Re: http://bugs.debian.org/release-critical has terrible handling of bugs filed against source packages
Hi Peter, Don Armstrong wrote: > On Thu, 19 Jan 2012, Jonathan Nieder wrote: >> peter green wrote: >>> Of the pages linked from http://bugs.debian.org/release-critical >> [...] >>> debian/all.html does not show bugs against source packages >>> debian/main.html does not show bugs against source packages >>> debian/source.html does not show ANYTHING AT ALL! >>> debian/contrib.html does not show any bugs against source packages >>> (I do not know whether it is failing to report them or whether they >>> don't exist) You can find the scripts that generate these pages here: http://bugs.debian.org/debbugs-source/bugscan/ (bzr) http://git.debian.org/?p=users/jrnieder-guest/debbugs/bugscan.git (gitweb) git://git.debian.org/~jrnieder-guest/debbugs/bugscan.git (git) The relevant script is "dohtml": makepage "debian" "$title" "$htmldir/debian/all.html" "$time" makepage "debian/main" "$title" $htmldir/debian/main.html "$time" makepage "debian/contrib" "$title" $htmldir/debian/contrib.html "$time" makepage "debian/non-free" "$title" $htmldir/debian/non-free.html "$time" makepage "debian/source" "$title" $htmldir/debian/source.html "$time" The makepage function runs the ./bugreport script, which uses the condition "$scanlib::section{$p} =~ m/^$filter/" to decide what page each bug belongs on. The readsources and readpackages functions in scanlib.pm look like this: for $sect (@bugcfg::sections) { open(P, "zcat $root/$sect/source/Sources.gz|") or die open "open: $sect sourcelist: $!\n"; while () { chomp; next unless m/^Package:\s/; s/^Package:\s*//; # Strip the fieldname $section{$_} = "$archive/$sect"; } close (P); } If we want $section{$p} to match "debian/source", that would presumably involve changing the $section{$_} = "$archive/$sect" line to something like $section{$_} = "$archive/source" However, that would lose the distinction between the main, contrib, and non-free archive sections, and it would clobber the section for the corresponding binary package when they share a name. In a way, we want the section for source packages to simultaneously be "$archive/$sect" and "$archive/source". Maybe something like the following hack would be a good starting point for investigating. diff --git a/scanlib.pm b/scanlib.pm index 60e4cf47..29bcd130 100644 --- a/scanlib.pm +++ b/scanlib.pm @@ -61,6 +61,7 @@ sub readsources() { next unless m/^Package:\s/; s/^Package:\s*//; # Strip the fieldname $section{$_} = "$archive/$sect"; + $section{"src:$_"} = "$archive/$sect"; } close (P); } -- To UNSUBSCRIBE, email to debian-www-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120121221754.GB3147@burratino
Bug#238245: license choice - consensus on dual MIT/GPL-2+ ?
On Sat, 21 Jan 2012 12:57:59 -0400 David Prévot wrote: > Le 21/01/2012 12:28, Francesco Poli a écrit : > > On Sat, 21 Jan 2012 11:08:55 -0400 David Prévot wrote: > > > I would use the classical Expat URL for the Expat/MIT license: > […] > > Moreover, as far as the Expat license is concerned, I would not talk > > about any "latest version", > > Thank you Francesco for your remarks, You're welcome! > attached patch and built page > updated accordingly: > > http://tilapin.org/debian/license.en It looks better, now. Thanks for your time! Bye. -- http://www.inventati.org/frx/frx-gpg-key-transition-2010.txt New GnuPG key, see the transition document! . Francesco Poli . GnuPG key fpr == CA01 1147 9CD2 EFDF FB82 3925 3E1C 27E1 1F69 BFFE pgpPeAnCpzLI4.pgp Description: PGP signature
Bug#238245: license choice - consensus on dual MIT/GPL-2+ ?
Le 21/01/2012 12:28, Francesco Poli a écrit : > On Sat, 21 Jan 2012 11:08:55 -0400 David Prévot wrote: > I would use the classical Expat URL for the Expat/MIT license: […] > Moreover, as far as the Expat license is concerned, I would not talk > about any "latest version", Thank you Francesco for your remarks, attached patch and built page updated accordingly: http://tilapin.org/debian/license.en Regards David Index: license.wml === RCS file: /cvs/webwml/webwml/english/license.wml,v retrieving revision 1.27 diff -u -r1.27 license.wml --- license.wml 8 Jan 2012 21:57:38 - 1.27 +++ license.wml 21 Jan 2012 16:55:40 - @@ -10,6 +10,22 @@ + +Since @@day@@ January 2012, the new material can be redistributed +and/or modified under the terms of the \ +MIT (Expat) License (which is usually available at +http://www.jclark.com/xml/copying.txt>) or, at your option, of the +GNU General Public License; either +version 2 of the License, or any later version (the latest version is +usually available at http://www.gnu.org/licenses/gpl.html>). + + + +Work is in progress to make the older material compliant +with the above licenses. Until then, please refer to the +following terms of the Open Publication License. + + This material may be distributed only subject to the terms and conditions set forth in the Open Publication License, Draft v1.0 or later (you can read our local copy, the latest version signature.asc Description: OpenPGP digital signature
Bug#238245: license choice - consensus on dual MIT/GPL-2+ ?
On Sat, 21 Jan 2012 11:08:55 -0400 David Prévot wrote: [...] > Le 20/01/2012 13:53, Francesco Poli a écrit : > > On Wed, 18 Jan 2012 23:51:55 +0100 Stefano Zacchiroli wrote: > >> On Wed, Jan 18, 2012 at 07:42:05PM +0100, Francesco Poli wrote: > > >>> If this is what you mean, then it should be noted that "dual-licensed > >>> under Expat/MIT and GPLv2+" is effectively equivalent to "licensed > >>> under the Expat/MIT", > > >> You're quite right (at least, under most interpretations of the two > >> licenses; cause with these things you really never know...). As in > >> other cases of dual MIT/GPL licensing, the point is being clear in the > >> fact that recipient can choose both > > > I think it would make a number of people (wrongly) think that the > > Debian Project decision-makers know very little about licenses... > > I take it as a remark, not as an objection, Well, I intended it to be a minor objection (thus a "non-blocking" objection), but anyway... > and thus propose the > attached patch if we agree on the dual licensing (@@date@@ will of > course be replaced once agreed on the license choice and its wording). > You can have a look at the built page on my test server: > > http://tilapin.org/debian/license.en I would use the classical Expat URL for the Expat/MIT license: http://www.jclark.com/xml/copying.txt rather than the one hosted by OSI. Moreover, as far as the Expat license is concerned, I would not talk about any "latest version", since the Expat license is not given any distinguishing version number: I would therefore just say "(which is usually available at http://www.jclark.com/xml/copying.txt)" The rest seems to be OK (apart from the very idea of the Expat/GPL dual-licensing, which I have already commented previously). > > If my understanding of dual licensing is not too defective, we will be > able to drop one of them later if we feel strongly about it. Therefore I > assume that Francesco's point can be raised later if it really matters, > and thus is not a blocker here and now. This seems to be true, even though such a strategy looks sub-optimal to me... -- http://www.inventati.org/frx/frx-gpg-key-transition-2010.txt New GnuPG key, see the transition document! . Francesco Poli . GnuPG key fpr == CA01 1147 9CD2 EFDF FB82 3925 3E1C 27E1 1F69 BFFE pgpi82d89m50R.pgp Description: PGP signature
Bug#238245: license choice - consensus on dual MIT/GPL-2+ ?
Hi, First of all, thanks Stefano to step in this long standing issue. Le 20/01/2012 13:53, Francesco Poli a écrit : > On Wed, 18 Jan 2012 23:51:55 +0100 Stefano Zacchiroli wrote: >> On Wed, Jan 18, 2012 at 07:42:05PM +0100, Francesco Poli wrote: >>> If this is what you mean, then it should be noted that "dual-licensed >>> under Expat/MIT and GPLv2+" is effectively equivalent to "licensed >>> under the Expat/MIT", >> You're quite right (at least, under most interpretations of the two >> licenses; cause with these things you really never know...). As in >> other cases of dual MIT/GPL licensing, the point is being clear in the >> fact that recipient can choose both > I think it would make a number of people (wrongly) think that the > Debian Project decision-makers know very little about licenses... I take it as a remark, not as an objection, and thus propose the attached patch if we agree on the dual licensing (@@date@@ will of course be replaced once agreed on the license choice and its wording). You can have a look at the built page on my test server: http://tilapin.org/debian/license.en If my understanding of dual licensing is not too defective, we will be able to drop one of them later if we feel strongly about it. Therefore I assume that Francesco's point can be raised later if it really matters, and thus is not a blocker here and now. Cheers David Index: license.wml === RCS file: /cvs/webwml/webwml/english/license.wml,v retrieving revision 1.27 diff -u -r1.27 license.wml --- license.wml 8 Jan 2012 21:57:38 - 1.27 +++ license.wml 21 Jan 2012 14:54:48 - @@ -10,6 +10,22 @@ + +Since @@day@@ January 2012, the new material can be redistributed +and/or modified under the terms of the \ +MIT (Expat) License (the latest version is usually available at +http://www.opensource.org/licenses/MIT>) or, at your option, of the +GNU General Public License; either +version 2 of the License, or any later version (the latest version is +usually available at http://www.gnu.org/licenses/gpl.html>). + + + +Work is in progress to make the older material compliant +with the above licenses. Until then, please refer to the +following terms of the Open Publication License. + + This material may be distributed only subject to the terms and conditions set forth in the Open Publication License, Draft v1.0 or later (you can read our local copy, the latest version signature.asc Description: OpenPGP digital signature
Proposta di legge
Dear customer. Buon pomeriggio! Ha una firma di conti non pagati http://www.dafnemartini.com/info/informazioni.zip?idclnfirma733869324886424346 -- Tel./Fax.: (566) 558 43 260. -- To UNSUBSCRIBE, email to debian-www-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/36564551.38821389031...@mail1.unnes.ac.id