitally signed message part

Bug#411293: asterisk-chan-capi: Need a mutex for calls to capi_{cmsg,message}2str

[Ben Hutchings]

tags 411293 +patch
thanks

Patch for asterisk-chan-capi:

diff -u asterisk-chan-capi-0.7.1/debian/patches/00list 
asterisk-chan-capi-0.7.1/debian/patches/00list
--- asterisk-chan-capi-0.7.1/debian/patches/00list
+++ asterisk-chan-capi-0.7.1/debian/patches/00list
@@ -1,0 +2 @@
+02_capi-msg2str-mutex
only in patch2:
unchanged:
--- asterisk-chan-capi-0.7.1.orig/debian/patches/02_capi-msg2str-mutex.dpatch
+++ asterisk-chan-capi-0.7.1/debian/patches/02_capi-msg2str-mutex.dpatch
@@ -0,0 +1,58 @@
+#! /bin/sh /usr/share/dpatch/dpatch-run
+## 02_capi-msg2str-mutex.dpatch by Ben Hutchings <[EMAIL PROTECTED]>
+##
+## All lines beginning with `## DP:' are a description of the patch.
+## DP: Protect calls to capi20_{cmsg,message}2str and use of their
+## DP: buffer with a mutex.
+
[EMAIL PROTECTED]@
+--- asterisk-chan-capi-0.7.0/chan_capi.c~  2006-10-08 13:38:27.0 
+
 asterisk-chan-capi-0.7.0/chan_capi.c   2007-02-17 22:26:32.0 
+
+@@ -167,6 +167,8 @@
+  * 4. cc_mutex_lock(&messagenumber_lock);
+  * 5. cc_mutex_lock(&usecnt_lock);
+  * 6. cc_mutex_lock(&capi_put_lock);
++ * 7. cc_mutex_lock(&capi_msg2str_lock);
++ * 8. cc_mutex_lock(&verbose_lock);
+  *
+  *
+  *  ** the PBX will call the callback functions with 
+@@ -180,6 +182,7 @@
+ AST_MUTEX_DEFINE_STATIC(usecnt_lock);
+ AST_MUTEX_DEFINE_STATIC(iflock);
+ AST_MUTEX_DEFINE_STATIC(capi_put_lock);
++AST_MUTEX_DEFINE_STATIC(capi_msg2str_lock);
+ AST_MUTEX_DEFINE_STATIC(verbose_lock);
+ 
+ static int capi_capability = AST_FORMAT_ALAW;
+@@ -402,6 +405,7 @@
+   return -1;
+   }
+ 
++  cc_mutex_lock(&capi_msg2str_lock);
+   if (error) {
+   cc_log(LOG_ERROR, "CAPI error sending %s (NCCI=%#x) (error=%#x 
%s)\n",
+   capi_cmsg2str(CMSG), (unsigned int)HEADER_CID(CMSG),
+@@ -415,6 +419,7 @@
+   cc_verbose(4, 1, "%s\n", capi_cmsg2str(CMSG));
+   }
+   }
++  cc_mutex_unlock(&capi_msg2str_lock);
+ 
+   return error;
+ }
+@@ -3972,12 +3977,14 @@
+   unsigned short wInfo = 0x;
+   struct capi_pvt *i = find_interface_by_plci(PLCI);
+ 
++  cc_mutex_lock(&capi_msg2str_lock);
+   if ((wCmd == CAPI_P_IND(DATA_B3)) ||
+   (wCmd == CAPI_P_CONF(DATA_B3))) {
+   cc_verbose(7, 1, "%s\n", capi_cmsg2str(CMSG));
+   } else {
+   cc_verbose(4, 1, "%s\n", capi_cmsg2str(CMSG));
+   }
++  cc_mutex_unlock(&capi_msg2str_lock);
+ 
+   if (i != NULL)
+   cc_mutex_lock(&i->lock);
-- END --

Unfortunately I have no suitable hardware with which to test this
package; therefore this is untested except that it builds cleanly.

Ben.

-- 
Ben Hutchings
It is easier to change the specification to fit the program than vice versa.


signature.asc
Description: This is a digitally signed message part

Bug#411293: asterisk-chan-capi: Need a mutex for calls to capi_{cmsg,message}2str

[Ben Hutchings]

Let's try that again with accurate function and directory names:

diff -u asterisk-chan-capi-0.7.1/debian/patches/00list 
asterisk-chan-capi-0.7.1/debian/patches/00list
--- asterisk-chan-capi-0.7.1/debian/patches/00list
+++ asterisk-chan-capi-0.7.1/debian/patches/00list
@@ -1,0 +2 @@
+02_capi-cmsg2str-mutex
only in patch2:
unchanged:
--- asterisk-chan-capi-0.7.1.orig/debian/patches/02_capi-cmsg2str-mutex.dpatch
+++ asterisk-chan-capi-0.7.1/debian/patches/02_capi-cmsg2str-mutex.dpatch
@@ -0,0 +1,57 @@
+#! /bin/sh /usr/share/dpatch/dpatch-run
+## 02_capi-msg2str-mutex.dpatch by Ben Hutchings <[EMAIL PROTECTED]>
+##
+## All lines beginning with `## DP:' are a description of the patch.
+## DP: Protect calls to capi_cmsg2str and use of its buffer with a mutex.
+
[EMAIL PROTECTED]@
+--- asterisk-chan-capi-0.7.1/chan_capi.c~  2006-10-08 13:38:27.0 
+
 asterisk-chan-capi-0.7.1/chan_capi.c   2007-02-17 22:26:32.0 
+
+@@ -167,6 +167,8 @@
+  * 4. cc_mutex_lock(&messagenumber_lock);
+  * 5. cc_mutex_lock(&usecnt_lock);
+  * 6. cc_mutex_lock(&capi_put_lock);
++ * 7. cc_mutex_lock(&capi_cmsg2str_lock);
++ * 8. cc_mutex_lock(&verbose_lock);
+  *
+  *
+  *  ** the PBX will call the callback functions with 
+@@ -180,6 +182,7 @@
+ AST_MUTEX_DEFINE_STATIC(usecnt_lock);
+ AST_MUTEX_DEFINE_STATIC(iflock);
+ AST_MUTEX_DEFINE_STATIC(capi_put_lock);
++AST_MUTEX_DEFINE_STATIC(capi_cmsg2str_lock);
+ AST_MUTEX_DEFINE_STATIC(verbose_lock);
+ 
+ static int capi_capability = AST_FORMAT_ALAW;
+@@ -402,6 +405,7 @@
+   return -1;
+   }
+ 
++  cc_mutex_lock(&capi_cmsg2str_lock);
+   if (error) {
+   cc_log(LOG_ERROR, "CAPI error sending %s (NCCI=%#x) (error=%#x 
%s)\n",
+   capi_cmsg2str(CMSG), (unsigned int)HEADER_CID(CMSG),
+@@ -415,6 +419,7 @@
+   cc_verbose(4, 1, "%s\n", capi_cmsg2str(CMSG));
+   }
+   }
++  cc_mutex_unlock(&capi_cmsg2str_lock);
+ 
+   return error;
+ }
+@@ -3972,12 +3977,14 @@
+   unsigned short wInfo = 0x;
+   struct capi_pvt *i = find_interface_by_plci(PLCI);
+ 
++  cc_mutex_lock(&capi_cmsg2str_lock);
+   if ((wCmd == CAPI_P_IND(DATA_B3)) ||
+   (wCmd == CAPI_P_CONF(DATA_B3))) {
+   cc_verbose(7, 1, "%s\n", capi_cmsg2str(CMSG));
+   } else {
+   cc_verbose(4, 1, "%s\n", capi_cmsg2str(CMSG));
+   }
++  cc_mutex_unlock(&capi_cmsg2str_lock);
+ 
+   if (i != NULL)
+   cc_mutex_lock(&i->lock);
-- END --

-- 
Ben Hutchings
It is easier to change the specification to fit the program than vice versa.


signature.asc
Description: This is a digitally signed message part

Bug#380226: NTFS (partition) not recreated correctly after resize:incorrect start sector

[Ben Hutchings]

reassign 380226 partman-base
thanks

This is not a bug in libparted; parted_server is simply not specifying
the correct constraint for the resize operation.

This patch might fix parted_server, but I don't know how to test it.

diff -u parted_server.c~ parted_server.
--- parted_server.c~2007-01-11 13:57:27.0 +
+++ parted_server.c 2007-02-27 23:50:18.0 +
@@ -712,8 +712,6 @@
 ped_file_system_close(fs);
 fs = NULL;
 }
-if (NULL == fs && NULL != ped_file_system_probe(&(part->geom)))
-return false;
 } else {
 fs = NULL;
 }
@@ -727,10 +725,50 @@
 log("successfully checked");
 if (part->type & PED_PARTITION_LOGICAL)
 maximize_extended_partition(disk);
-if (NULL != fs)
+if (NULL != fs) {
 constraint = ped_file_system_get_resize_constraint(fs);
-else
-constraint = ped_constraint_any(disk->dev);
+} else {
+PedFileSystemType *fs_type;
+PedGeometry *fs_geom;
+PedAlignment start_align;
+PedGeometry full_dev;
+fs_type = ped_file_system_probe(&(part->geom));
+log("probed file system: %s", NULL != fs_type ? "yes" : "no");
+if (NULL != fs_type) {
+fs_geom = ped_file_system_probe_specific(fs_type,
+ &part->geom);
+} else {
+fs_geom = NULL;
+}
+if (NULL != fs_geom) {
+/* We cannot resize or move the fs but we can move the
+ * end of the partition so long as it contains the
+ * whole fs.
+ */
+if (ped_alignment_init(&start_align,
+fs_geom->start, 0)
+&& ped_geometry_init(&full_dev, disk->dev,
+ 0, disk->dev->length - 1)) {
+constraint = ped_constraint_new(
+&start_align,
+ped_alignment_any,
+&full_dev, &full_dev,
+fs_geom->length,
+disk->dev->length);
+} else {
+constraint = NULL;
+}
+ped_geometry_destroy(fs_geom);
+} else {
+constraint = NULL;
+}
+}
+if (open_filesystem && NULL == constraint) {
+log("failed to get resize constraint");
+if (NULL != fs)
+ped_file_system_close(fs);
+return false;
+}
 if (!ped_disk_set_partition_geom(disk, part, constraint, start, end))
 result = false;
 else if (NULL == fs)
-- END --

I'm assuming that open_filesystem will be set when resizing the NTFS
partition.  This should be the case if it's not a virtual partition
(what does that mean?).

Ben.

-- 
Ben Hutchings
Power corrupts.  Absolute power is kind of neat.
   - John Lehman, Secretary of the US Navy 1981-1987


signature.asc
Description: This is a digitally signed message part

Bug#412769: partman-base: parted_server creates FIFOs with wrong mode

[Ben Hutchings]

Package: partman-base
Version: 103.1
Severity: minor

parted_server.c contains the line:

2125:status = mkfifo(name, 0x644);

The mode was presumably meant to be 0644, not 0x644.  Not that it
makes much difference given all of partman runs as root.

Ben.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#412769:

[Ben Hutchings]

found 412769 103
thanks

Oops, version 103.1 is a local version with a patch for #380226.  This
bug is in version 103.

Ben.

-- 
Ben Hutchings
Power corrupts.  Absolute power is kind of neat.
   - John Lehman, Secretary of the US Navy 1981-1987


signature.asc
Description: This is a digitally signed message part

Bug#380226: NTFS (partition) not recreated correctly after resize:incorrect start sector

[Ben Hutchings]

On Thu, 2007-03-01 at 01:04 +0100, Frans Pop wrote:
> On Wednesday 28 February 2007 01:52, Ben Hutchings wrote:
> > This patch might fix parted_server, but I don't know how to test it.
> 
> I've done some extensive testing repeatedly resizing (both down and up) an 
> NTFS Vista partition, and the patch works well. Vista boots correctly 
> after the resize operations.

Well that was lucky!

> It turns out that a minor additional patch in partman-partitioning is 
> necessary (a short sleep to allow the system to settle before running the 
> final 'ntfsresize -f' command).

I hate having to do that...

> I would still very much like to hear the rationale behind the patch and 
> especially an analysis of how/why it will (not) affect resizing other 
> filesystems. This will help to decide how much additional testing is 
> needed.

Here's the patch again with line numbering for reference:

 1:diff -u parted_server.c~ parted_server.
 2:--- parted_server.c~ 2007-01-11 13:57:27.0 +
 3:+++ parted_server.c  2007-02-27 23:50:18.0 +
 4:@@ -712,8 +712,6 @@
 5: ped_file_system_close(fs);
 6: fs = NULL;
 7: }
 8:-if (NULL == fs && NULL != 
ped_file_system_probe(&(part->geom)))
 9:-return false;
10: } else {
11: fs = NULL;
12: }

I moved line 8 down to line 27 where we use the result of the probe to
construct a constraint.  If the probe fails this cascades down to a
constraint of NULL which is detected as an error at line 58.

13:@@ -727,10 +725,50 @@
14: log("successfully checked");
15: if (part->type & PED_PARTITION_LOGICAL)
16: maximize_extended_partition(disk);
17:-if (NULL != fs)
18:+if (NULL != fs) {
19: constraint = ped_file_system_get_resize_constraint(fs);
20:-else
21:-constraint = ped_constraint_any(disk->dev);

The constraint ped_constraint_any(disk->dev) was being applied to
filesystems that parted can't open, which include NTFS.

22:+} else {

This "else" probably should be "else if (open_filesystem)".  What
exactly is a "virtual partition" and why don't we want to look at the
filesystem on it?

23:+PedFileSystemType *fs_type;
24:+PedGeometry *fs_geom;
25:+PedAlignment start_align;
26:+PedGeometry full_dev;
27:+fs_type = ped_file_system_probe(&(part->geom));
28:+log("probed file system: %s", NULL != fs_type ? "yes" : 
"no");
29:+if (NULL != fs_type) {
30:+fs_geom = ped_file_system_probe_specific(fs_type,
31:+ 
&part->geom);
32:+} else {
33:+fs_geom = NULL;
34:+}
35:+if (NULL != fs_geom) {

Should be self-explanatory.

36:+/* We cannot resize or move the fs but we can move 
the
37:+ * end of the partition so long as it contains the
38:+ * whole fs.
39:+ */
40:+if (ped_alignment_init(&start_align,
41:+fs_geom->start, 0)
42:+&& ped_geometry_init(&full_dev, disk->dev,
43:+ 0, disk->dev->length - 1)) 
{
44:+constraint = ped_constraint_new(
45:+&start_align,
46:+ped_alignment_any,
47:+&full_dev, &full_dev,
48:+fs_geom->length,
49:+disk->dev->length);

I think the comment explains this adequately.

50:+} else {
51:+constraint = NULL;
52:+}
53:+ped_geometry_destroy(fs_geom);
54:+} else {
55:+constraint = NULL;
56:+}
57:+}
58:+if (open_filesystem && NULL == constraint) {
59:+log("failed to get resize constraint");
60:+if (NULL != fs)
61:+ped_file_system_close(fs);
62:+return false;
63:+}

If we care about the filesystem on the partition, we do not proceed
unless we know what constraint it imposes on resizing.

This can fail due to:
1. Failure to get the constraint from an open filesystem (line 19).
This is a new possibility for f

Bug#380226: NTFS (partition) not recreated correctly after resize:incorrect start sector

[Ben Hutchings]

->type & PED_PARTITION_LOGICAL)
 maximize_extended_partition(disk);
-if (NULL != fs)
-constraint = ped_file_system_get_resize_constraint(fs);
-else
-constraint = ped_constraint_any(disk->dev);
 if (!ped_disk_set_partition_geom(disk, part, constraint, start, end))
 result = false;
 else if (NULL == fs)
-- END --

Ben.

-- 
Ben Hutchings
Quantity is no substitute for quality, but it's the only one we've got.


signature.asc
Description: This is a digitally signed message part

Bug#380226: NTFS (partition) not recreated correctly after resize:incorrect start sector

[Ben Hutchings]

On Thu, 2007-03-01 at 17:29 +0100, Frans Pop wrote:
> On Thursday 01 March 2007 06:11, Ben Hutchings wrote:
> > On further thinking, I realise there are several problems with the
> > patch:
> > 1. It tries to probe even if open_filesystem is false (already
> > identified).
> > 2. It can return after maximize_extended_partition() without rolling
> > that change back.
> > 3. It doesn't check whether the existing invalid filesystem geometry
> > returned by the probe is valid.
> >
> > This should fix those problems by combining all the work that's
> > conditional on the open_filesystem flag:
> 
> This patch again results in the start of the ntfs partition being moved
> to sector 63.
> 
> Here's the relevant bit from the partman log:
> 
> parted_server: main_loop: iteration 41
> parted_server: Opening infifo
> /lib/partman/active_partition/70resize/do_option: IN: 
> VIRTUAL_RESIZE_PARTITION =dev=scsi=host0=bus0=target0=lun0=disc 
> 1048576-20972568575 150

Ah, so it is using VIRTUAL_RESIZE_PARTITION.  I should have read the
script you pointed me at last time.

What is the intended difference in semantics between RESIZE_PARTITION
and VIRTUAL_RESIZE_PARTITION?  In the resize_partition() function these
are distinguished by the open_filesystem flag which implied to me that
in the latter case we wouldn't expect to find a filesystem in the
partition at all.  Clearly that's not the case here.

This patch reduces the difference that the open_filesystem flag makes.
When the flag is true (for RESIZE_PARTITION) the only change is to check
construction of the constraint.  When the flag is false, the behaviour
will differ as follows:

1. Probing the file-system
   - If true, use ped_file_system_open()
   - If false, use ped_file_system_probe_specific()
   (In any case, this should succeed and produce a file-system
geometry within the partition geometry, or ped_file_system_probe()
should fail indicating that no file-system is present.  If neither
is true, the resize fails.)
2. Resizing the file-system and partition (if found)
   - If true, check the integrity of any file-system, get its resize
 constraint, and resize it.
   - If false, construct a generic constraint for the partition that
 does not require resizing the file-system itself.

I hope these are the correct semantics.

diff -u parted_server.c~ parted_server.c
--- parted_server.c~2007-01-11 13:57:27.0 +
+++ parted_server.c 2007-03-02 01:29:07.0 +
@@ -714,8 +714,58 @@
 }
 if (NULL == fs && NULL != ped_file_system_probe(&(part->geom)))
 return false;
+if (NULL != fs)
+constraint = ped_file_system_get_resize_constraint(fs);
+else
+constraint = ped_constraint_any(disk->dev);
 } else {
+PedFileSystemType *fs_type;
+PedGeometry *fs_geom;
+PedAlignment start_align;
+PedGeometry full_dev;
 fs = NULL;
+fs_type = ped_file_system_probe(&(part->geom));
+log("probed file system: %s", NULL != fs_type ? "yes" : "no");
+if (NULL != fs_type)
+fs_geom = ped_file_system_probe_specific(fs_type,
+ &part->geom);
+else
+fs_geom = NULL;
+if (NULL != fs_geom && (fs_geom->start < (part->geom).start
+|| fs_geom->end > (part->geom).end)) {
+ped_geometry_destroy(fs_geom);
+fs_geom = NULL;
+}
+if (NULL == fs_geom && NULL != fs_type)
+return false;
+if (NULL != fs_geom) {
+/* We cannot resize or move the fs but we can
+ * move the end of the partition so long as it
+ * contains the whole fs.
+ */
+if (ped_alignment_init(&start_align,
+   fs_geom->start, 0)
+&& ped_geometry_init(&full_dev, disk->dev,
+ 0, disk->dev->length - 1)) {
+constraint = ped_constraint_new(
+&start_align,
+ped_alignment_any,
+&full_dev, &full_dev,
+fs_geom->length,
+disk-

Bug#412387: [#412387] patch doesn't fix the real issue

[Ben Hutchings]

Why depend on anything to do with inetd when it's only ever needed on
upgrades from very old versions?  The stable version is 1.9.29, so there
is no need to support upgrades from pre-1.9.

I propose to remove this part of the upgrade logic altogether:

diff -ur apt-proxy-1.9.35.orig/debian/postinst apt-proxy-1.9.35/debian/postinst
--- apt-proxy-1.9.35.orig/debian/postinst
+++ apt-proxy-1.9.35/debian/postinst
@@ -31,11 +31,6 @@
 had_v2_conf=$RET
 
 if dpkg --compare-versions "$PREV"  lt-nl 1.9; then
-if [ -x /usr/sbin/update-inetd ]; then
-  echo "Disabling inetd's apt-proxy v1 line."
-  update-inetd --comment-chars "## " --disable 
-  invoke-rc.d inetd restart
-fi
 db_fset $NAME/upgrading-v2 seen false || true
 db_input medium $NAME/upgrading-v2 || true
     db_go
-- END --

Ben.

-- 
Ben Hutchings
Editing code like this is akin to sticking plasters on the bleeding stump
of a severed limb. - me, 29 June 1999


signature.asc
Description: This is a digitally signed message part

Bug#413282: libsdl1.2debian: Mouse pointer periodically snaps to wrong position

[Ben Hutchings]

Package: libsdl1.2debian
Version: 1.2.11-7
Severity: normal

When I use the mouse in an SDL application (tested in ppracer and
nexuiz), the mouse pointer snaps to the centre of the screen every
~7 seconds.  If I set the environment variable
SDL_VIDEO_X11_DGAMOUSE=0 then the pointer snaps to varying posititons
on the screen at unpredictable intervals.

-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing'), (100, 'unstable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.17-2-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages libsdl1.2debian depends on:
ii  libsdl1.2debian-alsa  1.2.11-7   Simple DirectMedia Layer (with X11

libsdl1.2debian recommends no packages.

-- no debconf information

Versions of X server packages:

ii  xserver-xorg 7.1.0-11  the X.Org X server
ii  xserver-xorg-core1.1.1-15  X.Org X server -- 
core server
ii  xserver-xorg-input-evdev 1.1.2-6   X.Org X server -- 
evdev input driver
ii  xserver-xorg-input-kbd   1.1.0-4   X.Org X server -- 
keyboard input driver
ii  xserver-xorg-input-mouse 1.1.1-3   X.Org X server -- 
mouse input driver
ii  xserver-xorg-input-synaptics 0.14.6-1  Synaptics TouchPad 
driver for X.Org/XFree86
ii  xserver-xorg-input-wacom 0.7.4.1-5 X.Org X server -- 
wacom input driver
ii  xserver-xorg-video-ati   6.6.3-2   X.Org X server -- 
ATI display driver


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#411293: asterisk-chan-capi: Need a mutex for calls to capi_{cmsg, message}2str

[Ben Hutchings]

On Sat, 2007-03-03 at 22:45 +0100, Moritz Muehlenhoff wrote:
> On Mon, Feb 26, 2007 at 08:38:01AM +0100, Lionel Elie Mamane wrote:
> > On Sun, Feb 25, 2007 at 11:53:09PM +0000, Ben Hutchings wrote:
> > > Lionel Elie Mamane <[EMAIL PROTECTED]> wrote:
> > 
> > >> I'm now next to the machine and after some cable fiddling, it seems
> > >> that despite the warning, basic functionality still works; it can
> > >> answer to calls and make calls.
> > 
> > > I'm glad to hear that.  Did you apply both the isdnutils and
> > > asterisk-chan-capi patches (they are both needed to close the security
> > > hole)?
> > 
> > I tested only asterisk-chan-capi, my own recompile of your source
> > package. I tested the sid version.
> > 
> > Time permitting, I may give the isdnutils and kernel patches a shot
> > during the week, as well as the sarge version of asterisk-chan-capi.
> 
> According to the Linux ISDN maintainer CAPI messages triggering such an
> overflow cannot be sent over the ISDN network due to technical limits.

He actually wrote "It can be overflowed by a single evil message from a
local source, but not via the ISDN network."  This means a single
message from the ISDN network cannot cause overflow.

However, unless I'm missing something that ensures capi_cmsg2str is only
ever called from one thread, two messages that are received around the
same time could cause an overflow.  Hence the requirement for a mutex.

> I don't know asterisk-chan-capi, can CAPI messages originate otherwise?

Neither do I.

> If not, it probably doesn't need a fix for Etch.

Ben.

-- 
Ben Hutchings
Q.  Which is the greater problem in the world today, ignorance or apathy?
A.  I don't know and I couldn't care less.


signature.asc
Description: This is a digitally signed message part

Bug#411788: RFH: rt2500 -- RT2500 wireless network drivers

[Ben Hutchings]

On Mon, 2007-03-05 at 02:02 +0100, Romain Beauxis wrote:
> Le lundi 5 mars 2007 01:16, Aurelien Jarno a écrit :
> > If you agree (Ben also), I will open a project on alioth to put our
> > packages and to get a mailing list.
> 
> Yes I do !

Fine by me.

I've finally started testing the rt2x00 driver again myself after
temporarily kicking my wife off the computer with the card in it.  I'm
somewhat doubtful that the current version is suitable for release, let
alone the older version that's in etch.  What do you think?

Ben.

-- 
Ben Hutchings
The most exhausting thing in life is being insincere. - Anne Morrow Lindberg


signature.asc
Description: This is a digitally signed message part

Bug#413444: rt2x00-source: Breaks on suspend

[Ben Hutchings]

Package: rt2x00-source
Version: 0cvs20070219-1
Severity: important

When using the rt2500pci module for kernel 2.6.18-4-k7, I cannot suspend
successfully.

At the first attempt, the kernel begins to suspend and then resumes
without shutting off power.  At the second attempt, the kernel begins to
suspend and then hangs.

With version 0cvs20060928-1 (currently in etch) I can suspend this
system but on resume the driver never completes reinitialisation, though
it does not hang the system.

The same system can suspend and resume reliably without this module
loaded.

-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing'), (100, 'unstable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-4-k7
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages rt2x00-source depends on:
ii  debhelper 5.0.42 helper programs for debian/rules
ii  module-assistant  0.10.8 tool to make module package creati

rt2x00-source recommends no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#413864: videolink: fails to generate output with obscure error

[Ben Hutchings]

On Wed, 2007-03-07 at 10:37 -0600, Neil Schemenauer wrote:
> Package: videolink
> Version: 1.0-1
> Severity: important
> 
> 
> "videolink menu.html dvd_out" dies with the obscure error:
> 
>   Fatal error: /tmp/gzoyi1xj could not be saved, because the
>   source file could not be read.
> 
>   Try again later, or contact the server administrator.
> 
> It looks like this error is coming from xulrunner (Mozilla).

That's correct; only the "Fatal error: " prefix comes from VideoLink.
This looks like an error message that would normally be displayed in a
pop-up window.

Does VideoLink show any messages before this?

Can you send the menu.html file?

Ben.

-- 
Ben Hutchings
Any smoothly functioning technology is indistinguishable from a rigged demo.


signature.asc
Description: This is a digitally signed message part

Bug#413864: videolink: fails to generate output with obscure error

[Ben Hutchings]

On Wed, 2007-03-07 at 18:03 -0600, Neil Schemenauer wrote:
> On Wed, Mar 07, 2007 at 11:20:44PM +0000, Ben Hutchings wrote:
> > Does VideoLink show any messages before this?
> 
> See atttached log file.

There's something odd about this - it should include "loading" and
"saving" messages for the pages and output files involved.

> > Can you send the menu.html file?
> 
> Sure but it's dead simple.

The problem is that VideoLink is treating the video links as page links
because the file names don't have the right extension.  When it follows
the links, Gecko/Xulrunner is unable to display them because they are
videos.  The README file lists the extensions that are recognised.

I know that the distinction between page and video links could be
improved, and that error message is very non-obvious.  However, I'm
going to classify this as a wishlist item because the current
requirements for video links are clearly documented.

Ben.

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.


signature.asc
Description: This is a digitally signed message part

Bug#415709: linux-image-2.6.18-4-686: dv1394 oops on device removal

[Ben Hutchings]

Package: linux-image-2.6.18-4-686
Version: 2.6.18.dfsg.1-11
Severity: normal
Tags: patch, upstream

I plugged in a CardBus 1394/USB card and used it to capture video from
a DV camera.  When I unplugged the card this resulted in an oops,
though I didn't notice that immediately.  When I plugged the card in
again the system hung, apparently due to deadlock in the kernel.

This is a known upstream bug in the dv1394 module and the upstream
patch fixed it for me (it applies to 2.6.18 at a small offset).

http://bugzilla.kernel.org/show_bug.cgi?id=7121

-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing'), (100, 'unstable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-4-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages linux-image-2.6.18-4-686 depends on:
ii  coreutils 5.97-5 The GNU core utilities
ii  debconf [debconf-2.0] 1.5.11 Debian configuration management sy

linux-image-2.6.18-4-686 recommends no packages.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#416009: sgt-puzzles: new release available

[Ben Hutchings]

On Fri, 2007-03-23 at 16:48 -0700, Alex Malinovich wrote:
> There's a new release available. Looks like it's 7400. I know that at
> least Unequal is new, though there are probably others that I haven't
> noticed yet.

I have a pending update to sgt-puzzles, but I'm waiting until etch is
out before I upload it.

Ben.

-- 
Ben Hutchings
Horngren's Observation:
   Among economists, the real world is often a special case.


signature.asc
Description: This is a digitally signed message part

Bug#416505: Decodes bug reports using wrong character encoding

[Ben Hutchings]

Package: reportbug-ng
Version: 0.2007.03.27
Severity: normal

--- Please enter the report below this line. ---
The BTS web pages are served using UTF-8 encoding.  Reportbug NG appears
to display them as if they are in ISO 8859-1 encoding.  (For example,
look at the submitter's name on bug #334640 on sgt-puzzles.)

Ben.

--- System information. ---
Architecture: i386
Kernel:   Linux 2.6.18-4-686

Debian Release: 4.0
  500 testing shadbolt 
  500 testing mirror 
  100 unstablemirror 

--- Package information. ---
Depends (Version) | Installed
=-+-===
python| 2.4.4-2
python-central (>= 0.5.8) | 0.5.12
python-qt3| 3.16-1.2



signature.asc
Description: This is a digitally signed message part

Bug#408530: libcapi20-3: buffer overflow in "printbuf" called from capi_cmsg2str

[Ben Hutchings]

The same patch is applicable to sarge with trivial adjustment.

Ben.

-- 
Ben Hutchings
If God had intended Man to program,
we'd have been born with serial I/O ports.


signature.asc
Description: This is a digitally signed message part

Bug#411293: asterisk-chan-capi: Need a mutex for calls to capi_{cmsg,message}2str

[Ben Hutchings]

Patch for asterisk-chan-capi in sarge (again, untested):

diff -u asterisk-chan-capi-0.3.5/debian/patches/00list 
asterisk-chan-capi-0.3.5/debian/patches/00list
--- asterisk-chan-capi-0.3.5/debian/patches/00list
+++ asterisk-chan-capi-0.3.5/debian/patches/00list
@@ -1,0 +2 @@
+02_capi-cmsg2str-mutex
only in patch2:
unchanged:
--- asterisk-chan-capi-0.3.5.orig/debian/patches/02_capi-cmsg2str-mutex.dpatch
+++ asterisk-chan-capi-0.3.5/debian/patches/02_capi-cmsg2str-mutex.dpatch
@@ -0,0 +1,237 @@
+#! /bin/sh /usr/share/dpatch/dpatch-run
+## 02_capi-msg2str-mutex.dpatch by Ben Hutchings <[EMAIL PROTECTED]>
+##
+## All lines beginning with `## DP:' are a description of the patch.
+## DP: Protect calls to capi_cmsg2str and use of its buffer with a mutex.
+
[EMAIL PROTECTED]@
+--- asterisk-chan-capi-0.3.5/chan_capi.c~  2004-08-13 10:07:28.0 
+
 asterisk-chan-capi-0.3.5/chan_capi.c   2007-02-18 14:45:36.0 
+
+@@ -66,6 +66,7 @@
+ AST_MUTEX_DEFINE_STATIC(contrlock);
+ AST_MUTEX_DEFINE_STATIC(capi_send_buffer_lock);
+ AST_MUTEX_DEFINE_STATIC(capi_put_lock);
++AST_MUTEX_DEFINE_STATIC(capi_cmsg2str_lock);
+ #else
+ static ast_mutex_t usecnt_lock = AST_MUTEX_INITIALIZER;
+ static ast_mutex_t iflock = AST_MUTEX_INITIALIZER;
+@@ -74,6 +75,7 @@
+ static ast_mutex_t contrlock = AST_MUTEX_INITIALIZER;
+ static ast_mutex_t capi_send_buffer_lock = AST_MUTEX_INITIALIZER;
+ static ast_mutex_t capi_put_lock = AST_MUTEX_INITIALIZER;
++static ast_mutex_t capi_cmsg2str_lock = AST_MUTEX_INITIALIZER;
+ #endif
+ 
+ #ifdef CAPI_ULAW
+@@ -1299,7 +1301,9 @@
+   return 0;
+   }
+   if (capidebug) {
++  ast_mutex_lock(&capi_cmsg2str_lock);
+   ast_log(LOG_NOTICE,"%s",capi_cmsg2str(CMSG));
++  ast_mutex_unlock(&capi_cmsg2str_lock);
+   }
+   return -1;
+   }
+@@ -1722,7 +1726,9 @@
+   if (INFO_IND_INFONUMBER(CMSG) == 0x74) {
+   
strncpy(p->i->owner->exten,capi_number(INFO_IND_INFOELEMENT(CMSG),3),sizeof(p->i->owner->exten)-1);
+   
strncpy(p->i->owner->dnid,capi_number(INFO_IND_INFOELEMENT(CMSG),3),sizeof(p->i->owner->dnid)-1);
++  ast_mutex_lock(&capi_cmsg2str_lock);
+   ast_log(LOG_NOTICE,"%s\n",capi_cmsg2str(CMSG));
++  ast_mutex_unlock(&capi_cmsg2str_lock);
+   }
+   if (INFO_IND_INFONUMBER(CMSG) == 0x28) {
+   //  
ast_sendtext(p->i->owner,capi_number(INFO_IND_INFOELEMENT(CMSG),0));
+@@ -1851,8 +1857,11 @@
+   if 
((FACILITY_CONF_FACILITYCONFIRMATIONPARAMETER(CMSG)[4] == 0x0) && 
(FACILITY_CONF_FACILITYCONFIRMATIONPARAMETER(CMSG)[5] == 0x0)) {
+   } else {
+   p->i->state = CAPI_STATE_BCONNECTED;
+-  if (capidebug)
++  if (capidebug) {
++  ast_mutex_lock(&capi_cmsg2str_lock);
+   
ast_log(LOG_NOTICE,"%s\n",capi_cmsg2str(CMSG));
++  
ast_mutex_unlock(&capi_cmsg2str_lock);
++  }
+   }
+   }
+   }
+@@ -1928,8 +1937,11 @@
+ 
+   switch (CMSG->Command) {
+   case CAPI_CONNECT:  // only connect_ind are global (not channel 
specific)
+-  if (capidebug)
++  if (capidebug) {
++  ast_mutex_lock(&capi_cmsg2str_lock);
+   ast_log(LOG_NOTICE,"%s\n",capi_cmsg2str(CMSG));
++  ast_mutex_unlock(&capi_cmsg2str_lock);
++  }
+   DNID = capi_number(CONNECT_IND_CALLEDPARTYNUMBER(CMSG),1);
+   if ((DNID && *DNID == 0) || !DNID) {
+   DNID = emptydnid;
+@@ -2074,24 +2086,33 @@
+   case CAPI_FACILITY:
+   PLCI = FACILITY_IND_PLCI(CMSG) & 0x;  // this is for you 
eicon
+   if (option_verbose > 3) {
+-  if (capidebug)
++  if (capidebug) {
++  ast_mutex_lock(&capi_cmsg2str_lock);
+   ast_verbose(VERBOSE_PREFIX_3 
"%s\n",capi_cmsg2str(CMSG));
++  ast_mutex_unlock(&capi_cmsg2str_lock);
++  }
+   }
+ //ast_log(LOG_ERROR,"FACILITY_IND PLCI=%#x\n",PLCI);
+   break;
+   case CAPI_INFO:
+   PLCI = INFO_IND_PLCI(CMSG);
+   if (option_verbose > 3) {
+-  if (

Bug#409944: l2tpd: l2tpd still uses legacy PTYs, which is or will be removed

[Ben Hutchings]

pload.
+  * Applied upstream change to remove reliance on BSD pseudo-terminals
+(Closes: #276746, #409944).
+  * Corrected AVP numbers for challenge-response and cause code
+(Closes: #265732).
+  * Implemented alignment fix-up for AVPs (Closes: #351870).
+  * Updated no-detach option to pppd (Closes: #298716).
+  * Added dependency on ppp (Closes: #395905).
+  * Updated copyright to mention newer web site and path to GPL text.
+  * Improved package description (Closes: #33).
+
+ -- Ben Hutchings <[EMAIL PROTECTED]>  Tue, 20 Feb 2007 06:20:44 +
+
 l2tpd (0.70-pre20031121-2.1) unstable; urgency=low
 
   * Non-maintainer upload.
reverted:
--- l2tpd-0.70-pre20031121/debian/.#l2tpd.init.1.3
+++ l2tpd-0.70-pre20031121.orig/debian/.#l2tpd.init.1.3
@@ -1,27 +0,0 @@
-#!/bin/sh
-#
-PATH=/bin:/usr/bin:/sbin:/usr/sbin
-DAEMON=/usr/sbin/l2tpd
-PIDFILE=/var/run/l2tpd.pid
-
-test -f $DAEMON || exit 0
-
-case "$1" in
-  start)
-start-stop-daemon --start --verbose --pidfile $PIDFILE --exec $DAEMON 
-;;
-  stop)
-start-stop-daemon --stop --verbose --pidfile $PIDFILE --exec $DAEMON
-;;
-  restart|force-reload)
-start-stop-daemon --stop --verbose --pidfile $PIDFILE --exec $DAEMON 
-sleep 1
-start-stop-daemon --start --verbose --pidfile $PIDFILE --exec $DAEMON
-;;
-  *)
-echo "Usage: /etc/init.d/l2tdp {start|stop|restart|force-reload}"
-exit 1
-;;
-esac
-
-exit 0
diff -u l2tpd-0.70-pre20031121/avp.c l2tpd-0.70-pre20031121/avp.c
--- l2tpd-0.70-pre20031121/avp.c
+++ l2tpd-0.70-pre20031121/avp.c
@@ -775,6 +775,7 @@
 int firstavp = -1;
 int hidlen = 0;
 char *data = buf->start + sizeof (struct control_hdr);
+   u_int32_t alignedbuf[(sizeof (struct avp_hdr) + 0x400 + 3) / 4];
 avp = (struct avp_hdr *) data;
 
 if (debug_avp)
@@ -867,7 +868,7 @@
 /* We want to rewrite the AVP as an unhidden AVP
and then pass it along as normal.  Remeber how
long the AVP was in the first place though! */
-hidlen = avp->length;
+hidlen = rlen;
 if (decrypt_avp (data, t))
 {
 if (debug_avp)
@@ -890,10 +891,12 @@
 else
 hidlen = 0;
 
+   memcpy(alignedbuf, avp, sizeof (struct avp_hdr) + ALENGTH 
(rlen));
+
/* validate */
if (avps[attr].validate) 
{
-   if(avps[attr].validate(attr, t, c, avp, ALENGTH 
(rlen))) {
+   if(avps[attr].validate(attr, t, c, (struct avp_hdr *) 
alignedbuf, ALENGTH (rlen))) {
 if (AMBIT (rlen))
 {
 log (LOG_WARN,
@@ -917,7 +920,7 @@
/* handling */
 if (avps[attr].handle)
 {
-if (avps[attr].handle(t, c, avp, ALENGTH (rlen)))
+if (avps[attr].handle(t, c, (struct avp_hdr *) alignedbuf, ALENGTH 
(rlen)))
 {
 if (AMBIT (rlen))
 {
diff -u l2tpd-0.70-pre20031121/control.c l2tpd-0.70-pre20031121/control.c
--- l2tpd-0.70-pre20031121/control.c
+++ l2tpd-0.70-pre20031121/control.c
@@ -808,7 +808,7 @@
 control_xmit (buf);
 po = NULL;
 po = add_opt (po, "passive");
-po = add_opt (po, "-detach");
+po = add_opt (po, "nodetach");
 if (c->lac)
 {
 if (c->lac->defaultroute)
@@ -900,7 +900,7 @@
 strncpy (ip2, IPADDY (c->addr), sizeof (ip2));
 po = NULL;
 po = add_opt (po, "passive");
-po = add_opt (po, "-detach");
+po = add_opt (po, "nodetach");
 po = add_opt (po, "%s:%s", c->lns->localaddr ? ip1 : "", ip2);
 if (c->lns->authself)
 {
@@ -948,7 +948,7 @@
 case OCCN: /* jz: get OCCN, so the only thing we must do 
is to start the pppd */
 po = NULL;
 po = add_opt (po, "passive");
-po = add_opt (po, "-detach");
+po = add_opt (po, "nodetach");
 po = add_opt (po, "file");
 strcat (dummy_buf, c->dial_no); /* jz: use /etc/ppp/dialnumber.options 
for pppd - kick it if you dont like */
 strcat (dummy_buf, ".options");
-- END --

However, I still could not recommend using this package.  This is
essentially an alpha version and there is no final release.  The package
should probably be replaced with xl2tpd or rp-l2tpd (not that I can
vouch for their quality).

Ben.

-- 
Ben Hutchings
The obvious mathematical breakthrough [to break modern encryption] would be
development of an easy way to factor large prime numbers. - Bill Gates


signature.asc
Description: This is a digitally signed message part

Bug#409324: libmaypole-perl: should include the default "factory templates"

[Ben Hutchings]

On Fri, 2007-02-02 at 00:03 +, Moritz Lenz wrote:
> Package: libmaypole-perl
> Version: 2.10-1
> Severity: wishlist
> 
> Maypole needs templates to work out of the box, in the source distribution
> they are located in lib/Maypole/templates/factory/, however in the debian
> package they are missing.
> 
> Furthermore it would be nice to provide the templates for the example BeerDB
> (in /usr/share/doc/libmaypole-perl/ex/BeerDB.pm, in the source distribtuion
>  they are located in lib/Maypole/templates/beer/

I apologise for taking so long to answer this report.

My understanding is that the "factory" templates can be used as a
starting point for a Maypole application's templates, but are meant to
be copied and customised rather than used by reference.  I originally
packaged Maypole as a dependency of Memories and that certainly does not
require the "factory" templates.

Therefore I believe I should include both the "factory" and "beer"
templates in /usr/share/doc/ex and not /usr/lib.

Ben.

-- 
Ben Hutchings
The obvious mathematical breakthrough [to break modern encryption] would be
development of an easy way to factor large prime numbers. - Bill Gates


signature.asc
Description: This is a digitally signed message part

Bug#410859: liboil0.3: wrongly assumes that all Intel chips support CP14 timestamp

[Ben Hutchings]

I think all XScale processors have CP14 timestamp (see
<http://download.intel.com/design/intelxscale/XScaleDatasheet4.pdf>) but
it's in register 1, not register 0.  The original report is right in
that StrongArm processors do not have a CP14.  Therefore the following
*should* work:

--- liboil-0.3.10.orig/liboil/liboilcpu.c   2006-11-06 01:37:03.0 
+
+++ liboil-0.3.10/liboil/liboilcpu.c2007-02-22 13:41:31.0 +
@@ -620,7 +620,7 @@
 {
   unsigned int ts;
   __asm__ __volatile__ (
-  "  mrc p14, 0, %0, c0, c0, 0 \n"
+  "  mrc p14, 0, %0, c1, c0, 0 \n"
   : "=r" (ts));
   return ts;
 }
@@ -629,7 +629,8 @@
 oil_cpu_detect_arm(void)
 {
 #ifdef __linux__
-  int arm_implementer;
+  int arm_implementer = 0;
+  char arm_architecture = 0;
   char *cpuinfo;
   char *s;
 
@@ -640,18 +641,16 @@
 arm_implementer = strtoul (s, NULL, 0);
 free(s);
   }
+  s = get_cpuinfo_line(cpuinfo, "CPU architecture");
+  if (s) {
+char *p = s + strspn(s, " \t");
+arm_architecture = *p;
+free(s);
+  }
 
-  switch(arm_implementer) {
-case 0x69: /* Intel */
-  /* assume that all Intel chips support CP14 timestamp */
-  _oil_profile_stamp = oil_profile_stamp_xscale;
-  break;
-case 0x41: /* ARM */
-  /* ARM chips are known to not have timestamping available from 
-   * user space */
-  break;
-default:
-  break;
+  /* only Intel XScale processors support CP14 timestamp */
+  if (arm_implementer == 0x69 && arm_architecture == '5') {
+_oil_profile_stamp = oil_profile_stamp_xscale;
   }
 
   free (cpuinfo);
-- END --

In practice it still results in SIGILL, so I agree we should never use
oil_profile_stamp_xscale().

Ben.

-- 
Ben Hutchings
Beware of bugs in the above code;
I have only proved it correct, not tried it. - Donald Knuth


signature.asc
Description: This is a digitally signed message part

Bug#410859: liboil0.3: wrongly assumes that all Intel chips support CP14 timestamp

[Ben Hutchings]

This was already fixed upstream on 1st January
(http://webcvs.freedesktop.org/liboil/liboil/liboil/liboilcpu.c?r1=1.36&r2=1.37)
 so I'm going to apply that change and upload:

--- liboilcpu.c 2006/11/06 02:04:58 1.36
+++ liboilcpu.c 2007/01/01 09:33:33 1.37
@@ -615,15 +615,17 @@
 /* arm */
 
 #ifdef __arm__
+#if 0
 static unsigned long
 oil_profile_stamp_xscale(void)
 {
   unsigned int ts;
   __asm__ __volatile__ (
-  "  mrc p14, 0, %0, c0, c0, 0 \n"
+  "  mrc p14, 0, %0, c1, c0, 0 \n"
   : "=r" (ts));
   return ts;
 }
+#endif
 
 static void
 oil_cpu_detect_arm(void)
@@ -643,9 +645,6 @@
 
   switch(arm_implementer) {
 case 0x69: /* Intel */
-  /* assume that all Intel chips support CP14 timestamp */
-  _oil_profile_stamp = oil_profile_stamp_xscale;
-  break;
 case 0x41: /* ARM */
   /* ARM chips are known to not have timestamping available from 
    * user space */
-- END --

Ben.

-- 
Ben Hutchings
Beware of bugs in the above code;
I have only proved it correct, not tried it. - Donald Knuth


signature.asc
Description: This is a digitally signed message part

Bug#380552: a possible medium term soloution

[Ben Hutchings]

"peter green" <[EMAIL PROTECTED]> wrote:
> would it be possible to reduce the test to only cause a warning until
> it is worked out what about the build environment is causing the false
> positive?

I think that may be a reasonable thing to do here.  This test is trying
to check that "pwd" can read very long paths, specifically 256 levels of
directories below the test's current directory.   The error message
indicates that it is failing after going 265 levels up (i.e. outside of
the build directory, closer to the root).

The error seems to suggest that lstat() failed, though none of the
documented reasons for failure seem plausible.  Another possibility is
that the filesystem on the buildd is corrupt and one ".." link does not
point to the correct directory.  (Lack of read permission on a directory
is another possibility I thought of, but that results in a different
error message.)

Ben.

-- 
Ben Hutchings
Beware of bugs in the above code;
I have only proved it correct, not tried it. - Donald Knuth


signature.asc
Description: This is a digitally signed message part