[Peter Lundblad]
This is an error in the book. Hook script error output is only
marshalled if the scripts returns a non-zero value.
Can I close this bug? It seems the behavior is intentional. (Also, we
don't ship the Subversion Book at this time, so we can't fix it.)
Peter
signature.asc
Package: texlive-doc-cs+sk
Version: 2005-1
Severity: minor
The description for texlive-doc-cs+sk includes quot; which looks
like a remnant of HTML.
Also, the sub-packages in texlive-doc-sk+sk and quite a few of the
other texlive-doc-* packages are not described well at all:
texlive-cz -- The
tags 354877 upstream
thanks
[Vincent Lefevre]
The Subversion manual says:
If the hook program writes data to stderr, it will be marshalled
back to the client.
But this message is output only in case of failure (instead of
everytime).
Thanks for reporting this upstream, saving me
reassign 354510 libaprutil1
thanks
[Adrian Bunk]
Please switch from db4.3 to the more recent db4.4.
Subversion must use the same Berkeley DB version as libaprutil (the
Apache Portable Runtime utility library from the Apache Group).
Currently we link to libapr0, but soon we will switch to
[Paul Pergamenshchik]
When I run equivs-build faketwisted, where faketwisted contains the line
Provides: python2.2-twisted python-twisted python2.3-twisted
Package names must be comma-separated:
dpkg-gencontrol: warning: can't parse dependency python2.2-twisted
python-twisted
tags 350141 patch
thanks
Debian diff.gz files don't have timestamps in the patch headers, so the
patched 'configure' may or may not get a timestamp newer than the files
it was generated from (some of which are also patched). A simple
workaround follows.
--- debian/rules
+++ debian/rules
@@
[Paul Wise]
I recently upgraded gpm (running debian sid), and was very annoyed
when my mouse froze in gnome/X. Simply switching to a console, moving
the mouse a bit till the cursor showed, then switching back to X
fixed this, but it is still annoying.
This is why we have a debconf question:
Package: miscfiles
Version: 1.4.2-1
The miscfiles copyright file says GNU-manifesto.gz is distributed under
the GPL. The file itself says Modified versions may not be made.
Please edit the file to remove the misleading statement, perhaps
replacing it with a standard GPL notice header. (Or, if
[Stefani Banerian]
* Package name: sqlitemanager
Version : x.y.z
Upstream Author : Name [EMAIL PROTECTED]
* URL : http://www.example.org/
* License : (GPL, LGPL, BSD, MIT/X, etc.)
The information content here is rather thin.
signature.asc
Description:
[Philip Martin]
Is there a Subversion packaging problem ahead? If libsvn0 starts to
use libapr1 this will change the ABI of libsvn0 and any package built
against the existing libsvn0 ABI may not work correctly with the new
libsvn0.
Way ahead of you. We're planning to call it libsvn1.
[Philip Martin]
Isn't libsvn1 a new upstream version number?
Why don't you check the SONAMEs for yourself? You can do this with
objdump -p. I won't spoil the surprise, but I will say that I think
you'll discover the remarkable fact that neither libsvn0 nor
libsvn1 is an exact match for the
[Ph. Marek]
Does it make sense to upgrade to libapr1? This would need a
recompiled subversion too, doesn't it?
Yes, but more importantly, if we upgrade to libapr1 we break
mod_dav_svn. It needs to use the same apr that apache is using. Thus
we will upgrade to libapr1 when apache 2.2 hits
[Tollef Fog Heen]
Either that, or make sure YESSTR/NOSTR (from I18N::Langinfo) are
usefully populated and use those. (They seemed generally not to be
based on the discussion on #debian-devel yesterday).
It's not clear to me whether YESSTR/NOSTR are supposed to be
single-letter abbreviations,
reassign 345639 libc6
retitle 345639 nl_langinfo(YESEXPR) ignores LANGUAGE, no apparent workaround
thanks
[Claudio Nieder]
$ env | grep -e LC -e LANG
LANG=en_GB.UTF-8
LANGUAGE=de_CH:de_DE:de:en_GB:en
In this setup, a prompt Is this information correct? [y/N] does not
work.
- The prompt is
[Denis Barbier]
On Sun, Feb 12, 2006 at 06:27:35PM +0100, Claudio Nieder wrote:
Otherwise, if for libc supported language xx_YY no message catalog
is supplied, the user would receive the prompt in untranslated
english (y/N), but expected to type either ä or è because thats
what
[Martin v. Löwis]
I also tried compiling/linking with ncursesw instead; this didn't
change anything.
Right, python Modules/_cursesmodule.c has no explicit support for
ncursesw functions.
The reason the python curses code displays ä is purely accidental:
python believes it is sending *two*
[Peter Samuelson]
In the ISO-8859 family, bytes 0x80-0xbf are invalid - and the UTF-8
encoding of Ä is 0xc3 0x84.
Doh! Of course I meant to say bytes 0x80-0x9f are invalid. Anyway,
ncurses seems to reject that same range of bytes even when LC_CTYPE
indicates UTF-8.
signature.asc
[Anand Kumria]
my $subcommand;
+ my $subcommandopt;
if ($prog eq 'cvs' || $prog eq 'svn' || $prog eq 'svk' ||
$prog eq 'bzr' || $prog eq 'git' ) {
$subcommand = 'diff';
} else {
$subcommand = 'file-diff';
}
- foreach my $line
[Frank Lichtenheld]
| 1. ERROR in
/build/buildd/subversion-1.3.0/subversion/bindings/java/javahl/src/org/tigris/subversion/javahl/CommitMessage.java
| (at line 1)
| /**
| ^
| The type java.lang.Object cannot be resolved. It is indirectly referenced
from required .class files
[Erik Schanze]
I have seen your pending tag on this bug and wait for an upload, but
nothing happens for a half a year now.
Oh, sorry about that. I applied the bugfix but then forgot to actually
upload it! I'll see if I can do that in the next day or so.
Thanks for the reminder,
Peter
[dann frazier]
* Package name: slimscrobbler
Description : SlimServer plugin that submits listening data to Last.FM
OK...
Version : x.y.z
Upstream Author : Name [EMAIL PROTECTED]
* URL : http://www.example.org/
* License : (GPL, LGPL, BSD, MIT/X,
[dann frazier]
Version : x.y.z
Upstream Author : Name [EMAIL PROTECTED]
* URL : http://www.example.org/
* License : (GPL, LGPL, BSD, MIT/X, etc.)
(Include the long description here.)
I couldn't really come up with anything longer - the short
Package: havp
Version: 0.76-2
Severity: minor
The bullet list in the long description of this package is formatted
badly. To illustrate, run aptitude show havp. To fix it, indent
each line starting with - by one extra space.
signature.asc
Description: Digital signature
Package: wnpp
Severity: normal
I intend to hijack 'equivs', as its maintainer (Fabio Rafael da Rosa
[EMAIL PROTECTED]) is believed MIA.
I will wait a week or so to see he shows up.
Peter
signature.asc
Description: Digital signature
[Andreas Hoenen]
* Package name: dblatex
DocBook to LaTeX Publishing that transforms your SGML/XML DocBook
documents to DVI, PostScript or PDF by translating them in pure LaTeX
as a first process. MathML 2.0 markups are supported, too. It is a
clone of DB2LaTeX.
Please explain in your
[Christoph Berg]
Concerning the naming, I was inclined to name the script also
madison, but that would collide with the original, as devscripts is
also installed on *.d.o hosts.
I'd go with a name like 'rmadison' (remote madison) or 'wmadison' (web
scraper madison).
signature.asc
[Britton Leo Kerin]
Now since fldiff can be invoked 'fldiff file1 file2' just like diff,
you would expect it to work here, or if not, you would expect a
specification somewhere from svn saying just how it intends to invoke
the supplied diff command. But I didn't find anything like that.
merge 324511 350133
thanks
[Arthur de Jong]
I would like to make svn2cl available in Debian. I'm a DD so I can
package it (and someone else also did an ITP [2]) but I wonder,
because it's such a small package, if it could be made part of the
subversion-tools package.
Yes, it probably should
[Louis-David Mitterrand]
A strace shows that it aborts upon encountering a file over 2GB in
/home/ldm. If no large files are present then the checkout completes.
Yes, this is a problem with libapr0, which is fixed in libapr1.0.
subversion can be fixed by compiling it against libapr1.0, but
[Olaf van der Spek]
Is there a (good) reason these headers are comments?
If not, please uncomment them.
No good reason. I agreed with upstream's reply to you two days ago,
but hadn't thought about patching Debian's copy.
I expect this fix will get into upstream 1.3.x eventually, but if not,
[Max Bowsher]
From 1.3.0 onwards, adequate Makefile support is in place to make this
fairly easy.
Attached is a patch which adds a new libsvn0-doc subpackage containing
doxygen docs.
Thanks! Looks pretty straightforward.
+Package: libsvn0-doc
+Section: libdevel
+Priority: extra
[Henning Makholm]
... Note that I am the upstream author myself. I wrote this tool
because I need it myself and there seems to be nothing comparable
available on the net. Then I put some extra work into polishing it up
for general consumption, because it seemed to be the Right Thing to
do -
[Max Bowsher]
Oh, OK. I freely admit to being inexperienced with the Debian BTS.
My interpretation of the above was that it was not applicable in this
case, since I don't know which version fixed it, only that is was
fixed at some point between the two versions that I tested. It is
quite
[Norbert Tretkowski]
1.3.0-1 fails to build when using swig 1.3.24-1 (the version in
sarge), after upgrading to 1.3.27-1 from testing it builds fine.
I think Blair Zajac fixed this as part of the patch he submitted in
#347775 - the $(MAKE) -C $(DEB_BUILDDIR) extraclean-swig-headers line
he
Package: busybox-udeb
Version: 1:1.01-4
Severity: wishlist
I had a desire tonight to mount an NFS volume from inside
debian-installer. This doesn't seem unreasonable or unusual to me, so
could you consider enabling CONFIG_NFSMOUNT in config-udeb-linux?
Note, I haven't actually verified that
[Skliarouk Arieh]
This time the segfault is not related to locale (as previous ones).
From my observations it is related to directory depths.
Thank you for the precise reproduction recipe! Unfortunately, I was
unable to reproduce this - either with the 1.2.3dfsg1-3 or with the
1.3.0-1 in
[Skliarouk Arieh]
libapr0 2.0.55-3
libc6 2.3.5-6
libdb4.3 4.3.28-3
libxml2 2.6.22-1
zlib1g 1.2.3-4
Thanks - I'm running the same of everything except libc6 2.3.5-8.1, but
that really doesn't seem likely to be the problem.
I still can't reproduce the problem. I tried nesting
[Skliarouk Arieh]
Yes, the version of subversion and libsvn0 were 1.2.3dfsg1-3
For some reason package subversion-tools was REMOVEd and not upgraded.
Does it matter?
No, doesn't matter.
$ uname -a
Linux desktop 2.4.27-1-386 #1 Fri Sep 3 06:24:46 UTC 2004 i686 GNU/Linux
I installed
[Peter Samuelson]
diff -urN src/main.c src/main.c
--- src/main.c
+++ src/main.c
@@ -99,6 +99,7 @@
cloudlist_t *curcloud;
obslist_t *curobs;
int n = 0;
+ float qnh;
I actually meant 'double qnh;' there, sorry. I thought I'd tested the
patch, but apparently
[Blair Zajac]
I'm a Subversion committer and working on taking the Debian Etch
Subversion packages and back-porting them for Ubuntu Breezy Badger.
This is a small patch to Debian's build system to make it easier and
to document issues I ran into.
s/Etch/Experimental/, I see. Which is fine.
[Thorsten Gunkel]
No, I can live with that. If I want to overwrite that files anyway I
can delete them manually if svn complains.
So in short the solution to my problem is
URL:http://subversion.tigris.org/faq.html#in-place-import?
That's it, yeah.
There have been some requests for
[Thorsten Gunkel]
However when I want to do a checkout on another machine svn will complain
because etc already exists:
# svn update
svn: Working copy 'etc' is missing or not locked
# cd /etc
# svn checkout http://your.server/your/etc .
This will store the files directly in /etc
Package: metar
Version: 20050807.1-1
Tags: patch
metar has this idea that my city has atmospheric pressure of 3007
inches of mercury. Since I haven't suffocated yet, this seems
unlikely. In reality, of course, it is 30.07 inches.
I don't know if it gets the other units right - if not, it's
[Jim Paris]
This is definitely the openssl bug. It appears that the fix in
subversion 1.2.3dfsg1-3 only postponed the problem until libneon24
upgraded to openssl 0.9.8.
That seems unlikely since libneon24 in unstable uses openssl 0.9.8.
...E, wait, are you saying openssl 0.9.7 has the
[Peter Samuelson]
That seems unlikely since libneon24 in unstable uses openssl 0.9.8.
...E, wait, are you saying openssl 0.9.7 has the bug, or 0.9.8?
Never mind. Having read #338006, all is clear now. Thanks again for
the information.
Peter
signature.asc
Description: Digital signature
[Ross Boylan]
Minimally, the instructions could just say to be aware of ownership
and permission issues. I know that's pretty obvious, but I managed
to miss it :)
I think I would have missed it too! Here is what I've put in for the
next upload - let me know if you think it's still
[Ross Boylan]
Instead, how about this?
Sounds good, I just made minor adjustments at the end.
-'hooks' directories into the new copy of the repository. Then, after
+'hooks' directories into the new copy of the repository, and use
+'chown' and 'chmod' to make the access permissions
[Aurelien Jarno]
subversion fails to build on GNU/kFreeBSD because java is not (yet)
available on this platform. Please find attached a patch to disable
java for this architecture, as it is already done for other
architectures.
Applied. Next upload should be within a day or two.
Thanks,
severity 319165 normal
reassign 319165 subversion-tools
merge 311440 319165
thanks
A patch for this bug will be in the next upload, which I hope to get
out in the next day or so.
Peter
signature.asc
Description: Digital signature
tags 304541 upstream wontfix
thanks
[Yann Dirson]
Here is an excerpt from a load_dirs session involving UTF8-encoded
filenames that appear to confuse some of the tools.
Please correct me if I misunderstood your bug report!
It's a consequence of a fundamental design decision in Subversion: it
I found a bug today in the unneeded-explicit-linking lintian test: I
was building a major new release of a package with several libraries in
it, and an old release was also installed on the system:
libfoo_x in the package contains a symbol foo_x_new_function
/usr/lib/libfoo_x.so.0 does not
[Peter Samuelson]
The following patch creates the trivial option variable
DPEP_OMIT_TIMESTAMPS which goes into effect if set (normally set in
00options). I don't know if a corresponding command line option is
useful. I'll add one if you think so.
[Junichi Uekawa]
This feature sounds
[Justin Pryzby]
$ bts found 322253 6.8.99.901.dfsg.1-2
DB_File needs compatible versions of libdb db.h
you have db.h version 4.3.29 and libdb version 4.3.28
#343468, fixed in perl 5.8.7-10.
signature.asc
Description: Digital signature
Package: tar
Version: 1.15.1-2
tar forgets to take --strip-components into account when extracting
hard links:
$ mkdir links
$ touch links/a
$ ln links/a links/b
$ tar cf links.tar links
$ rm -r links
$ tar xf links.tar --strip-components=1
tar: b: Cannot hard link to `links/a':
-*-
+#
+# Test suite for GNU tar.
+# Written by Peter Samuelson, who abandons this file to the public domain.
+#
+# Test extraction of hard links with a --strip-components flag.
+
+AT_SETUP([extracting hard links])
+AT_KEYWORDS([extract06])
+
+AT_TAR_CHECK([
+mkdir blim blim/tibbek
+genfile --length
Package: dpatch
Version: 2.0.16
Severity: wishlist
Tags: patch
As a sort of followup to #321320, I'd like the ability to tell
dpatch-edit-patch not to put timestamps on patch header lines at all.
They're not very useful, and as noted in the bug log for #321320, they
create extra interdiff noise
[Tilman Koschnick]
the attached patch to svn_load_dirs adds support for symbolic links.
Should probably be tested some more, but Works For Me(tm).
The patch looks correct. I tweaked it a bit (for example, not to
depend on /bin/readlink, which won't be on all non-Debian systems) and
I'll add
[Steve Halasz]
I'd like to throw in a vote here too. I want to use these bindings.
I've been updating the packaging for Subversion 1.3.0, which upstream
is expected to release in a couple of weeks. I'll try to add the Ruby
bindings, since multiple people have started to ask for them.
Package: dpatch
Version: 2.0.16
Severity: wishlist
Tags: patch
I suddenly realised a couple weeks ago that I think -p0 format patches
look nicer than -p1 format patches, especially in a closed system like
dpatch where there's no possible confusion about what project or
version a patch applies
[intrigeri]
The package does not put the usual files into
/etc/apache2/mods-available/
mod_dav_svn and mod_authz_svn are loaded by a single .load file. Do
you think it would ever be useful to enable one but not both? If you
can give a good reason, I'll consider changing it.
APT policy:
[Justin Pryzby]
$ svnadmin
--version hotcopy recover
create list-dblogs rmlocks
d list-unused-dblogs rmtxns
deltify loadsetlog
dumplslocks verify
help
Package: swig
Version: 1.3.27-1
Severity: important
Justification: causes FTBFS of unrelated software (subversion)
swig has a builtin C preprocessor which apparently is doing something
wrong when a .i file includes apr-0/apr.h from libapr0-dev:
$ cat apr.c
#include apr.h
$ gcc -c
[A Mennucc]
$ bts fixed #bugnumber #version [why]
that sends an email to -done with a Version: pseudo-header ;
I don't like the idea of encouraging people to fit the reason for
closing a bug onto a single line.
(I'm also the guy who very rarely uses svn commit -m message, for the
same
[Jean-Marc Ranger]
Everyone agrees that CC isn't free, and removing svnbook from main
was the right thing to do. However, I'd personaly like to be able to
install a local copy of svnbook, preferably as a Debian package. At
least one other package's documentation (svn-buildpackage) currently
[Wesley J. Landaker]
As described by the upstream website (the rest of this is a quote):
What is sendcard?
Sendcard is a multi-database (It currently supports 9 different
databases!) e-card or virtual postcard program written in PHP. Suitable
for large or small sites, it is very easy to
[Lior Kaplan]
* Package name: culmus-fancy
Description : Type1 Fancy Hebrew Fonts for X11
I understand that the 'culmus' package already exists, and other
packages like 'lmodern' don't follow any particular name convention
either, but could you consider naming this thing
[Henning Makholm]
I have written a Lintian check which attempts to flag instances of
this problem. It looks for ELF objects that flag shared libraries in
the default search path as NEEDED without actually importing symbols
that the library exports.
This produces a lot of noise in a case
Package: notification-daemon
Version: 0.2.2-1
Severity: minor
The short description (first line of the Description field) and the
long description (rest of the Description field) need to be
independent. I suggest:
Description: daemon to display pop-up notification windows
[Erinn Clark]
It'd rock to have audioscrobbler support for cplay since all of the
other players that support it are... not as good. I have no idea how
this could be implemented, but maybe it's easy.
Yeah, I don't know anything about the AS protocol, and a bit of
preliminary research doesn't
[Steffen Joeris]
This also means that 915resolution must be run every time the
computer boots inorder for it's changes to take effect.
its
signature.asc
Description: Digital signature
[Jean-Marie Thomas]
Subversion book is missing in the aforementioned package, while
packages.debian.org mentions it as packaged.
Please explain what you mean when you say packages.debian.org mentions
it as packaged. Where is it mentioned?
signature.asc
Description: Digital signature
I'll go one farther than the submitter of #313371 and claim that *all*
the Killed by signal N messages are unneeded. Most Unix programs
don't print such messages; that's the shell's job.
The reason I care is that subversion can use a ssh pipe, and needs to
be able to kill the pipeline when it
tags 338781 pending
thanks
[Daniel Nylander]
Here is the swedish translation for the debconf template for gpm.
Thanks! It looks good, so I've added it for the next upload.
Peter
signature.asc
Description: Digital signature
[Philip Martin]
Install the db4.2-util package and replace
svnadmin recover path/to/repo
with
db4.2_recover -h path/to/repo/db
That's what I've been leaning toward, but I am still trying to read
enough source and docs to be reasonably sure this does everything.
signature.asc
[Charles Fry]
All of the standard svn command that I can think of (update, commit,
diff) default to the current directory if no PATH is specified.
Unfortunately, svn revert is inconsistent in this regard in that it
insists on the explicit specification of a PATH, rather than
defaulting to
severity 336781 minor
retitle 336781 subversion: harmless warning in commit-email.pl hook script
thanks
[Michael Nacey]
On line 224, the commit-email.pl file has the line:
my $tmp_dir = ( -d $ENV{'TEMP'} ? $ENV{'TEMP'} : '/tmp' );
This fails when TEMP is not defined.
It shouldn't fail,
[Brian May]
See bug #87371. The title is wrong. devfs is not the issue,
/dev/pts is (I think). I don't think there is any intention to fix
the bug. Over 4 years old. Perhaps this is grounds to drop the
package from Debian.
[Miles Bader]
I'd say so. Or fix the bug.
Kind of quick and
[Philip Martin]
The Subversion FAQ has an item that was written when Brane's Windows
build was upgraded from 4.2 to 4.3.
http://subversion.tigris.org/faq.html#bdb43-upgrade
Yes, well, that procedure assumes you still have a /usr/bin/svnadmin
linked to db4.2. For various reasons that would
[Laszlo Boszormenyi]
Can you please try out the 0.25.4 packages located at
http://www.barcikacomp.hu/deb/ and report back if they are working
right for you?
libtool: link: `/usr/lib/libgssapi.la' is not a valid libtool archive
This is because /usr/lib/libneon.la mentions
Package: libssl0.9.8
Version: 0.9.8-3
libssl0.9.8.shlibs is not strict enough. It should mention (= 0.9.8a)
because of the addition of symbol versioning.
I had a previous version of libssl0.9.8 on my system, and got a failure
trying to link to a library that uses libssl0.9.8, because of lack
Package: libneon24-dev
Version: 0.24.7.dfsg-2
Severity: minor
'neon-config --libs' commits the sin made famous by libtool: it assumes
we don't have an ELF runtime linker that can resolve dependencies
recursively, so drags in extra libraries we don't really need.
The correct solution apparently
svnadmin create repos and other svn commands that need UUIDs don't
work entropy isn't available. They hang.
Not to deny your claim of a bug in apr, but in subversion's case, the
real bug is that subversion doesn't particularly need random numbers.
For the next upload we're considering
[Peter Samuelson]
Oh, doh! /usr/bin/svn will not use libneon25, so that accomplished
nothing.
I came up with several ways around this - the latest is to use
libneon24 but *not* link libssl0.9.8. There was never any reason for
us to link to openssl at all; this was a packaging bug.
Can you
Package: libssl0.9.8
Version: 0.9.8-3
[Kurt Roeckx]
You mean 0.9.8a-1 has the bug since 0.9.8-3 didn't have symbol
versioning yet.
Uh, yeah. Sorry. I started to file the bug before upgrading, since
that's when I noticed the effects.
It looks like we did something wrong, or something
[Sven-Haegar Koch]
Having had the same problem, your test-packages recompiled locally
(just decreased the version number, I want the official one to
upgrade the test one when its released) fixes it for me.
Great! We'll close the bug on the next upload, then, unless Thomas
reports that this
[Michael Biebl]
is there any progress with this bug? I got a package which depends on
libsvn0-dev and wanted to upload a new version. Unfortunately the
db4.2db4.3 conflict makes it impossible.
We've got a db4.3-enabled subversion pretty much ready to go, *except*
for tracking down a simple
[Thomas Petazzoni]
Since you have a ready test case for this, can you try the neon
package at
http://www.barcikacomp.hu/deb/libneon25_0.25.4.dfsg-1_i386.deb, as
mentioned in Bug #335574?
With a Debian sid updated yesterday and your package, it still doesn't
work:
Oh, doh!
[Peter Samuelson]
Oh, doh! /usr/bin/svn will not use libneon25, so that accomplished
nothing.
I'll build a new set of packages that use libneon25 instead
Well, now I see why Debian ships both neon24 and neon25. subversion
won't yet compile against neon25, and I don't have the time
First, apologies for the delayed response - I didn't get the BTS mail
until the other day, possibly because of the known mail backlog
affecting certain Debian services.
svn: MKCOL of
'/svn/thomas/!svn/wrk/b82d4a0a-4a04-0410-8ac2-c33f329d32ff/uclibc-sos/trunk/test/string':
Could not read
[Thomas Petazzoni]
Maybe it's a problem around libneon24 (linked against openssl0.9.7)
and the fact that subversion is linked against openssl 0.9.8.
Since you have a ready test case for this, can you try the neon package
at http://www.barcikacomp.hu/deb/libneon25_0.25.4.dfsg-1_i386.deb, as
The fact that neon links to libssl0.9.7 *might* be causing #336373
against subversion, due to svn (indirectly) linking to both openssl
0.9.7 and 0.9.8. (apt-cache tells me that bazaar, davfs2 and kdesvn
are in the same situation.) I've asked the submitter of #336373 to try
your new package.
[Roland Stigge]
It would also be possible to return the differences-found state via a
return/exit value of the script. But that would diverge from common
practice (see e.g., diff(1)).
Heh - did you read diff(1)? diff *does* in fact return 1 iff any
differences are found. So does cmp. grep
severity 330824 wishlist
retitle 330824 disagreement about example scripts
thanks
[micah]
BTW: Maybe you can move the example files to /u/s/d/s-tools/example/
Thanks so much for not changing the severity there. I guess you
believe this is release-critical. I require a bit of convincing on
Were you able to reproduce the bug using the testcase I supplied on
Friday evening?
I never received a test case. And for some odd reason, I only just now
(November 5) got this message from you from October 25. Could you
resend the test case?
Thanks,
Peter
signature.asc
Description:
[Kobayashi Noritada]
Subversion Ruby binding is available since 1.2.0.
It will be fully implemented in the upcoming release, 1.3.0.
Please package it.
Is this just a general request (this exists, it ought to be compiled)
or do you know of anyone actually writing tools in ruby that want to
Package: figfonts
Version: 2.2.1-1.1
Severity: serious
/usr/share/doc/figfonts/copyright does not tell us anything about the
license terms by which Debian distributes the package. Even though
it's in non-free, we still need some assurance that Debian has the
right to distribute these fonts at
[asdf]
The svn cmd line client segfaults, on an up-to-date Sid/unstable
system, when doing a checkout of a project just added to the
repository.
Thanks for the report. I tried and failed to reproduce your bug, using
a 400 MB backup of an old home directory. That contains a little of
[asdf]
Can you provide me with debug .debs of subversion? IE .debs that
contain debug binaries and libraries of subversion. I can try and run
subversion in GDB and provide you with stack trace information, etc.
I built a set with debugging enabled and pointed him privately to it.
He may also
[Mattia Dongili]
* Package name: acpica-unix
iasl compiles ASL (ACPI Source Language) into AML (ACPI Machine
Language). This AML is suitable for inclusion as a DSDT in system
firmware. It also can disassemble AML, for debugging purposes.
The name is a bit silly, IMO. It's not as
[Daniel Burrows]
you'd have to do something like aptitude keep ~Rdepends:^gnome$.
Too arcane. (:
I've occasionally wanted a simple command in aptitude for remove the
auto flag from all the depends of this package - not only for
metapackages but also for dummy upgrade packages. Another way to
601 - 700 of 793 matches
Mail list logo
