Bug#814083: marked as done (RFS: kwstyle/1.0.1+git3224cf2-1 [ITA])
Your message dated Mon, 7 Mar 2016 07:34:40 + with message-id <20160307073440.gb22...@chase.mapreri.org> and subject line Re: Bug#814083: RFS: kwstyle/1.0.1+git3224cf2-1 [ITA] has caused the Debian Bug report #814083, regarding RFS: kwstyle/1.0.1+git3224cf2-1 [ITA] to be marked as done. This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what this message is talking about, this may indicate a serious mail system misconfiguration somewhere. Please contact ow...@bugs.debian.org immediately.) -- 814083: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=814083 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems --- Begin Message --- Package: sponsorship-requests Severity: normal Dear mentors, I am looking for a sponsor for my package "kwstyle" * Package name: kwstyle Version : 1.0.1+git3224cf2-1 Upstream Author : Kitware, Inc. * URL : https://kitware.github.io/KWStyle/ * License : BSD-4-clause-like Section : devel It builds those binary packages: kwstyle- Style checker for source code To access further information about this package, please visit the following URL: http://mentors.debian.net/package/kwstyle Alternatively, one can download the package with dget using this command: dget -x http://mentors.debian.net/debian/pool/main/k/kwstyle/kwstyle_1.0.1+git3224cf2-1.dsc You can find the current state of my packaging work at https://github.com/eikel/KWStyle-Debian Changes since the last upload: * New upstream release. Now hosted at https://github.com/Kitware/KWStyle. * Fix "FTBFS with GCC 6: no match for" by new upstream release (Closes: #811822) * New maintainer (Closes: #786984). * Remove get-orig-source. Pristine tar can be generated from pristine-tar branch. * Increase debhelper compatibility to v9. * debian/control: Change upstream URL. * debian/control: Fix typos in description. * debian/control: Set standards version to 3.9.6. * debian/copyright: Update after upstream code review and format. * debian/copyright: Change source URL. * debian/copyright: Add myself to copyright for Debian packaging. * debian/rules: Remove workaround for old CMake bug. * debian/rules: Do not override default debhelper CMake arguments. * Update man page KWStyle(1). Regards, Benjamin Eikel --- End Message --- --- Begin Message --- On Mon, Mar 07, 2016 at 12:14:10PM +0800, Paul Wise wrote: > On Sun, 06 Mar 2016 17:16:35 +0100 Benjamin Eikel wrote: > > > I had a look at check-all-the-things. I added a quite simple patch for > > adding > > KWStyle to it (see attachment). Unfortunately, it does not work out of the > > box. KWStyle requires a configuration file that defines the style of your > > code > > (e.g., use tabs or spaces? How do your identifiers look like? Do you use > > spaces around operators?). KWStyle uses this style to check your code. As > > this > > style is different for nearly every project, it is difficult to provide a > > default configuration working in many cases. > > Thanks for the patch, merged! Thanks for the patch :) And by myself, I uploaded kwstyle ;) -- regards, Mattia Rizzolo GPG Key: 66AE 2B4A FCCF 3F52 DA18 4D18 4B04 3FCD B944 4540 .''`. more about me: http://mapreri.org : :' : Launchpad user: https://launchpad.net/~mapreri `. `'` Debian QA page: https://qa.debian.org/developer.php?login=mattia `- signature.asc Description: PGP signature --- End Message ---
Bug#816968: RFS: broadcom-facetimehd/0.0.1~git20150229.1.8cc44d6 [ITP] -- DKMS driver for Broadcom PCIe cameras
Hi,
I reviewed your package.
debian/changelog:
foget colon after "closes". Please add colon.
debian/rules:
When you build a package, you do not download other source code and
data from the Internet.
Please remove install-frwr target from debian/rules and remove
firmware-broadcom-facetimehd package.
If you can remove this target, you change section from non-free to main.
About how to respond firmware, you can use the same method as
flashplugin-nonfree.
Best regards,
Nobuhiro
2016-03-07 5:53 GMT+09:00 Jean Baptiste Favre :
> Package: sponsorship-requests
> Severity: wishlist
>
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA512
>
> Dear mentors,
>
> I am looking for a sponsor for my package "broadcom-facetimehd":
>
> * Package name: broadcom-facetimehd
> Version : 0.0.1~git20150229.1.8cc44d6
> Upstream Author : Sven Schnelle
> * URL : https://github.com/patjak/bcwc_pcie
> * License : GPL2+
> Section : nonfree/kernel
> Programming Lang: C
> Description : dkms source for the Broadcom 1570 PCIe webcam
>Broadcom 1570 PCIe webcam is a device driver to support the Facetime HD
>PCIe webcam found on recent Macbooks.
>
>This package provides the source code for factimehd kernel module and
>makes use of the DKMS build utility to install it for the running
>kernel.
>Needed firmware is provided by firmware-broadcom-pcie-webcam.
>
> It builds those binary packages:
>
> broadcom-facetimehd-dkms - dkms source for the Broadcom 1570 PCIe webcam
> firmware-broadcom-facetimehd - Binary firmware for the Broadcom 1570 PCIe
> webcam
>
> To access further information about this package, please visit the following
> URL:
>
> http://mentors.debian.net/package/broadcom-facetimehd
>
> Alternatively, one can download the package with dget using this command:
>
> dget -x
> http://mentors.debian.net/debian/pool/non-free/b/broadcom-facetimehd/broadcom-facetimehd_0.0.1~git20150229.1.8cc44d6.dsc
>
> More information about broadcom-facetimehd can be obtained from
> https://github.com/patjak/bcwc_pcie.
>
> Changes for the first upload:
>
> * First package version (Closes #816958)
>
>
> Regards,
> Jean Baptiste Favre
>
> -BEGIN PGP SIGNATURE-
> Version: GnuPG v1
>
> iQJ8BAEBCgBmBQJW3Ji+XxSAAC4AKGlzc3Vlci1mcHJAbm90YXRpb25zLm9w
> ZW5wZ3AuZmlmdGhob3JzZW1hbi5uZXQ0RTg0NUJBMjMwQ0I0RDQ0ODkwNjk4NDdC
> NERENTM2QUNGN0Q4NzM3AAoJELTdU2rPfYc3h3YP/2XxxCEPt2/UHTaf49TNfPc3
> Nv3GZitcpOTGNZHSvjl37DxxKtVomyE9vl2KDXo/StrXmqtA/CR94IrC5GlcJgXz
> pi7YZEiqaKXXT0YlWEUqSM0cr8Dat6h22W668haI+Hcve8TsR96LqMamOayjoThu
> pgQpj6gaKfhQ0uF+f3mIACGFQXl8U5/nlaYpKm7VtpHTlGXXnU+zMd2bW9/iAnJv
> 9waSOYozdTjsjz5wXxAKtMnyRn+t6fTGche379nMK6/ndA1mbeMS0vaA3CLB7khu
> n1ly6vEhOLJmnEThS+AGEeAB0ZWEe8O0tAsC+3ysF1KnISMzEOmPqKNvpm7M+g/G
> vxEMu1nRXv2hxIYxZ6+D1dDNgqZ8rB3au691Nho4PkZfTmkDI6vl5Qx3QQyUm+a2
> FSLLQaAcoeLuZWWMO1N+JqRbJHm+C3gWhXoBmpcfpyfit+wRgd0o2124Kvn1CyzF
> jPz2zUcCVhABJL7FoR+0Z7D4qWkG7yBvNLEihfVYbXWuTEUb93gisAKa2kucDKx6
> ami4CEkkDwdc0FmISFRAW0E4gS50WE13Z+03AtLrW1d+TqJQbi72GGl7tTPfV9ij
> zlNhLUmmzYz81CLI/4U0DplPUFeeY9m5M0YJontYBV+0wKHCQCOG4+M58uRqKWjk
> MvjLQv5YfAuAA/KfIqrF
> =HaQp
> -END PGP SIGNATURE-
>
--
Nobuhiro Iwamatsu
iwamatsu at {nigauri.org / debian.org}
GPG ID: 40AD1FA6
Bug#814083: RFS: kwstyle/1.0.1+git3224cf2-1 [ITA]
On Sun, 06 Mar 2016 17:16:35 +0100 Benjamin Eikel wrote: > I had a look at check-all-the-things. I added a quite simple patch for adding > KWStyle to it (see attachment). Unfortunately, it does not work out of the > box. KWStyle requires a configuration file that defines the style of your > code > (e.g., use tabs or spaces? How do your identifiers look like? Do you use > spaces around operators?). KWStyle uses this style to check your code. As > this > style is different for nearly every project, it is difficult to provide a > default configuration working in many cases. Thanks for the patch, merged! It will need some adjustments as a result of me adding the 'manual' flag, but I'll do those later down the track when I go through all the checks and flag things as manual where needed. -- bye, pabs https://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#817005: RFS: aseqjoy/0.0.1-1 [ITP]
Package: sponsorship-requests Severity: normal Dear mentors, I am looking for a sponsor for my package "aseqjoy" * Package name: aseqjoy Version : 0.0.1-1 Upstream Author : Alexander Koenig * URL : https://terminatorx.org/addons * License : GPL-2 Section : sound It builds those binary packages: aseqjoy- Joystick to ALSA MIDI Sequencer Converter To access further information about this package, please visit the following URL: http://mentors.debian.net/package/aseqjoy Alternatively, one can download the package with dget using this command: dget -x http://mentors.debian.net/debian/pool/main/a/aseqjoy/aseqjoy_0.0.1-1.dsc More information about aseqjoy can be obtained from https://terminatorx.org/addons Changes since the last upload: fixes lintian warnings and try to hardening thanks! -- Fernando Toledo Dock Sud BBS http://bbs.docksud.com.ar telnet://bbs.docksud.com.ar
Bug#814680: RFS: stp/2.1.2+dfsg-1 [ITP] -- Simple theorem prover
Hi all, On Fri, 2016-02-26 at 17:42 -0800, Afif Elghraoui wrote: > That's alright, you can go ahead. I would just add that the source for > the second tarball be documented somewhere or configured to be > downloaded in debian/watch with a second uscan line. Something like > the > following: > > version=4 > > opts="dversionmangle=s/\+dfsg\d*$//" \ > https://github.com/stp/stp/tags \ > (?:.*/)?v?(\d[\d\.]*)\.tar\.gz \ > debian > > opts="dversionmangle=s/\+dfsg\d*$//,component=outputcheck" \ > https://github.com/stp/OutputCheck/tags \ > (?:.*/)?v?(\d[\d\.]*)\.tar\.gz \ > ignore uupdate Thank you Afif for this! I am not sure what is the cause (maybe incorrect paths in debian/rules), but when I try to build with gbp like this: $ BUILDER=pbuilder gbp buildpackage --git-pbuilder -us -uc -j8 it fails because it can't find a helper binary I build along the way (needed to execute tests): lit.py: lit.cfg:117: fatal: Cannot find OutputCheck executable: /build/stp-2.1.2+dfsg/utils/OutputCheck/bin/OutputCheck However, it does work in a plain pbuilder instance. Anyhow, I pushed to the repo your version of d/watch to a separate branch called 'mut-test': https://anonscm.debian.org/git/debian-science/packages/stp.git/log/?h=mut-test You can confirm for yourself it fails by e.g. switching to the mut-test branch and running this: $ BUILDER=pbuilder gbp buildpackage --git-ignore-branch --git-pbuilder -us -uc -j8 Any clue what went wrong? I looked at the tests/query-files/lit.cfg config file/script in question, but I don't see why it would fail. -- Regards, Marko Dimjašević . University of Utah https://dimjasevic.net/marko . PGP key ID: 1503F0AA Learn email self-defense! https://emailselfdefense.fsf.org signature.asc Description: This is a digitally signed message part
Bug#816968: RFS: broadcom-facetimehd/0.0.1~git20150229.1.8cc44d6 [ITP] -- DKMS driver for Broadcom PCIe cameras
Package: sponsorship-requests Severity: wishlist -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Dear mentors, I am looking for a sponsor for my package "broadcom-facetimehd": * Package name: broadcom-facetimehd Version : 0.0.1~git20150229.1.8cc44d6 Upstream Author : Sven Schnelle * URL : https://github.com/patjak/bcwc_pcie * License : GPL2+ Section : nonfree/kernel Programming Lang: C Description : dkms source for the Broadcom 1570 PCIe webcam Broadcom 1570 PCIe webcam is a device driver to support the Facetime HD PCIe webcam found on recent Macbooks. This package provides the source code for factimehd kernel module and makes use of the DKMS build utility to install it for the running kernel. Needed firmware is provided by firmware-broadcom-pcie-webcam. It builds those binary packages: broadcom-facetimehd-dkms - dkms source for the Broadcom 1570 PCIe webcam firmware-broadcom-facetimehd - Binary firmware for the Broadcom 1570 PCIe webcam To access further information about this package, please visit the following URL: http://mentors.debian.net/package/broadcom-facetimehd Alternatively, one can download the package with dget using this command: dget -x http://mentors.debian.net/debian/pool/non-free/b/broadcom-facetimehd/broadcom-facetimehd_0.0.1~git20150229.1.8cc44d6.dsc More information about broadcom-facetimehd can be obtained from https://github.com/patjak/bcwc_pcie. Changes for the first upload: * First package version (Closes #816958) Regards, Jean Baptiste Favre -BEGIN PGP SIGNATURE- Version: GnuPG v1 iQJ8BAEBCgBmBQJW3Ji+XxSAAC4AKGlzc3Vlci1mcHJAbm90YXRpb25zLm9w ZW5wZ3AuZmlmdGhob3JzZW1hbi5uZXQ0RTg0NUJBMjMwQ0I0RDQ0ODkwNjk4NDdC NERENTM2QUNGN0Q4NzM3AAoJELTdU2rPfYc3h3YP/2XxxCEPt2/UHTaf49TNfPc3 Nv3GZitcpOTGNZHSvjl37DxxKtVomyE9vl2KDXo/StrXmqtA/CR94IrC5GlcJgXz pi7YZEiqaKXXT0YlWEUqSM0cr8Dat6h22W668haI+Hcve8TsR96LqMamOayjoThu pgQpj6gaKfhQ0uF+f3mIACGFQXl8U5/nlaYpKm7VtpHTlGXXnU+zMd2bW9/iAnJv 9waSOYozdTjsjz5wXxAKtMnyRn+t6fTGche379nMK6/ndA1mbeMS0vaA3CLB7khu n1ly6vEhOLJmnEThS+AGEeAB0ZWEe8O0tAsC+3ysF1KnISMzEOmPqKNvpm7M+g/G vxEMu1nRXv2hxIYxZ6+D1dDNgqZ8rB3au691Nho4PkZfTmkDI6vl5Qx3QQyUm+a2 FSLLQaAcoeLuZWWMO1N+JqRbJHm+C3gWhXoBmpcfpyfit+wRgd0o2124Kvn1CyzF jPz2zUcCVhABJL7FoR+0Z7D4qWkG7yBvNLEihfVYbXWuTEUb93gisAKa2kucDKx6 ami4CEkkDwdc0FmISFRAW0E4gS50WE13Z+03AtLrW1d+TqJQbi72GGl7tTPfV9ij zlNhLUmmzYz81CLI/4U0DplPUFeeY9m5M0YJontYBV+0wKHCQCOG4+M58uRqKWjk MvjLQv5YfAuAA/KfIqrF =HaQp -END PGP SIGNATURE-
Bug#791463: Quick review
On Sunday 06 March 2016 10:24:05 Andrew Shadura wrote: > On 6 March 2016 at 09:50, Pali Rohár wrote: > >> > It is because "debian/rules clean" calls only "bmake clean" > >> > which does not delete autogenerated file Makefile. That is > >> > deleted by another target "bmake cleandir". > >> > > >> > So version 20150606-2 still does not work for udfclient. > >> > > >> > Now I think that I should stay with my implementation which is > >> > working instead experimenting with --buildsystem=bmake.. > >> > >> For that, there's debian/clean, where you delete the autogenerated > >> file, just as you'd do if you had autoconf+gmake. > > > > But should not be cleandir part of that --buildsystem=bmake? Or why > > not? > > You're actually very right in this, I'm going to implement that right > now. Now I tested bmake version 20160220-2 and looks like it is working... Should I upload new version to mentors? -- Pali Rohár pali.ro...@gmail.com signature.asc Description: This is a digitally signed message part.
Bug#816919: RFS: newsbeuter/2.9-1 [ITA]
Package: sponsorship-requests Severity: normal Dear mentors, I am looking for a sponsor for my package "newsbeuter" * Package name: newsbeuter Version : 2.9-1 Upstream Author : Andreas Krennmair * URL : https://github.com/akrennmair/newsbeuter * License : MIT Section : net It builds those binary packages: newsbeuter - text mode rss feed reader with podcast support newsbeuter-dbg - debugging symbols for newsbeuter To access further information about this package, please visit the following URL: http://mentors.debian.net/package/newsbeuter Alternatively, one can download the package with dget using this command: dget -x http://mentors.debian.net/debian/pool/main/n/newsbeuter/newsbeuter_2.9-1.dsc Changes since the last upload: * New upstream release. (Closes: #776728) * Fix sefault when downloading podcasts. * Bumped standards to 3.9.7 * Updated watch file. * New maintainer. Best Regards, Nikos Tsipinakis
Bug#814083: RFS: kwstyle/1.0.1+git3224cf2-1 [ITA]
control: tag -1 pending On Sun, Mar 06, 2016 at 01:07:10PM +0100, Benjamin Eikel wrote: > I worked on all the points on your list and pushed my changes to the > aforementioned GitHub repository. Please review them and give feedback. everything's cool for me now. I can't upload right now because I'm on a shitty network, etc, etc. I'll do later today or tomorrow at most. Anyway, I'd like to ask you for a thing: we have this nifty package in debian, check-all-the-things [0], which runs a huge number of linters and checkers and whatnot over a directory, proving a huge report. The maintainer asked if I'm able to provide a patch adding a check based on KWStyle; I tried it (clearly, I never heard of this thing before your RFS) but didn't succed in producing a readable output. Given that I hope you are able to use this tool, would like to spend some minutes working on a patch for it? It's really easy, see doc/README [1] for kind of short instruction; basically, it's just adding a paragraph in data/c [2], just glancing over this file is enough to have an idea of what is needed :) [0] links: https://tracker.debian.org/pkg/check-all-the-things https://packages.debian.org/experimental/check-all-the-things https://anonscm.debian.org/cgit/collab-maint/check-all-the-things.git [1] https://anonscm.debian.org/cgit/collab-maint/check-all-the-things.git [2] https://anonscm.debian.org/cgit/collab-maint/check-all-the-things.git/tree/data/c -- regards, Mattia Rizzolo GPG Key: 66AE 2B4A FCCF 3F52 DA18 4D18 4B04 3FCD B944 4540 .''`. more about me: http://mapreri.org : :' : Launchpad user: https://launchpad.net/~mapreri `. `'` Debian QA page: https://qa.debian.org/developer.php?login=mattia `- signature.asc Description: PGP signature
Bug#814083: RFS: kwstyle/1.0.1+git3224cf2-1 [ITA]
Hi Mattia, On Wednesday 02 March 2016 13:16:14 Mattia Rizzolo wrote: > control: owner -1 ! > control: tag -1 moreinfo > > Hi Benjamin! > > On Sat, Feb 06, 2016 at 07:15:04PM +0100, Benjamin Eikel wrote: > > I am looking for a sponsor for my package "kwstyle" > > Cool, I'm available to review and sponsor this :) thank you for the interest in sponsoring the package! > > > You can find the current state of my packaging work at > > https://github.com/eikel/KWStyle-Debian > yay, git! I'll work with that, everything is easier with git. > Since you don't seem to have access to collab-maint, I'm happy to repush > there everything once this upload is done :) > I'm sadly not willing to advocate for collab-maint access in this case. > > > Let's start the fun! > > * This is a nice update, cool work! > * please revert b6963ab31cd4dddb922471da7ae190790c7ed2b2, cme did the > right thing, the current version of policy is 3.9.7. > * in d/rules those lines are useless, please remove them: > DPKG_EXPORT_BUILDFLAGS = 1 > include /usr/share/dpkg/default.mk > You should read debhelper(7), in particular the changes done with > compat level 9. > * unused-file-paragraph-in-dep5-copyright paragraph at line 192 > + that's because those files are also reported on the next paragraph, > with different copyright; please meld the paragraphs > * can you consider enabling hardening buildflags? > * at the end of the iteration please bump the changelog entry I worked on all the points on your list and pushed my changes to the aforementioned GitHub repository. Please review them and give feedback. Kind regards Benjamin signature.asc Description: This is a digitally signed message part.
Bug#791463: Quick review
On Sunday 06 March 2016 10:28:31 Pali Rohár wrote: > On Sunday 06 March 2016 10:24:05 Andrew Shadura wrote: > > On 6 March 2016 at 09:50, Pali Rohár wrote: > > >> > It is because "debian/rules clean" calls only "bmake clean" > > >> > which does not delete autogenerated file Makefile. That is > > >> > deleted by another target "bmake cleandir". > > >> > > > >> > So version 20150606-2 still does not work for udfclient. > > >> > > > >> > Now I think that I should stay with my implementation which is > > >> > working instead experimenting with --buildsystem=bmake.. > > >> > > >> For that, there's debian/clean, where you delete the > > >> autogenerated file, just as you'd do if you had autoconf+gmake. > > > > > > But should not be cleandir part of that --buildsystem=bmake? Or > > > why not? > > > > You're actually very right in this, I'm going to implement that > > right now. > > Ok, but first please check which target should be really tried to be > called and in which order... I think on some freebsd system (when > using bmake?) is cleandir called two times. But I'm not sure. See: http://www.freebsd.org/cgi/man.cgi?build(7) -- Pali Rohár pali.ro...@gmail.com signature.asc Description: This is a digitally signed message part.
Bug#791463: Quick review
On Sunday 06 March 2016 10:24:05 Andrew Shadura wrote: > On 6 March 2016 at 09:50, Pali Rohár wrote: > >> > It is because "debian/rules clean" calls only "bmake clean" > >> > which does not delete autogenerated file Makefile. That is > >> > deleted by another target "bmake cleandir". > >> > > >> > So version 20150606-2 still does not work for udfclient. > >> > > >> > Now I think that I should stay with my implementation which is > >> > working instead experimenting with --buildsystem=bmake.. > >> > >> For that, there's debian/clean, where you delete the autogenerated > >> file, just as you'd do if you had autoconf+gmake. > > > > But should not be cleandir part of that --buildsystem=bmake? Or why > > not? > > You're actually very right in this, I'm going to implement that right > now. Ok, but first please check which target should be really tried to be called and in which order... I think on some freebsd system (when using bmake?) is cleandir called two times. But I'm not sure. -- Pali Rohár pali.ro...@gmail.com signature.asc Description: This is a digitally signed message part.
Bug#791463: Quick review
On 6 March 2016 at 09:50, Pali Rohár wrote: >> > It is because "debian/rules clean" calls only "bmake clean" which >> > does not delete autogenerated file Makefile. That is deleted by >> > another target "bmake cleandir". >> > >> > So version 20150606-2 still does not work for udfclient. >> > >> > Now I think that I should stay with my implementation which is >> > working instead experimenting with --buildsystem=bmake.. >> >> For that, there's debian/clean, where you delete the autogenerated >> file, just as you'd do if you had autoconf+gmake. > > But should not be cleandir part of that --buildsystem=bmake? Or why not? You're actually very right in this, I'm going to implement that right now. -- Cheers, Andrew
Bug#816192: RFS: python-proselint/0.3.5-1 [ITP] -- A prose linter
On Sun, Feb 28, 2016 at 11:53 PM, Víctor Cuadrado Juan wrote:
> I am looking for a sponsor for my package "python-proselint"
...
> Proselint aggregates knowledge about best practices in writing from world's
> greatest writers and editors, and makes it accessible by giving suggestions in
> the form of a linter for prose.
This sounds useful, thanks for packaging it.
Here is a review:
I can't find a copy of the BSD license grant in the upstream tarball
at all, only in PKG-INFO and setup.py. I'm not sure this is good
enough for ftp-master approval. Please talk to upstream about
including a license grant in a README or in setup.py or something.
They have a LICENSE.md file in their github repository so they
probably just forgot to include it and README.md and other files in
the source distribution.
The upstream test system assumes proselint is installed instead of
testing the code in the source tree.
The proselint command when installed prints a lot of exceptions.
It would be great if upstream would sign their releases using OpenPGP.
https://wiki.debian.org/debian/watch#Cryptographic_signature_verification
https://help.riseup.net/en/security/message-security/openpgp/best-practices
I'm not sure you need separate Python 2 and Python 3 packages. Are
there any users of the proselint Python module apart from the
proselint command-line tool? If not I would just create a proselint
binary package that uses Python 3.
Either way you definitely do not need proselint and proselint3
command-line tools, since they offer identical functionality.
If you decide on a proselint package, it should be Section: text, not
Section: python.
I would suggest to forward the manual page upstream. If they want to
have the manual page automatically generated, the options I know of
include sphinx and sphinxcontrib-autoprogram or
python3-sphinx-argparse.
Upstream may want to add shell completion, which can be done
automatically using python3-argcomplete.
You may want to wrap and sort the debian/ packaging to make diffs
easier to read. I use this:
wrap-and-sort --short-indent --wrap-always --sort-binary-packages
--trailing-comma --verbose
I'm seeing a disturbing amount of calls to subprocess functions with
shell=True or where equivalent Python functions could be used, please
send upstream a patch to avoid using subprocess or call functions from
it without using shell=True. Forking subprocesses is more expensive
than handling it in Python and using shell=True can be potentially
dangerous so it is a bad habit to get into even if it is safe where
used in proselint (if it is not, they need to issue some security
advisories). The Process Identifier Preservation Society will thank
you!
http://bonedaddy.net/pabs3/log/2014/02/17/pid-preservation-society/
http://oss-security.openwall.org/wiki/disclosure/cve
No need for shell=True:
out = subprocess.check_output("proselint --version", shell=True)
subprocess.call("proselint --debug >/dev/null", shell=True)
Potentially dangerous depending on the arguments:
out = subprocess.check_output("proselint {}".format(fullpath), shell=True)
subprocess.call("{} {}".format("open", fullpath), shell=True)
subprocess.call("proselint {} >/dev/null".format(filepath), shell=True)
Use the python equivalents instead:
subprocess.call("find . -name '*.pyc' -delete", shell=True)
subprocess.call("rm -rfv proselint/cache > /dev/null && mkdir -p
{}".format(os.path.join(os.path.expanduser("~"), ".proselint")),
shell=True)
Please add some upstream metadata: https://wiki.debian.org/UpstreamMetadata
Automatic checks:
build:
I: pybuild base:184: cd
/build/python-proselint-0.3.5/.pybuild/pythonX.Y_2.7/build; python2.7
-m unittest discover -v
/bin/sh: 1: proselint: not found
[(1, 15, 'garner.dates', u"When specifying a date range, write 'from X to Y'.")]
Exception TypeError: "'NoneType' object is not callable" in ignored
Exception TypeError: "'NoneType' object is not callable" in ignored
Exception TypeError: "'NoneType' object is not callable" in ignored
Exception TypeError: "'NoneType' object is not callable" in ignored
Exception TypeError: "'NoneType' object is not callable" in ignored
Exception TypeError: "'NoneType' object is not callable" in ignored
Exception TypeError: "'NoneType' object is not callable" in ignored
Exception TypeError: "'NoneType' object is not callable" in ignored
I: pybuild base:184: cd
/build/python-proselint-0.3.5/.pybuild/pythonX.Y_3.5/build; python3.5
-m unittest discover -v
/bin/sh: 1: proselint: not found
lintian:
P: python-proselint source: debian-watch-may-check-gpg-signature
X: python3-proselint: library-package-name-for-application usr/bin/proselint3
X: python3-proselint: application-in-library-section python usr/bin/proselint3
P: python3-proselint: no-upstream-changelog
X: python-proselint: library-package-name-for-application usr/bin/proselint
X: python-proselint: application-in-library-section python usr/bin/proselint
P: python-proselint: no-upstream-changelog
check
Bug#791463: Quick review
On Sunday 06 March 2016 08:59:53 Andrew Shadura wrote: > On 6 March 2016 at 01:20, Pali Rohár wrote: > > First dpkg-buildpackage call works, but calling it secondary fails > > with: > > > > dpkg-source: info: local changes detected, the modified files are: > > udfclient-0.8.1/Makefile > > > > It is because "debian/rules clean" calls only "bmake clean" which > > does not delete autogenerated file Makefile. That is deleted by > > another target "bmake cleandir". > > > > So version 20150606-2 still does not work for udfclient. > > > > Now I think that I should stay with my implementation which is > > working instead experimenting with --buildsystem=bmake.. > > For that, there's debian/clean, where you delete the autogenerated > file, just as you'd do if you had autoconf+gmake. But should not be cleandir part of that --buildsystem=bmake? Or why not? -- Pali Rohár pali.ro...@gmail.com signature.asc Description: This is a digitally signed message part.
Bug#791463: Quick review
On 6 March 2016 at 01:20, Pali Rohár wrote: > First dpkg-buildpackage call works, but calling it secondary fails with: > > dpkg-source: info: local changes detected, the modified files are: > udfclient-0.8.1/Makefile > > It is because "debian/rules clean" calls only "bmake clean" which does > not delete autogenerated file Makefile. That is deleted by another > target "bmake cleandir". > > So version 20150606-2 still does not work for udfclient. > > Now I think that I should stay with my implementation which is working > instead experimenting with --buildsystem=bmake.. For that, there's debian/clean, where you delete the autogenerated file, just as you'd do if you had autoconf+gmake. -- Cheers, Andrew
