Bug#636840: O: jaranalyzer

2011-08-07 Thread Florian Grandel
Hi Torsten, I'm still available to maintain this package if necessary. I agree with you, though, that the package has a low popcon value and I myself do not use it right now. So for me it's not a problem to orphane it. Florian On 08/06/2011 05:41 AM, Torsten Werner wrote: Package: wnpp

Bug#543089: Should be ok in 1.2-4

2009-08-23 Thread Florian Grandel
Hi Lucas, thanks for you bug report! This bug should be fixed with 1.2-4 which is waiting for sponsorship on mentors.debian.net: - URL: http://mentors.debian.net/debian/pool/main/j/jaranalyzer - Source repository: deb-src http://mentors.debian.net/debian unstable main contrib non-free - dget

Bug#543301: override: jaranalyzer:section/priority

2009-08-23 Thread Florian Grandel
Package: ftp.debian.org Severity: normal jaranalyzer_1.2-4_all.deb: package says section is java, override says devel. I think the package should be moved to section java as it is mainly useful for java developers (determining dependencies between jars). Other java development packages are being

Bug#540175: jaranalyzer: depends on java-gcj-compat

2009-08-14 Thread Florian Grandel
A fixed package can be found on mentors.debian.net: - URL: http://mentors.debian.net/debian/pool/main/j/jaranalyzer - Source repository: deb-src http://mentors.debian.net/debian unstable main contrib non-free - dget http://mentors.debian.net/debian/pool/main/j/jaranalyzer/jaranalyzer_1.2-4.dsc

Bug#540175: jaranalyzer: depends on java-gcj-compat

2009-08-06 Thread Florian Grandel
Hi Marcus, jaranalyzer has an unconditional dependency on java-gcj-compat. Is there a reason for this? No there is no reason for this. Thanks for spotting the error. I'll fix it. Florian -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe.

Bug#511879: Public Subversion Repository

2009-07-13 Thread Florian Grandel
A public subversion repository with the current state of affairs is now available at: You can get the code from [1] (web interface) or [2] (anonymous svn) Help and collaboration is welcome. Those willing to contribute should read [3]. If you don't have an Alioth account yet then register at

Bug#535306: Package uploaded to mentors

2009-07-10 Thread Florian Grandel
Hi Cyril, thank you very much for pointing this error out. A patched package has been created and will be uploaded soon. Florian -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#529365: Workaround

2009-05-19 Thread Florian Grandel
I now use the following workaround until the issue has been resolved: *In debian/config*: db_get package/purge if [ $RET = true ]; then ACTION=reconfigure else ACTION=$1 fi dbc_go package $ACTION $2 *In debian/postinst*: THIS_VERSION=installed package version if [ -n $2 ]; then

Bug#529365: remove + re-install same version: dbconfig-common does not reconfigure

2009-05-19 Thread Florian Grandel
i'm a bit short on time so the patch is appreciated. Sure! It's already 4 in the morning here, so I'll do the patch tomorrow. ;-) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#529365: bug529365.patch

2009-05-19 Thread Florian Grandel
/debian/changelog 2009-05-19 14:11:58.0 -0300 *** *** 1,3 --- 1,9 + dbconfig-common (1.8.42) unstable; urgency=low + + * closes: #529365 (user cannot re-install database after remove) + + -- Florian Grandel jerico@gmail.com Tue, 19 May 2009 14:10:50 -0300

Bug#529365: remove + re-install same version: dbconfig-common does not reconfigure

2009-05-18 Thread Florian Grandel
Package: dbconfig-common Version: 1.8.38ubuntu1 Severity: normal Hi Sean, Please follow these steps to reproduce the problem: - install a package that uses dbconfig-common to install a database - remove the package (but do not purge) and uninstall the database with dbconfig-common when you're

Bug#511879: packaging waits for upstream response

2009-05-16 Thread Florian Grandel
See http://pkp.sfu.ca/support/forum/viewtopic.php?f=9t=4514p=17576#p17576 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#509557: ITP: xdoclet -- attribute oriented programming

2008-12-23 Thread Florian Grandel
Package: wnpp Severity: wishlist Owner: Florian Grandel jerico@gmail.com * Package name: xdoclet Version : 1.2.3 Upstream Author : XDoclet Team * URL : http://xdoclet.sourceforge.net/xdoclet/index.html * License : non-standard OSS license Programming

Bug#509544: ITP: libxjavadoc-java -- a re-implementation of sun's javadoc, required as a dependency for xdoclet

2008-12-22 Thread Florian Grandel
Package: wnpp Severity: wishlist Owner: Florian Grandel jerico@gmail.com * Package name: libxjavadoc-java Version : 1.1 Upstream Author : XDoclet authors * URL : http://xdoclet.sourceforge.net/xjavadoc/ * License : own license Programming Lang: Java

Bug#471680: libaspell-dev

2008-12-09 Thread Florian Grandel
@Tiago: libaspell-dev is a dependency of libgtkspell-dev so no need to explicitly include it in the control file see: # apt-cache depends libgtkspell-dev libgtkspell-dev Depends: libaspell-dev ... -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble?

Bug#471679: patch

2008-12-09 Thread Florian Grandel
(2.0.0-2) unstable; urgency=low + + * Non-maintainer upload. + * fixed missing dependency for switchspell (closes: #471679, #471680) + + -- Florian Grandel [EMAIL PROTECTED] Tue, 09 Dec 2008 15:12:07 -0200 + purple-plugin-pack (2.0.0-1) unstable; urgency=low * New upstream release (Closes

Bug#506173: syslog-ng should not conflict with klogd

2008-11-18 Thread Florian Grandel
Package: syslog-ng Version: 2.0.9-1ubuntu5 Severity: wishlist When running as an unprivileged user there currently only exists the dd workaround I proposed in [1] (see [2] and [3] for a discussion of alternatives). IMO it would be a lot cleaner to use klogd for kernel ringbuffer forwarding to a

Bug#503210: expanded patch, includes kmsg solution

2008-11-17 Thread Florian Grandel
Hi everybody, using syslog-ng with a non-privileged user causes trouble when /proc/kmsg is defined as a source. Access to /proc/kmsg is limited to processes with the CAP_SYS_ADMIN capability. So even resetting /proc/kmsg's file permissions will not solve the problem. The non-privileged user

Bug#505791: syslog-ng doesn't chdir before chroot

2008-11-15 Thread Florian Grandel
Package: syslog-ng Version: 2.0.9-1ubuntu3 Severity: grave Tags: security Justification: user security hole I have not had the time to analyze all of syslog-ng code. But by reading the code section near the chroot call and looking at strace results I believe that syslog-ng does not chdir to the

Bug#505797: syslog-ng: typo in postrm dissappear

2008-11-15 Thread Florian Grandel
Package: syslog-ng Version: 2.0.9-1ubuntu3 Severity: normal The postrm-script contains a typo (see bug title) -- System Information: n/a -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#503210: patch for init script

2008-11-15 Thread Florian Grandel
diff -u syslog-ng-2.0.9/debian/syslog-ng.init syslog-ng-2.0.9/debian/syslog-ng.init --- syslog-ng-2.0.9/debian/syslog-ng.init +++ syslog-ng-2.0.9/debian/syslog-ng.init @@ -1,4 +1,4 @@ -#! /bin/sh +#! /bin/bash ### BEGIN INIT INFO # Provides: syslog # Required-Start:$local_fs

Bug#505797: patch

2008-11-15 Thread Florian Grandel
--- syslog-ng-2.0.9.orig/debian/syslog-ng.postrm +++ syslog-ng-2.0.9/debian/syslog-ng.postrm @@ -9,7 +9,7 @@ fi # remove disabled files on purge or complete overwrite. -if [ $1 = purge -o $1 = dissappear ]; then +if [ $1 = purge -o $1 = disappear ]; then # main file [ -f

Bug#505791: patch

2008-11-15 Thread Florian Grandel
--- syslog-ng-2.0.9.orig/src/main.c +++ syslog-ng-2.0.9/src/main.c @@ -275,7 +275,15 @@ { if (chroot_dir) { - if (chroot(chroot_dir) 0) + if (chdir(chroot_dir) 0) +{ + msg_error(Error during chdir() before chroot(), +

Bug#503210: oups, patch was buggy

2008-11-15 Thread Florian Grandel
The previous patch was buggy, here a corrected one. I couldn't come up with a more elegant solution... The variables SYSLOGNG_USER, SYSLOGNG_GROUP and SYSLOGNG_CHROOT may be set in /etc/default/syslog-ng --- syslog-ng-2.0.9/debian/syslog-ng.init +++ syslog-ng-2.0.9/debian/syslog-ng.init @@

Bug#503210: syslog-ng should start as non-privileged user by default

2008-10-23 Thread Florian Grandel
Package: syslog-ng Version: 2.0.9-4 Severity: wishlist I think that syslog-ng should make use of the -u and -g options on daemon startup by default or otherwise provide good documentation and support for configuring the user/group and possibly chroot. If I have not overlooked s.th. then the init

Bug#495147: This bug was probably introduced by the fix for #310781

2008-08-15 Thread Florian Grandel
Hi Javier, Your fix for #310781 seems to have introduced this bug (see there): The original inline copyStat did a os.chown(target,statInfos[stat.ST_UID],statInfos[stat.ST_GID]) shutil.copystat does not do this. Florian -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of

Bug#310781: Bug only partly fixed (and introduces new bug)

2008-08-15 Thread Florian Grandel
Hi Javier, sorry for bugging you even more. :-| 1) I think this part of the original bug report has not been fixed: After a quick look to the code, it seems like *ACL or extended attributes* wouldn't be reproduced either 2) And I am afraid that your fix for this bug (using shutil.copystat)

Bug#495277: makejail produces non-predictable results without strace -ff

2008-08-15 Thread Florian Grandel
Package: makejail Version: 0.0.5-7 Severity: normal by default makejail writes all strace output into a single file. This however produces unpredictable results whenever child processes are forked. In this case strace may produce output like: 13578 access(/etc/ld.so.preload, R_OK unfinished

Bug#495277: the corresponding strace config

2008-08-15 Thread Florian Grandel
I forgot to post the corresponding strace commands: straceCommand=strace -e trace=file,connect -e signal=none -f -ff -o %file %command /dev/null 21 straceCommandPid=strace -e trace=file,connect -e signal=none -f -ff -o %file -p %pid /dev/null 21 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED]

Bug#495112: makejail does not allow to define default values

2008-08-14 Thread Florian Grandel
Package: makejail Version: 0.0.5-7 Severity: normal The makejail package installs /etc/makejail as a directory. This path should however be free for a configuration file containing configuration default values. (See man page!!) Please rename the directory to /etc/makejail.d or similar to free up

Bug#495116: makejail must use ldconfig.real by default

2008-08-14 Thread Florian Grandel
Package: makejail Version: 0.0.5-7 Severity: serious makejail defines as a default: self.pathToLdConfig=/sbin/ldconfig On Ubuntu this is not the real ldconfig but a bash shell script. Therefore the config should point to self.pathToLdConfig=/sbin/ldconfig.real. Otherwise a bash shell will be

Bug#495118: makejail: documentation error in man page

2008-08-14 Thread Florian Grandel
Package: makejail Version: 0.0.5-7 Severity: normal The makejail man page says: cleanJailFirst ... Default: 1 This does not seem to be correct. I think the real default is 0. See self.cleanJailFirst=0 -- System Information: n/a -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL

Bug#495147: makejail does not chown copied files

2008-08-14 Thread Florian Grandel
Package: makejail Version: 0.0.5-7 Severity: normal When the copied file has ownership other than root:root then makejail fails to correctly copy such files to the jail. makejail should preserve file ownership (as does jailkit f.ex.). Otherwise subtle errors may result when trying to start the

Bug#489900: Cleanup in .orig.tar.gz or debian/rules?

2008-07-17 Thread Florian Grandel
Hi Kalle, Kalle Kivimaa schrieb: You can either modify the .orig.tar.gz not to include the jars, or, IMO a better choice, modify the building process to use the Debian provided jars. I personally prefer keeping the orig.tar.gz as close to the original as possible (for JSPWiki all the Debian

Bug#489214: javahelper: jh_manifest produces non-spec

2008-07-13 Thread Florian Grandel
I have added a work around (new packages at the same URL), where I do (effectively): fastjar xf $jar rm -f $jar fastjar ufm $jar newmanifest * I can't use any other jar and have it in main still. This works fine now! :-) You can close bug #489432 as well. I have already updated my

Bug#489214: javahelper: jh_manifest produces non-spec MANIFEST.MF

2008-07-12 Thread Florian Grandel
Hi Matthew, Can you please try these fixed packages, I've put them up at: http://mjj29.matthew.ath.cx/debian-upload/java-tools/ The jar filesearch is now correct. IMO you may close bug #489214. Anyway: jh_manifest still produces an invalid jar due to a bug in fastjar -u option (see fastjar

Bug#489214: javahelper: jh_manifest produces non-spec MANIFEST.MF

2008-07-09 Thread Florian Grandel
Matthew Johnson schrieb: I shall try building it and reproducing that error, but in order to test the rest of the program, can you try applying this diff: --- jh_manifest.orig 2008-07-08 15:13:23.970424750 + +++ jh_manifest 2008-07-08 15:14:02.756848750 + @@ -73,8 +73,9 @@

Bug#489214: javahelper: jh_manifest produces non-spec MANIFEST.MF

2008-07-09 Thread Florian Grandel
Hi Matthew, I think I found the problem. I ran your script in the python debugger and found out that in line 99 scriptutil.ffind(path, shellglobs=('*.jar')) returns the following: ['debian/jaran...n/jaranalyzer', 'debian/jaran...README.Debian', 'debian/jaran...zer/copyright',

Bug#489214: javahelper: jh_manifest produces non-spec MANIFEST.MF

2008-07-07 Thread Florian Grandel
Hi Matthew, Hi, please can you try out this new version I have just written in python. Hopefully it should fix all these bugs. Packages at: http://mjj29.matthew.ath.cx/debian-upload/java-tools/ You will also need this package which I will upload to NEW, but you can get from me:

Bug#489214: javahelper: jh_manifest produces non-spec MANIFEST.MF

2008-07-05 Thread Florian Grandel
Matthew Johnson schrieb: Why should Debian-Java-Home not be in the main section? it applies to the jar as a whole? You are completely right. I was confused about this point. Sure, any attribute applying to the jar as a whole should be in the main section. Sorry about that. -- To

Bug#489418: Incompatibility fastjar-java-6-openjdk/jar

2008-07-05 Thread Florian Grandel
Package: fastjar Version: 2:0.95-1 Severity: normal A jar updated with fastjar is not correctly read by the jar application provided by openjdk's jar application. Once the bug number is established I'll send in an example jar. Steps to reproduce the problem: - produce a jar using openjdk jar -

Bug#489418: Acknowledgement (Incompatibility fastjar-java-6-openjdk/jar)

2008-07-05 Thread Florian Grandel
Here comes the test jar (see attachment). jaranalyzer-1.2.jar Description: Binary data

Bug#489418: Acknowledgement (Incompatibility fastjar-java-6-openjdk/jar)

2008-07-05 Thread Florian Grandel
Ok, I've tried. But Google doesn't let me send any executable (including .jar) through their SMTP. Anybody who wants the test jar please contact me directly. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#489418: Acknowledgement (Incompatibility fastjar-java-6-openjdk/jar)

2008-07-05 Thread Florian Grandel
Ok, I have seen now that the jar passed the SMTP, it just took some time. I further researched the issue and found out that the problem occurs when updating the manifest file with fastjar using the m-Option. Maybe this helps to find the error. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED]

Bug#489432: javahelper: combining jh_manifest/jh_depends hits Bug#489418

2008-07-05 Thread Florian Grandel
Package: javahelper Version: 0.7 Severity: normal Using jh_manifest to update a manifest file hits Bug#489418. The following jh_depends will therefore not be able to correctly extract the manifest. A quick workaround: Index: jaranalyzer-1.2/jh_manifest

Bug#489164: ITP: jaranalyzer -- A Dependency management utility for jar files.

2008-07-03 Thread Florian Grandel
Package: wnpp Severity: wishlist Owner: Florian Grandel [EMAIL PROTECTED] * Package name: jaranalyzer Version : 1.2 Upstream Author : Kirk Knoernschild [EMAIL PROTECTED] (remove the _filter) * URL : http://www.kirkk.com/main/Main/JarAnalyzer * License : non

Bug#489214: javahelper: jh_manifest produces non-spec MANIFEST.MF

2008-07-03 Thread Florian Grandel
Package: javahelper Version: 0.7 Severity: normal jh_manifest may produce a MANIFEST.MF that does not conform to the jar specification. See http://java.sun.com/j2se/1.3/docs/guide/jar/jar.html. The following problems may occur: Blank lines are significant. They separate sections within the

Bug#489214: Quick workaround

2008-07-03 Thread Florian Grandel
Hi Matthew, here is a patch that works around the first part of the problem (blank lines) and implements the nice-to-have part. This however is not yet really satisfying as the resulting MANIFEST.MF will still not be to spec. The non-main attributes will be ignored and the 72-bytes problem