Bug#1025206: RFP: bespoke -- Software modular music synthesizer
Package: wnpp Severity: wishlist X-Debbugs-Cc: j...@joshtriplett.org * Package name: bespoke Version : 1.1.0 Upstream Author : Ryan Challinor * URL : https://www.bespokesynth.com/ https://github.com/BespokeSynth/BespokeSynth/ * License : GPLv3 Programming Lang: C++ Description : Software modular music synthesizer Bespoke is a graphical tool to synthesize music, by wiring together modular components.
Bug#933579: ITP: openvr -- openvr sdk
On Thu, Aug 01, 2019 at 02:23:21AM +0800, Andrew Lee (李健秋) wrote: > Package: wnpp > Severity: wishlist > Owner: Andrew Lee (李健秋) > > * Package name: openvr > Version : 1.4.18 > Upstream Author : Valve Corporation > * URL : https://github.com/ValveSoftware/openvr > * License : Expat > Programming Lang: C > Description : openvr sdk > > OpenVR is an API and runtime that allows access to VR hardware from > multiple vendors without requiring that applications have specific > knowledge of the hardware they are targeting. This repository is an > SDK that contains the API and samples. The runtime is under SteamVR > in Tools on Steam. Can OpenVR work with any Open Source backend? (And do you intend to package such a backend?) It appears to only work with SteamVR, which is proprietary. If this only works with SteamVR or other proprietary backends, or in general if Debian doesn't have any Open Source backend for this, it'll have to go to contrib, not main.
Bug#913285: RFP: oxipng -- Parallel lossless PNG compression optimizer
Package: wnpp Severity: wishlist * Package name: oxipng Version : 2.1.5 Upstream Author : Joshua Holmer * URL : https://github.com/shssoichiro/oxipng * License : MIT Programming Lang: Rust Description : Parallel lossless PNG compression optimizer Oxipng optimizes PNG images to make them use less storage and bandwidth. In parallel, oxipng evaluates possible PNG encoding parameters, zlib DEFLATE encoding parameters, and optionally Zopfli-optimized DEFLATE encoding parameters, writing out the optimized image that uses the least storage and bandwidth.
Bug#842943: Also interested in this
I'd love to see a package of the Signal standalone desktop client in main. As far as I can tell, the github repository at https://github.com/signalapp/Signal-Desktop has the source of the standalone desktop client. - Josh Triplett
Bug#883701: RFP: turtl -- Note-taking application with secure online syncing
Package: wnpp Severity: wishlist * Package name: turtl Version : 0.6.0 Upstream Author : i...@turtlapp.com * URL : https://turtlapp.com/ * License : GPLv3 for the client, AGPLv3 for the server Programming Lang: JavaScript and Rust Description : Note-taking application with secure online syncing Turtl is a note-taking application that supports sharing and syncing notes securely, with client-side encryption, so that the server has no access to your notes. The client and server are both FOSS; I'd love to see both packaged, though the client would be higher priority. The server uses node.js; the client uses NW.js. Both have a common library written in Rust.
Bug#860116: [Pkg-rust-maintainers] Bug#860116: RFH: cargo -- Rust package manager
On Thu, Apr 20, 2017 at 08:16:00AM +, Ximin Luo wrote: > Josh Triplett: > > [..] > > > > ...and looking at git from 20 minutes ago, it looks like you've switched > > over to directory registries now. > > > > Thanks for the tip, yes I figured this out by looking at both your dh-cargo > code and also the Fedora cargo package. [1] > > The new version is uploaded to experimental, but there's some more things I'd > like to tidy up before uploading it to unstable. > > - I get errors about missing "cargotest" and "hamcrest" when I try to `make > test`, it looks like cargotest is actually in the rustc package: > > https://github.com/rust-lang/rust/tree/master/src/tools/cargotest > > Any suggestions on how to deal with this? I guess the obvious thing to do is > to package a librust-cargotest-xxx-dev package but I wonder if there are > better options. > > - cargo still embeds libgit2-sys source code, and I can see that in the > deps-tarball-filter.txt Luca was explicitly leaving it in. Is that because > they patch the source code? > > Can we just get rid of it, and link to libgit2 instead? git2-rs doesn't modify libgit2, but it does use a snapshot from git, via a submodule. And unfortunately, while it supports building against a system version via pkg-config, it doesn't include any version number in its dependency. libgit2 itself doesn't have a stable ABI, so either an older *or* a newer version can potentially fail due to incompatibilities. (Also, like many libraries, libgit2 doesn't really update its SONAME and similar except when releasing, even though many people run git snapshots.) Fortunately, git2-rs uses ctest now to attempt to make sure its bindings match the actual signatures of the functions, so at least an attempt to build against an older (or sometimes newer) version of libgit2 will tend to produce compile errors (assuming you get all the environment variables set correctly to find it). However, I know of at least one case where git2-rs pulled in a new snapshot of libgit2 to obtain critical bugfixes rather than differences in ABI. In short, I think we *could* switch to pulling in the system version of libgit2, but we'd need to work very closely with the libgit2 maintainer, and sometimes package snapshots of libgit2 from git. We might also want to work with libgit2 and git2-rs upstreams to 1) get more regular updates to the version number that appears in the pkg-config file and 2) get libgit2-sys to declare a pkg-config dependency that matches its snapshot of libgit2. - Josh Triplett
Bug#860116: [Pkg-rust-maintainers] Bug#860116: RFH: cargo -- Rust package manager
On Wed, Apr 19, 2017 at 11:11:00PM +, Ximin Luo wrote: > Ximin Luo: > > [..] > > I've made some more process and have pushed 0.17.0 to git. However we're > getting build errors and I don't know how to proceed: > > https://anonscm.debian.org/cgit/pkg-rust/cargo.git/tree/debian/TODO > https://anonscm.debian.org/cgit/pkg-rust/cargo.git/tree/debian/cargo-vendor-pack.py > > I'm not familiar with the Cargo repo formats, and I don't know how > d/cargo-vendor-pack.py should be fixed. Help would be appreciated. > > Josh, is this anything related to how your dh-cargo works? Any code that we > can reuse? Not directly, until we're ready to package all the individual crate dependencies (which we'll want to do at some point). But you can definitely use the same mechanism; it'd be much easier to generate a directory registry than the full Cargo index format. Rather than generating an index file and repo, just: 1) Copy every crate directory, verbatim, into a subdirectory of a vendor directory. 2) Create a file .cargo-checksum.json in the crate directory, containing {"package":"$SHA256","files":{}} , where $SHA256 is the sha256 of the .crate file. 3) Create a directory containing a file "config", containing: [source.crates-io] replace-with = "my-registry" [source.my-registry] directory = "/path/to/the/vendor/directory" 4) Set CARGO_HOME to the directory containing that config file, and build. debcargo can help a bit with this. Run it on a crate, move debian/cargo-checksum.json to .cargo-checksum.json , aggregate debian/copyright, remove the rest of debian/ , and put the resulting directory into the registry directory. ...and looking at git from 20 minutes ago, it looks like you've switched over to directory registries now.
Bug#852451: ITP: rname -- invoke a program under a different name
On Tue, 24 Jan 2017 17:19:49 +0200 Peter Pentchev <r...@ringlet.net> wrote: > The rname utility invokes a specified program, passing a different name > instead of the name of the program executable. This could be useful in > a number of cases, both during software development, testing, and in > production use. There are many programs that do different things based > on what name they have been invoked under; the rname utility avoids > the need to e.g. create ephemeral symlinks to run these programs in > some conditions when they have not been completely installed. > > I originally wrote this tool in 2000 and I'm resurrecting it now for > the purpose of writing unit and integration tests for just such > a multifaceted program. What kind of test environment do you plan to use rname from? Not trying to suggest that you shouldn't package it, but I wanted to call attention to a little-known feature of the bash exec builtin: ~$ (exec sh -c 'echo $0') sh ~$ (exec -a differentname sh -c 'echo $0') differentname That unfortunately doesn't work in portable sh, though; rname seems like an appropriate option if you can more easily require rname than requiring bash (or a better scripting/testing language). - Josh Triplett
Bug#504266: closing RFP: darwin-calendar-server -- CalDAV calendar server
On Mon, Sep 12, 2016 at 10:01:57PM -0700, Josh Triplett wrote: > On Mon, Sep 12, 2016 at 10:20:15AM +, Bart Martens wrote: > > RFP 504266 has no visible progress for a long time, so closing. > > It looks like the software got packaged (as "calendarserver") back in > 2011, just without an update to the RFP. Correction, packaged back in 2008.
Bug#504266: closing RFP: darwin-calendar-server -- CalDAV calendar server
On Mon, Sep 12, 2016 at 10:20:15AM +, Bart Martens wrote: > RFP 504266 has no visible progress for a long time, so closing. It looks like the software got packaged (as "calendarserver") back in 2011, just without an update to the RFP.
Bug#829007: ITP: gnome-shell-extension-move-clock -- move clock extension for GNOME shell
On Wed, 29 Jun 2016 20:07:22 +0200 "Jonathan Carter (highvoltage)"wrote: > * Package name: gnome-shell-extension-move-clock > Version : 1.0.0 > Upstream Author : 2011-2013 Ron Yorston > 2016 Jonathan Carter > * URL : > https://git.bluemosh.com/highvoltage/gnome-shell-extension-move-clock > * License : GPL-2+ > Programming Lang: JavaScript > Description : move clock extension for GNOME shell > > This extension will, when enabled, move the clock on your GNOME shell panel > from the middel to the right. s/middel/middle/ Also, when you CC debian-devel on ITPs, please use X-Debbugs-CC, so that the BTS sends debian-devel a version of the bug with a bug number included.
Bug#827843: ITP: xlogo -- XLogo is an interpreter for the Logo programming language, written in Java.
On Tue, 21 Jun 2016 18:18:09 +0200 Wolf Bergenheimwrote: > * Package name: xlogo Note that x11-apps ships /usr/bin/xlogo (a tiny toy app that displays the X Window System logo), so you'll want to avoid a binary name conflict.
Bug#823672: Streaming SIMD Extensions (SSE) is an SIMD instruction set extension to the x86 architecture
Wookey <woo...@wookware.org> wrote: > +++ Christian Seiler [2016-05-07 16:14 +0200]: > > On 05/07/2016 03:59 PM, Geert Stappers wrote: > > > I now have a better idea _why_ a sse-suport package. > > I do think that this sort of ISA-level checking would be best done via > dpkg and package metadata, although this sse-support mechanism will > obviously work. i.e the cpontrol file says what ISA features are > wanted and dpkg can not install it if those HWCAPS are not available > (or indeed install alternate versions that will work if they are > available). > > We worked up a prototype spec for this a couple of years ago (at the > bootstrap sprint), but it needs the namespace and granularity fleshing > out: what is the set of HWCAPS listed in each package, what is > presumed to be 'base' that we don't list individually, and so on. I'd love to see this integrated with dpkg multiarch support. That way, for instance, specific packages could provide optimized versions for specific target CPU features. I think we'd need three things to make that happen: 1) Support in dpkg multiarch to understand "compatible" architectures that can satisfy each others' dependencies (e.g. a dependency on libfoo1:amd64 is satisfied by libfoo1:amd64+avx2, and vice versa). 2) Support for synthesizing such architectures in a canonical way from a set of CPU features, so that every combination of CPU features doesn't need its own unique architecture name (we don't want architecture names like "amd64+sse4+avx2+rdrand", even though a package might need that set of features). 3) Support for detecting available CPU features, as you mentioned above, to not even attempt to install a package that requires unavailable CPU features. Optionally, we could also detect if the set of CPU features has changed incompatibly. For instance, if you move your Debian filesystem to a system without some features that you have installed packages depending on, it'd be nice to handle that as gracefully as possible. - Josh Triplett
Bug#821318: RFP: vim-toml -- Vim filetype and syntax plugin for TOML files
Package: wnpp Severity: wishlist * Package name: vim-toml * URL : https://github.com/cespare/vim-toml * License : MIT Description : Vim filetype and syntax plugin for TOML files This package provides vim support for editing TOML files, including filetype detection and syntax highlighting. CCing pkg-rust-maintainers, since Rust's Cargo ecosystem uses TOML.
Bug#748321: ITP: raceintospace -- free software version of the Liftoff! board game
On Sun, 14 Sep 2014 09:49:27 +0200 Dariusz Dwornikowski dariusz.dwornikow...@cs.put.poznan.pl wrote: Dariusz Dwornikowski dariusz.dwornikow...@cs.put.poznan.pl writes: Since the package depends on physfs 2.1, which has not been release yet, we need to wait for the upstream of physfs to release 2.1 and for the package to be in Debian. For now, raceintospace is ready in pkg-games git and can go to Debian when physfs is ready. Thanks for your efforts in packaging raceintospace. I've noticed that Hedgewars used to have a similar problem, which they've solved by adding a compatibility layer for PhysFS 2.0. Using this as a starting point, I saw that only one more function had to be provided, which I backported from the 2.1 branch with some minor modifications. With the attached patch, raceintospace compiles and runs with the PhysFS library from sid. The compatibility layer checks for the version number of PhysFS, so once 2.1 is packaged, the code should automatically stop being included. Thanks for that Hendrik, I will test it and preppare a release if it is ok. The only slight drawback is that Hedgewars is GPL-2 only, so it will no longer be possible to ship raceintospace binaries as GPL-2+. But since this is only a temporary situation, I think this can be justified. Yeah I think so. On the other hand I can see that physfs is not very actively developed so we could wait a little bit. It's been 9 months, and physfs 2.1 still doesn't seem to be available in either sid or experimental. This patch seems quite reasonable; is someone still willing to upload raceintospace with that patch? - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150630004357.GA15976@jtriplet-mobl1
Bug#781984: ITP: libnacl -- ctypes wrapper for libsodium
On Sun, 05 Apr 2015 22:12:02 +0200 David Douard david.dou...@logilab.fr wrote: Package: wnpp Severity: wishlist Owner: David Douard david.dou...@logilab.fr * Package name: libnacl Version : 1.4.2 Upstream Author : Thomas S Hatch tha...@saltstack.com * URL : https://libnacl.readthedocs.org/ * License : Apache 2.0 Programming Lang: Python Description : ctypes wrapper for libsodium This library is used to gain direct access to the functions exposed by Daniel J. Bernstein's nacl library via libsodium or tweetnacl. It has been constructed to maintain extensive documentation on how to use nacl as well as being completely portable. This package is a dependency for raet, the new transport layer for salt. Would you consider including a brief note in the final package description saying This library is unrelated to Native Client (NaCl), the sandbox used in Chromium.? I've seen that particular confusion arise several times, particularly since people often want a library version of the Chromium NaCl sandbox. Thanks, Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150406213720.GA8782@jtriplet-mobl1
Bug#779232: RFP: screen-session -- Session management and other tools for GNU Screen
Package: wnpp Severity: wishlist * Package name: screen-session Version : (git) Upstream Author : Artur Skonecki skon...@gmail.com * URL : https://skoneka.github.io/screen-session/index.html * License : GPLv3 Programming Lang: Python Description : Session management and other tools for GNU Screen screen-session (scs) includes various commands to manage GNU Screen sessions, including saving and restoring sessions, and managing layouts. Personally, I'm primarily interested in this for the scs new-window command, which uses /proc to figure out the working directory of the current screen tab and opens a new window in that directory. If screen knew how to do that natively, I wouldn't need screen-session. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150225184437.3607.15632.reportbug@jtriplet-mobl1
Bug#570611: Bug#669920: Please provide a backport for squeeze
On Sun, Apr 22, 2012 at 12:49:24AM -0400, Andres Mejia wrote: On Sat, Apr 21, 2012 at 8:46 PM, Josh Triplett j...@joshtriplett.org wrote: On Sat, Apr 21, 2012 at 06:53:57PM -0400, Andres Mejia wrote: On Sat, Apr 21, 2012 at 6:10 PM, Josh Triplett j...@joshtriplett.org wrote: Package: libcrystalhd-dev Severity: wishlist Please consider providing a backport of libcrystalhd-dev for squeeze. Such a backport would make it easier to prepare MythTV packages for squeeze. Thanks, Josh Triplett -- System Information: Debian Release: wheezy/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-2-amd64 (SMP w/4 CPU cores) Locale: LANG=C.UTF-8, LC_CTYPE=C.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Hi Josh, Were you trying to package MythTV to unstable at all? IMO that should be done first before attempting to upload for squeeze (i.e. actually get accepted, get tested in unstable, migrate to testing, then backport to stable). At the moment I'm just building some local packages, not official backports. I'd love to see official MythTV packages, and I'd be happy to help work on them, but that wasn't my goal at the moment. - Josh Triplett I recommend starting off by porting the Ubuntu packaging of MythTV to Debian. I actually started from the debian-multimedia packaging, and hacked out all the bits that depended on packages outside of Debian, as well as dependency cruft that applied to previous versions but not to the current version. I've managed to successfully build working squeeze packages at this point, which only depend on packages in Debian to build and run; installing them requires a couple of packages from squeeze-backports (fonts-droid, libmp3lame0, libx264-120, libxvidcore4), and a locally built backport of libcrystalhd3, and all the rest of the dependencies come from squeeze. I presume if these are local packages, you're using reprepro or something similar to host a local archive. In such a case, you may as well build packages of libcrystalhd and upload them to your local archive. Then you can build mythtv. I did end up building local packages of crystalhd and libcec, which proved relatively straightforward. Note that mythtv is using it's own copy of ffmpeg. It needs to be modified so it can build/run with system libav before it can be uploaded to Debian. If that's something you're interested in doing, by all means, feel free to modify mythtv source tree, and preferably work with mythtv upstream to pass along patches. Yeah, I do know about that problem. Arguably not an issue for an upload to Debian experimental; only an issue for having a version in a security-supportable release. Considering that upstream went out of their way to remove support for system libav/ffmpeg (due to issues with different versions on end-user systems, particularly versions of the command-line tools), I suspect they might not like the idea of adding it back, or for that matter the idea of adding support for using avconv rather than ffmpeg. This needs a long conversation with upstream before attempting to write a patch; no sense writing a patch that'll never make it upstream. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120422093318.GA4379@leaf
Bug#570611: Bug#669920: Please provide a backport for squeeze
On Sat, Apr 21, 2012 at 06:53:57PM -0400, Andres Mejia wrote: On Sat, Apr 21, 2012 at 6:10 PM, Josh Triplett j...@joshtriplett.org wrote: Package: libcrystalhd-dev Severity: wishlist Please consider providing a backport of libcrystalhd-dev for squeeze. Such a backport would make it easier to prepare MythTV packages for squeeze. Thanks, Josh Triplett -- System Information: Debian Release: wheezy/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-2-amd64 (SMP w/4 CPU cores) Locale: LANG=C.UTF-8, LC_CTYPE=C.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Hi Josh, Were you trying to package MythTV to unstable at all? IMO that should be done first before attempting to upload for squeeze (i.e. actually get accepted, get tested in unstable, migrate to testing, then backport to stable). At the moment I'm just building some local packages, not official backports. I'd love to see official MythTV packages, and I'd be happy to help work on them, but that wasn't my goal at the moment. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120422004617.GA30987@leaf
Bug#642288: RFP: cabal-dev -- Manage sandboxed Haskell build environments
Package: wnpp Severity: wishlist * Package name: cabal-dev Version : 0.8 Upstream Author : Josh Hoyt j...@galois.com, Jonathan Daughtery j...@galois.com, Rogan Creswick cresw...@galois.com * URL : http://hackage.haskell.org/package/cabal-dev * License : BSD (3-clause) Programming Lang: Haskell Description : Manage sandboxed Haskell build environments cabal-dev is a tool for managing development builds of Haskell projects. It supports maintaining sandboxed cabal-install repositories, and sandboxed ghc package databases. For most packages, just use cabal-dev instead of cabal, and you will get a sandboxed build that will not install anything (even automatically installed dependencies) into the user or global ghc package databases. You can also add local sources for patched or unreleased libraries. -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110921064611.2421.86209.reportbug@leaf
Bug#637471: RFP: tj3 -- TaskJuggler 3, a project management tool with resource constraint solving
Package: wnpp Severity: wishlist * Package name: tj3 Version : 0.2.1 Upstream Author : Chris Schläger ch...@linux.com * URL : http://taskjuggler.org/ * License : GPLv2 Programming Lang: Ruby Description : TaskJuggler 3, a project management tool with resource constraint scheduling TaskJuggler 3 provides project management tools built around a text-based project description format. It supports project scoping, task tracking, resource assignment, and cost and revenue planing. TaskJuggler includes a resource constraint scheduler, which can automatically assign resources to tasks, and prevent conflicts between tasks that require the same resource. TaskJuggler compiles project descriptions, computes and produces reports, including task lists, schedules, timesheets, Gantt charts, iCal, and CSV. [ Note that this package explicitly needs to refer to TaskJuggler 3, to distinguish it from the obsolete TaskJuggler 2 for KDE3/Qt3. Also note that this package includes a vim mode, which will need installing in the appropriate path, and configuring for automatic use on ..tjp and .tji files. ] -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110811204319.16333.7386.reportbug@leaf
Bug#482938: Confirming
Just to confirm, I stopped using foxmarks a long time ago when it stopped using a FOSS license, and in any case Firefox Sync has made it entirely obsolete. So I definitely no longer care about this RFP. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110731054204.GA4378@leaf
Bug#570611: LAME now in Debian
As of the recent clearing of the NEW queue (http://blog.schmehl.info/Debian/NEW-empty), it looks like LAME and x264 both made it into Debian. That should make MythTV packaging much easier. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110729220331.GA5909@leaf
Bug#574800: nss-myhostname packaged for Debian
On Fri, May 07, 2010 at 10:03:36PM +0200, Joachim Breitner wrote: after being kindly asked to do it, I packaged nss-myhostname for Debian and uploaded it. It will have to pass through the NEW queue and will then be installable via apt-get on Debian. Thanks for creating this nice piece of software. I'll second the thanks for creating it, and add my thanks to Joachim for packaging it. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100508035547.ga11...@feather
Bug#574800: RFP: libnss-myhostname -- nss module to resolve the system hostname
Package: wnpp Severity: wishlist * Package name: libnss-myhostname Version : 0.2 Upstream Author : Lennart Poettering mzzlubfga...@0pointer.de * URL : http://0pointer.de/lennart/projects/nss-myhostname/ * License : LGPLv2.1 Programming Lang: C Description : nss module that resolves the system hostname instead of /etc/hosts nss-myhostname is a plugin for the GNU Name Service Switch (NSS) functionality of the GNU C Library (glibc) providing host name resolution for the locally configured system hostname as returned by gethostname(2). A lot of software relies on resolving the local host name via DNS to an IPv4 or IPv6 address. Most systems resolve the system hostname via /etc/hosts, which either duplicates /etc/hostname or requires dynamically editing /etc/hosts. nss-myhostname automatically resolves the current system hostname to the IPv4 address 127.0.0.2 (which is on the local loopback) and the IPv6 address ::1. -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100321054733.2727.47394.report...@feather
Bug#556135: RFP: xul-ext-weave -- Syncronize personal data between Mozilla browsers
Package: wnpp Severity: wishlist * Package name: xul-ext-weave Version : 0.8 Upstream Author : Mozilla * URL : https://mozillalabs.com/weave/ * License : Standard Mozilla tri-license: MPL, GPLv2+, LGPLv2.1+ Programming Lang: XUL, JavaScript, C++ Description : Syncronize personal data between Mozilla browsers Mozilla Weave securely shares and synchronizes personal data between Mozilla browsers running on different systems. In particular, Weave can synchronize bookmarks, history, passwords, open tabs, and user preferences. Weave uses encryption to keep data secure on the server, decrypting it using a passphrase. -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#551955: RFP: ttf-anonymous-pro -- A fixed-width sans-serif font designed especially for coders
Package: wnpp Severity: wishlist * Package name: ttf-anonymous-pro Version : 1.001 Upstream Author : Mark Simonson m...@ms-studio.com * URL : http://www.ms-studio.com/FontSales/anonymouspro.html * License : SIL Open Font License Description : A fixed-width sans-serif font designed especially for coders Anonymous Pro is a family of four fixed-width fonts designed especially with coding in mind. Characters that could be mistaken for one another (O, 0, I, l, 1, etc.) have distinct shapes to make them easier to tell apart in the context of source code. Anonymous Pro also features an international, Unicode-based character set, with support for most Western and European Latin-based languages, Greek, and Cyrillic. It also includes special box drawing characters for those who need them. /long-description I spoke with Mark Simonson previously about releasing his freeware font Anonymous Pro as FOSS. He had apparently considered the idea before that as well. I just discovered that Mark generously released Anonymous Pro under the Open Font License recently. To encourage widespread usage and distribution of this excellent font, I've filed this RFP requesting a package for Debian. Thank you, Mark! And thank you to whoever ends up packaging this for Debian. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#551955: RFP: ttf-anonymous-pro -- A fixed-width sans-serif font designed especially for coders
A followup note about packaging this font: Anonymous Pro includes embedded bitmaps for several sizes, so the font packaging should ideally include fontconfig rules that enable the embedded bitmaps as appropriate. - Josh Triplett -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#509850: RFP: haskell-curl -- Haskell bindings to the libcurl URL transfer library
Package: wnpp Severity: wishlist * Package name: haskell-curl Version : 1.3.3 Upstream Author : Sigbjorn Finne * URL : http://hackage.haskell.org/cgi-bin/hackage-scripts/package/curl * License : BSD (3-clause) Programming Lang: Haskell Description : Haskell bindings to the libcurl URL transfer library This package provides the Haskell Network.Curl bindings to the libcurl client-side URL transfer library. - Josh Triplett -- System Information: Debian Release: 5.0 APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#509182: RFP: haskell-fastcgi -- Haskell interface to FastCGI
Package: wnpp Severity: wishlist * Package name: haskell-fastcgi Version : 3001.0.2.2 Upstream Author : Bjorn Bringert bj...@bringert.net * URL : http://hackage.haskell.org/cgi-bin/hackage-scripts/package/fastcgi * License : 3-clause BSD Programming Lang: Haskell Description : Haskell interface to FastCGI This package provides the Haskell module Network.FastCGI, a Haskell interface to FastCGI. FastCGI provides an interface between web servers and CGI programs that runs the CGI program as a long-running server which services many requests. Network.FastCGI provides an interface almost identical to Network.CGI, making it easy to port existing Haskell CGI programs to FastCGI. -- To UNSUBSCRIBE, email to debian-wnpp-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#504266: RFP: darwin-calendar-server -- CalDAV calendar server
Package: wnpp Severity: wishlist * Package name: darwin-calendar-server Version : 1.3 Upstream Author : Apple * URL : http://calendarserver.org/ * License : Apache License 2.0 Programming Lang: Python Description : CalDAV calendar server Darwin Calendar Server provides a collaborative shared calendar using the standard CalDAV protocol. Many calendaring programs support synchronizing their calendar with a CalDAV server. Note that Darwin Calendar Server uses filesystem extended attributes (xattrs), so you must enable xattrs on the filesystem used to store the calendar data. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#482938: RFP: foxmarks -- Synchronize bookmarks between multiple systems running Firefox or Iceweasel
Package: wnpp Severity: wishlist * Package name: foxmarks Version : Stable 2.0.45, beta 2.0.44.20 Upstream Author : Foxmarks Inc. * URL : http://foxmarks.com/ * License : Apache License, Version 2.0 Programming Lang: XUL and JavaScript Description : Synchronize bookmarks between multiple computers running Firefox or Iceweasel Foxmarks synchronizes bookmarks between multiple computers running Firefox or Iceweasel. It stores bookmarks via your account on the Foxmarks server, also accessible via the web at http://my.foxmarks.com. It can also store bookmarks on any server running WebDAV. The stable version 2.0.45 seems like the safest choice to package. Foxmarks has a version for Firefox 3 in beta, and packaging that for experimental might make sense, but it seems easier to just wait for the stable version to support Firefox 3, which it almost certainly will by the time the stable release of Firefox 3 occurs. The binary package should have the name iceweasel-foxmarks or mozilla-foxmarks; probably the former. - Josh Triplett -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.25-2-686 (SMP w/1 CPU core) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#470369: RFP: haskell-tagsoup -- Haskell library to robustly parse non-standards-compliant HTML
Package: wnpp Severity: wishlist * Package name: haskell-tagsoup Version : 0.4 Upstream Author : Neil Mitchell [EMAIL PROTECTED] * URL : http://www-users.cs.york.ac.uk/~ndm/tagsoup/ * License : 3-clause BSD Programming Lang: Haskell Description : Haskell library to robustly parse unstructured HTML TagSoup extracts information out of unstructured HTML code, sometimes known as tag soup. TagSoup does not require well-formed or standards-compliant HTML, or HTML that renders correctly in any particular rendering engine. TagSoup transforms HTML into a list of open tags with attributes, close tags, and text, but makes no attempt to group these together into any kind of structure. -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.24-1-686 (SMP w/1 CPU core) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#456165: RFP: handbrake -- Rips and encodes DVDs
Package: wnpp Severity: wishlist * Package name: handbrake Version : 0.9.1 Upstream Author : Various; see http://handbrake.m0k.org/?article=development * URL : http://handbrake.m0k.org/ * License : GPL Programming Lang: C Description : Rips and encodes DVD videos Handbrake provides a friendly graphical interface for ripping DVD videos and encoding them as video files. It can show a preview of content before ripping, to verify titles and chapters. It can deinterlace, crop, and scale video. It supports DVD subtitles by encoding them directly in the video image. .. Handbrake supports numerous formats and codecs: - File format: MPEG-4 (MP4), Matroska (MKV), Ogg (OGM), or AVI - Video: MPEG-4 or H.264 (1 or 2 passes or constant quantizer/rate encoding) - Audio: AAC, MP3, Vorbis or AC-3 pass-through (supports encoding of several audio tracks) .. Handbrake also provides a command-line interface, HandbrakeCLI. Some notes on Handbrake: * Its build system needs some serious work. It wants to download and build all its dependencies at build time, and statically link to those copies. (I have not actually managed to get it to build, due to build errors in one of these libraries.) This will need to change to support dynamically linking to system libraries. * Some of the libraries it depends on do not exist in Debian due to legal problems such as patent issues or anti-circumvention laws. For libdvdcss, ideally it could do the same thing all the DVD players in Debian do: use libdvdread (which Handbrake already does), and let libdvdread pull in libdvdcss if installed. For encumbered codecs, ideally it could pick up libraries if installed and disable the corresponding codecs if not; however, if it cannot, then perhaps it could have a DEB_BUILD_OPTIONS=risky like ffmpeg. * Build-Depends available in Debian: jam, liba52-0.7.4-dev, libavcodec-dev, libdvdread3-dev, libmpeg2-4-dev, libogg-dev, libvorbis-dev, libsamplerate0-dev - Josh Triplett -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.22-rc1 (PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#452026: RFP: golly -- Game of Life simulator using hashlife algorithm
Package: wnpp Severity: wishlist * Package name: golly Version : 1.3 Upstream Author : Tomas Rokicki [EMAIL PROTECTED] Andrew Trevorrow [EMAIL PROTECTED] * URL : http://golly.sourceforge.net/ * License : GPL version 2 or later Programming Lang: C++, with Python scripting support Description : Game of Life simulator using hashlife algorithm Golly simulates Conway's Game of Life with an arbitrarily large grid of cells. It can optionally use a hashlife algorithm, which allows it to rapidly compute generations for huge patterns, and to compute many generations into the future at a time. . Golly provides a graphical interface for viewing and editing cellular automata. It supports copy and paste, zoom, auto-fit, multiple layers, and viewing different areas of a pattern simultaneously in different areas of a window. . Golly can load patterns from RLE, Life 1.05/1.06, dblife, and macrocell file formats; it can also interpret images as Life patterns. Golly provides integrated help, including a copy of the Life Lexicon. . Golly also supports other rules for 2D cellular automata with an 8-cell neighborhood, and supports 1D cellular automata. (end of description) Some notes: * Golly has some serious portability issues: in particular, it generates many compiler errors if sizeof(node *) != sizeof(int). Much of this seems to occur in the hash table code when hashing pointer values, so in theory it might not cause a problem if made to go away via appropriate casting. * Golly uses wxWidgets, and supports building for either X11 or GTK+; however, the authors specifically recommend against the X11 version, so I'd suggest only packaging the GTK+ version. * Golly runs extremely fast, and the hashlife algorithm makes many complex patterns easy to simulate. On my system, I could easily simulate a 3x3 grid of Unit Life cells running a blinker, with one Unit Life generation per step, in real-time. I could also take any of the patterns from Golly's database and run it to 1e10 generations with no problems, even those with asymptotically quadratic growth. * The interface beats xlife hands down. - Josh Triplett -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#396648: ITP: rt73 -- Linux device driver for Ralink RT73 a/b/g WLAN Card
Piotr Roszatycki wrote: The driver loads firmware from separate file and has included the blob with the default firmware if separate file is not available. I think the driver can go to contrib with removed the blob from its sources Yes. and with the firmware rt73.bin file can go to non-free as rt73-firmware package. The rt73.bin file is marked as GPL. There is no real source code included but this license implies that it is redistributable file at least as non-free. No. Distribution of a GPLed binary requires provision of the preferred form for modification; if Debian does not have that, Debian cannot distribute the firmware at all. Also, does the binary have an explicit GPL license attached (some file specifically says that the firmware falls under the GPL, and/or the version in the driver source has a GPL header above the firmware binary), or does it just reside in the driver next to all the GPLed source and a copy of the GPL? If the latter, it may not actually fall under the GPL, and you should request a specific license for it. (Whether or not upstream intended it as GPLed does not affect distributability, but it does affect how you approach upstream; in one case you just need to ask about a license for the firmware, and in the other case you need to explain the issues about GPLing a binary. Also, as usual, throwing in a request for the source as a lark couldn't hurt; worst case, they say no, no harm done.) One other question: do all variants of this device require firmware from the driver, or only some of them? (Just asking because several other ralink cards, such as rt2500, do not seem to require the driver to supply firmware.) If some variants of the device will work without supplied firmware, then the driver can go to main, regardless of where or if the firmware gets distributed. - Josh Triplett signature.asc Description: OpenPGP digital signature
Bug#358558: byzanz packaging
I also have Debian packaging for byzanz, available at http://minilop.net/~josh/byzanz/ . Among other things, it includes a manpage I wrote for the byzanz-record command-line utility, and a debian/watch file. Would you have any interest in co-maintaining Debian packaging for byzanz? CCed to Daniel Holbach since he may want to include the manpage in the Ubuntu package. I've already emailed the manpage upstream, along with two bugfixes: a typo in configure.ac, and the fact that COPYING contains the GPL but the source files indicate LGPL licensing. - Josh Triplett signature.asc Description: OpenPGP digital signature
Bug#284190: ITP: drdsl -- DSL Assistant for AVM DSL/ISDN-Controllers
Matthias Klose wrote: Josh Triplett writes: Matthias Klose wrote: CC'ing debian-legal, please could you have a look at the license? The question being is this acceptable to go into non-free? exactly. OK. I asked to clarify the paragraph, the current text now has append and the terms of the LGPL. The Proprietary Source Software, which is delivered in object code format only, such as the .o files, shall in no event be disassembled, reverse engineered, decompiled or otherwise be opened and the results realised insofar shall not be copied or distributed without the prior written approval of AVM except to the extent as may be expressly authorized under mandatory law and the terms of the LGPL. With this addition, the software can be distributed in non-free. - Josh Triplett signature.asc Description: OpenPGP digital signature
Bug#284190: ITP: drdsl -- DSL Assistant for AVM DSL/ISDN-Controllers
Matthias Klose wrote: CC'ing debian-legal, please could you have a look at the license? The question being is this acceptable to go into non-free? [...] The available AVM driver package consists of two portions, namely an Open Source Software portion and a Proprietary Source Software portion. The Proprietary Source Software portion is delivered in object code format only and includes i.e. the lib.o files which, again, include libraries as well as specific portions of the driver. The Open Source Software portion is licensed under the terms and conditions of the GNU Lesser GPL (LGPL). Please be aware of the requirements of the LGPL. You may easily download the terms of the LGPL as follows: http://www.gnu.org/copyleft/lesser.html. On Debian GNU/Linux systems, the complete text of the GNU Lesser General Public License can be found in `/usr/share/common-licenses/LGPL'. In general, you may distribute both portions of the available AVM driver package, i.e. on a distribution CD delivered in conjunction with your products. In addition to the terms of the LGPL and under all AVM intellectual property and proprietary rights, AVM grants you the worldwide, non-exclusive and royalty-free rights (1) to use and copy (2) to make (and have made), use, import, sell, offer for sale or otherwise distribute any of your (legal) products or services containing (portions of) the AVM driver package, and (3) to sublicense rights to the extend a license is necessary for using your products or services. Up to this point, the license seems acceptable for non-free; it seems to permit redistribution of the binary-only portion, and the LGPLed portion is of course acceptable. However: The Proprietary Source Software, which is delivered in object code format only, such as the .o files, shall in no event be disassembled, reverse engineered, decompiled or otherwise be opened and the results realised insofar shall not be copied or distributed without the prior written approval of AVM except to the extent as may be expressly authorized under mandatory law. This condition is incompatible with the GNU LGPL. Clause 6 of the LGPL states in part: 6. As an exception to the Sections above, you may also combine or link a work that uses the Library with the Library to produce a work containing portions of the Library, and distribute that work under terms of your choice, provided that the terms permit modification of the work for the customer's own use and reverse engineering for debugging such modifications. It goes on to state that distributing object code for relinking is acceptable, as AVM does. However, the license on the work as a whole, including the .o files, does not permit modification of the work for the customer's own use and reverse engineering for debugging such modifications. The result is that neither the .o files which link to the library nor any compiled binary from both the .o files and the library can be distributed. It is understood that you will be responsible/liable for the software which you offer/distribute/make available in conjunction with or which you combine with (portions of) the AVM driver package. For instance, we refer to the implications mentioned in the LGPL in case of a breach (your rights granted under LGPL would terminate automatically, see Art. 4 LGPL). You should state clearly that you offer any necessary support on your own. This clause doesn't look like a problem. Summary: non-distributable, because the license on the proprietary portion does not permit modification of the work for the customer's own use and reverse engineering for debugging such modifications, which is incompatible with the LGPLed portion. - Josh Triplett signature.asc Description: OpenPGP digital signature
Bug#278940: ITP: socket++ -- lightweight convenience library to handle low level BSD sockets in C++
Dan Weber wrote: * Package name: socket++ Version : 1.12.12 Upstream Author : Gnanasekaran(Sekar) Swaminathan [EMAIL PROTECTED], Herbert Straub [EMAIL PROTECTED] * URL : http://www.linuxhacker.at/socketxx * License : Pasted below with License Statement Description : lightweight convenience library to handle low level BSD sockets in C++ [snip long description] -- License Original Copyright Notice: Permission is granted to use at your own risk and distribute this software in source and binary forms provided the above copyright notice and this paragraph are preserved on all copies. This software is provided as is with no express or implied warranty. This license alone would not be sufficient, since it does not allow modification. However... Copyright Statement: Got your message. Please feel free to include it in any software and use it as you please. Let me know if you need any help with it. and further: +++ Hi Herbert, That was not the intention. It is a free code. You can modify, copy, and distribute and use it in anyway as you see fit. Other people are maintaining it and ported it to different OSes. Other than that, not much has changed as far as I know. I haven't looked at it recently. -Sekar ... this clarification makes the software clearly DFSG-Free (though it would be preferable if this mail explicitly referred to socket++ and not just it :) ). -- Side Note This ITP is being cross posted to debian-legal for the purpose of verifying that this is capable of being placed in main. Definitely. Be sure to include the clarification in the copyight file along with the original license. - Josh Triplett signature.asc Description: OpenPGP digital signature
Bug#251885: ITP: cgal -- C++ library for computational geometry
Joachim Reichel wrote: If the two licenses only applied to different libraries that linked to each other, they would be compatible, because the scope of the LGPL deliberately stops at the library boundary. However, the LGPL requires that all code which directly incorporates LGPLed code be LGPLed. [...] So any work based on the library must be licensed under the LGPL. Since the QPL is not compatible with the LGPL, the entire work is non-distributable. So what are the options for upstream? a) Relicense the code in a way that the work is redistributable, if possible DFSG-free. That would work. Note that if upstream likes the QPL, they would not need to give it up completely; they could choose to dual-license the work under the LGPL and the QPL, or under the GPL and the QPL, or under any other Free license and the QPL. As long as one of the license options is a Free Software license. b) Split the library in (at least) two libraries, one (or more) for the LGPL code and one (or more) for the QPL code. (By splitting, I mean creating two (or more) static/dynmaic libraries, not splitting the package itself.) That would work, but the resulting package would go into non-free, since it uses the QPL. c) What about clause 7 of the LGPL: 7. You may place library facilities that are a work based on the Library side-by-side in a single library together with other library facilities not covered by this License, and distribute such a combined library, provided that the separate distribution of the work based on the Library and of the other library facilities is otherwise permitted, and provided that you do these two things: a) Accompany the combined library with a copy of the same work based on the Library, uncombined with any other library facilities. This must be distributed under the terms of the Sections above. I think this would mean accompany in the same package, so this is basically the same as the previous option, except that you could also provide a combined library. Again, the resulting package would go into non-free. b) Give prominent notice with the combined library of the fact that part of it is a work based on the Library, and explaining where to find the accompanying uncombined form of the same work. For clarification, as I understand it: Library = LGPL-part of the code library = the dynamic/static library (including QPL code) Would this be another option? Is there a piece of software that actually uses this clause? d) What about an exception with respect to clause 2c) of the LGPL? Something similar like the GPL-waiver for libssl? This would also be an option, right? Assuming the authors hold the copyright to the entire work (meaning that they didn't incorporate any other LGPLed code), that would work. See http://www.gnu.org/licenses/gpl-faq.html#GPLIncompatibleLibs for details on the exception, and modify it to work with the LGPL instead of the GPL. Again, this option would only let the package go into non-free. - Josh Triplett
Bug#251885: ITP: cgal -- C++ library for computational geometry
Joachim Reichel wrote: License: The library consists of three modules. The lower layers (Kernel and the Support library) are licensed under LGPL, the upper layer (Basic Library) is licensed under QPL. Code under LGPL and code under QPL is combined in one library. I've CC'ed debian-legal and I would like to know whether both licenses are compatible. If the two licenses only applied to different libraries that linked to each other, they would be compatible, because the scope of the LGPL deliberately stops at the library boundary. However, the LGPL requires that all code which directly incorporates LGPLed code be LGPLed. From the LGPL, version 2.1: The precise terms and conditions for copying, distribution and modification follow. Pay close attention to the difference between a work based on the library and a work that uses the library. The former contains code derived from the library, whereas the latter must be combined with the library in order to run. [...] The Library, below, refers to any such software library or work which has been distributed under these terms. A work based on the Library means either the Library or any derivative work under copyright law: that is to say, a work containing the Library or a portion of it, either verbatim or with modifications and/or translated straightforwardly into another language. (Hereinafter, translation is included without limitation in the term modification.) [...] 2. You may modify your copy or copies of the Library or any portion of it, thus forming a work based on the Library, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: a) The modified work must itself be a software library. b) You must cause the files modified to carry prominent notices stating that you changed the files and the date of any change. c) You must cause the whole of the work to be licensed at no charge to all third parties under the terms of this License. [...] So any work based on the library must be licensed under the LGPL. Since the QPL is not compatible with the LGPL, the entire work is non-distributable. The other issue here is that the QPL is not a Free Software license at all. See the thread starting at http://lists.debian.org/debian-legal/2004/04/msg00233.html for details. The QPL requires that all changes are sent to the original author upon request, and that all license disputes are settled in Amsterdam City Court by the laws of the Netherlands. Both of these restrictions are non-DFSG-free. I would suggest asking the authors to dual-license their library under the QPL and the GPL, like Trolltech did with Qt. In addition to solving the freeness and compatibility problems, this would also makes the library GPL-compatible, which allows the huge number of GPLed programs to link with the library. (Incidentally, the libcwd package which raised the issue of the QPL seems to have been uploaded to main even after that discussion concluded that the license was non-free.) - Josh Triplett