Bug#339829: lintian wish 339829: check for homepages

2006-01-02 Thread Paul Wise
tag 339829 + patch usertag 339829 + sentpatch thanks I've attached a patch against lintian 1.23.14 that adds a check for the correctness of the Homepage/webpage part of the package description. It is by no means complete, but it is a reasonable check. Basically it checks for one blank line, then

Bug#339829: potential patch improvements

2006-01-04 Thread Paul Wise
On Wed, 2006-01-04 at 09:33 -0500, Justin Pryzby wrote: Packages for which this is a false-positive (such as slash, gnudip, and bake) These can be eliminated by checking for a url in the description too. It reduces some true positives also: abcmidi, achims-guestbook, airsnort,

Bug#339829: potential patch improvements

2006-01-05 Thread Paul Wise
On Thu, 2006-01-05 at 11:19 +0800, Paul Wise wrote: I'll add the last-para instead of last-line change and then send an updated patch. Patch attached. -- bye, pabs http://wiki.debian.org/PaulWise lintian-1.23.14-check-homepage-in-description-field.patch.gz Description: GNU Zip compressed

Bug#339829: potential patch improvements

2006-01-05 Thread Paul Wise
On Thu, 2006-01-05 at 14:59 -0500, Justin Pryzby wrote: How about the attached combination check - does my check and also does yours, with the changes that it checks a couple of other words, and checks for a url in the description too. $description =~

Bug#347510: lintian: menu-icon-not-in-xpm-format needs updating/removing

2006-01-11 Thread Paul Wise
Package: lintian Version: 1.23.14 Severity: normal The description for lintian's menu-icon-not-in-xpm-format check needs to be updated since it refers to section 3.4 of the menu policy, which does not currently exist. The alternative is to remove the check entirely, which seems appropriate,

Bug#347510: lintian: menu-icon-not-in-xpm-format needs updating/removing

2006-01-21 Thread Paul Wise
On Sat, 2006-01-21 at 15:03 -0800, Russ Allbery wrote: Aha, so that is where it is. Well, lintian should probably make what it is referring to more apparent, something like Debian Menu System 3.7. It does currently say menu 3.7; does that still seem unclear to you? To me, that suggested

Bug#438202: lintian: detect useless windows thumbnail databases (thumbs.db)

2007-08-16 Thread Paul Wise
Package: lintian Version: 1.23.34 Severity: wishlist Please add a check to search the .deb for windows thumbnail databases. These are generally useless in Debian binary packages and just take up space. They will be named Thumbs.db (or any variation in case) and file reports that they are

Bug#438203: lintian: detect useless MacOS Desktop Services Store files (.DS_Store)

2007-08-16 Thread Paul Wise
Package: lintian Version: 1.23.34 Severity: wishlist Please detect .DS_Store files in binary and source packages. They are generally not useful and just waste small amounts of space. More info: http://en.wikipedia.org/wiki/.DS_Store Packages containing these files: album

Bug#339829: Lintian home page check

2007-09-06 Thread Paul Wise
On Sun, 2006-03-26 at 17:14 -0800, Russ Allbery wrote: I really think the effort would be better spent standardizing an optional URL field in Policy so that people can start using that, package build tools can be updated where necessary to handle it, and package viewing and installation tools

Bug#438202: lintian: detect useless windows thumbnail databases (thumbs.db)

2007-10-14 Thread Paul Wise
On Sun, 2007-10-14 at 17:36 -0700, Russ Allbery wrote: If you do see ones with a variation in case, let me know. The glest-data package contains the lowercased variation: http://packages.debian.org/search?searchon=contentskeywords=Thumbs.dbmode=filenamesuite=unstablearch=any -- bye, pabs

architecture all packages

2008-01-01 Thread Paul Wise
Hi, It would be nice to have some stats on lintian.d.o about how many (and maybe which) source packages use each standards-version and how many use a specific minor version of policy in their standards version (perhaps that should be a lintian info/warning too). -- bye, pabs

Re: architecture all packages

2008-01-01 Thread Paul Wise
Woops, wrong subject. Initially I thought lintian.d.o wasn't checking arch all packages because the PTS link for the chromium-data lintian report is wrong (dash in place of underscore in the anchor). I'll following up to the qa.debian.org bug about that. -- bye, pabs

homepage in udebs?

2008-01-03 Thread Paul Wise
Hi all, I'm sponsoring ttf-khmeros, which produces a font udeb for d-i. The latest ttf-khmeros update moved the homepage from the package description to the new dpkg source package field. As a result, lintian complains with I: ttf-khmeros-udeb udeb: unknown-field-in-control homepage. Is this a

Re: homepage in udebs?

2008-01-03 Thread Paul Wise
On Thu, 2008-01-03 at 17:59 -0800, Russ Allbery wrote: I'd check with the d-i team and ask them their opinion. I'm happy to modify lintian to match if needed. Thanks for that. -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part

Re: homepage in udebs?

2008-01-04 Thread Paul Wise
On Fri, 2008-01-04 at 10:30 +, Colin Watson wrote: With my d-i hat on, I would certainly recommend not including the Homepage field in udebs. It's not actually forbidden as such, but any waste of space in udebs is frowned upon. I thought as much. I expect you could work around this by

dfsg/ds/etc versioning

2008-02-18 Thread Paul Wise
Hi, Does lintian check for and warn about versions like 1.2.3.dfsg1? The reason it should is this: 1.2.3 1.2.3+dfsg1 1.2.3.4 1.2.3.dfsg1 My NM found that more packages use the dot variant (621) than use the + variant (263). I also wonder if lintian checks for dsfg (vs dfsg) in the

Bug#466665: lintian: menu-command-not-in-package should not warn about the use of su-to-root

2008-02-20 Thread Paul Wise
Package: lintian Version: 1.23.45 Severity: normal Currently the menu-command-not-in-package warning is output when the menu file uses su-to-root. It should not do so, because su-to-root is available in the menu package, so it will be available to everyone who uses the menu item. -- bye, pabs

Bug#466665: lintian: menu-command-not-in-package should not warn about the use of su-to-root

2008-02-20 Thread Paul Wise
On Wed, 2008-02-20 at 10:07 -0800, Russ Allbery wrote: There's code in lintian specifically to not do this, and running lintian on two packages I have installed locally that use su-to-root in menus (magicfilter and alsa-utils) doesn't produce that tag. Could you point me at the package where

Bug#468876: lintian: please warn when icons are installed in the wrong dirs

2008-03-01 Thread Paul Wise
Package: lintian Version: 1.23.45 Severity: wishlist The password-gorilla package that I recently sponsored contains the following files: /usr/share/icons/hicolor/16x16/password-gorilla.png /usr/share/icons/hicolor/32x32/password-gorilla.png /usr/share/icons/hicolor/48x48/password-gorilla.png

Bug#468876: lintian: please warn when icons are installed in the wrong dirs

2008-03-02 Thread Paul Wise
On Sun, 2008-03-02 at 22:21 -0800, Russ Allbery wrote: That only lists some of the subdirectories of your list (it's missing all the stock/* ones) and doesn't use the same names (international instead of intl, applications instead of apps). All of those seem like things that could change on

pkg-games report is broken?

2008-04-25 Thread Paul Wise
Hi all, The pkg-games-devel report seems to be broken, mancala shows warnings on my page but not on the pkg-games-devel page: http://lintian.debian.org/reports/full/[EMAIL PROTECTED] http://lintian.debian.org/reports/full/[EMAIL PROTECTED] Any thoughts? -- bye, pabs

freedesktop and Debian menu files

2008-04-25 Thread Paul Wise
Hi all, Debian GNOME no longer shows the Debian menu by default, users have to set the Debian item in the menu to visible to get it. In my opinion the Debian-specific menu should be phased out in favour of the cross-distro freedesktop menu, with a transition like what is mentioned in #431118.

Re: freedesktop and Debian menu files

2008-04-25 Thread Paul Wise
On Fri, 2008-04-25 at 19:42 -0700, Russ Allbery wrote: I wish you the best of luck resolving this, but I'm not willing to touch that debate with a ten-foot pole, given the flamewar that happened on debian-devel the last time. (At least with my Lintian maintainer hat on; it might be

Re: pkg-games report is broken?

2008-04-25 Thread Paul Wise
On Fri, 2008-04-25 at 20:03 -0700, Russ Allbery wrote: I think I've fixed this. There were various logic bugs in the HTML page generation code introduced with the uploaders support. Thank you for the report! Hmm, seems that there is now a different bug. This page shows mancala:

Re: pkg-games report is broken?

2008-04-25 Thread Paul Wise
On Fri, 2008-04-25 at 20:35 -0700, Russ Allbery wrote: I'm not sure why what I just changed fixed that, but it seems to have. I think it's now okay. Thanks. One more minor issue, there is no anchor/id for packages where the person is an Uploader rather than a Maintainer: dtstrongmancala

Re: pkg-games report is broken?

2008-04-25 Thread Paul Wise
On Fri, 2008-04-25 at 20:57 -0700, Russ Allbery wrote: Right, that was intentional, since nothing linked to that for uploaders... although I guess it wouldn't hurt to add anyway. Thanks -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed

Bug#339829: Lintian home page check

2008-04-30 Thread Paul Wise
On Thu, 2008-05-01 at 00:43 +0200, Frank Lichtenheld wrote: I'm still not sure that the original patch in this bug report (trying to guess at whether a random URL found in the extended description should be a homepage) is really a good idea. I personally would vote for closing this bug.

Bug#504257: lintian: check for exit() or _exit() calls in shared libraries

2008-11-02 Thread Paul Wise
Package: lintian Version: 2.0.0 Severity: wishlist rpmlint added checks for exit() or _exit() calls in shared libraries: https://bugzilla.redhat.com/show_bug.cgi?id=450011 http://rpmlint.zarb.org/cgi-bin/trac.cgi/changeset/1448 I think this would be a useful addition to lintian. -- bye, pabs

Bug#504264: lintian: warn on empty files passed to dh_installdocs or in .deb

2008-11-02 Thread Paul Wise
Package: lintian Version: 2.0.0 Severity: wishlist Please warn about empty or whitespace-only files being listed in debian/*docs when dh_installdocs is being used, or on the command-line to the dh_installdocs command in debian/rules. Shipping useless files in .debs should be discouraged. Please

Bug#504264: lintian: warn on empty files passed to dh_installdocs or in .deb

2008-11-02 Thread Paul Wise
On Sun, 2008-11-02 at 21:22 -0800, Russ Allbery wrote: lintian already warns about empty files in the package in general, which would seem to cover this. Do you have a specific example that we can look at where this didn't work? Ah. I was reviewing an RFS, but only ran lintian on the source

Bug#504257: lintian: check for exit() or _exit() calls in shared libraries

2008-11-02 Thread Paul Wise
On Sun, 2008-11-02 at 21:21 -0800, Russ Allbery wrote: I have some packages that would trigger this because the library uses generic error handling routines that include functions that can call exit() or _exit() but which are never called in practice. Ideally, that dead code would be

Bug#512363: lintian: please detect and warn about swfobject.js

2009-01-19 Thread Paul Wise
Package: lintian Version: 2.1.6 Severity: wishlist swfobject.js is some JavaScript for using flash on the web. It is currently duplicated in 3 packages in Debian (and one HTML version??): http://packages.debian.org/search?suite=sidarch=anymode=filenamesearchon=contentskeywords=swfobject.js I

show override comments on the web page?

2009-01-24 Thread Paul Wise
Hi, Is it possible to show comments from the override file on the web? For example, /usr/share/lintian/overrides/apt contains this: # apt-mark is rarely used auxiliary script, we don't want to depend on # python-apt only for it. apt binary: python-script-but-no-python-dep ./usr/bin/apt-mark

Bug#512901: lintian: show override comments in the html?

2009-01-24 Thread Paul Wise
Package: lintian Severity: wishlist Is it possible to show comments from the override file on the web? For example, /usr/share/lintian/overrides/apt contains this: # apt-mark is rarely used auxiliary script, we don't want to depend on # python-apt only for it. apt binary:

Re: show override comments on the web page?

2009-01-24 Thread Paul Wise
On Sat, 2009-01-24 at 13:00 -0800, Russ Allbery wrote: Paul Wise p...@debian.org writes: Is it possible to show comments from the override file on the web? It's possible but a bit complex to implement since it requires parsing the comments out of the override file. Could you file a bug

Bug#298545: lintian: 298545: also detect free font duplicates?

2009-01-30 Thread Paul Wise
Would it be possible to also detect duplicates of free fonts already in Debian? As you can see from here, there are a ton of Bitstream Vera duplicates in Debian already (and these are just the ones with the same md5sum):

Bug#298545: lintian: 298545: also detect free font duplicates?

2009-01-30 Thread Paul Wise
On Fri, 2009-01-30 at 16:57 +0900, Atomo64 wrote in IRC: Atomo64 pabs: is it really needed to compare the md5 of the font files as shipped by the ttf- package and the duplicates? or can it just work like the current embedded-* checks? (i.e. by checking the file name) The name picks up most

Bug#513650: lintian: pedantic: check for fonts in non-font package

2009-01-30 Thread Paul Wise
Package: lintian Severity: wishlist Please check for *.ttf *.otf fonts in packages not named ttf-* and otf-*. These fonts should be generally be packaged separately and depended on because they are usually useful outside the package that embeds them (often they are decorative fonts, which are

Bug#514203: lintian: please check for version number badness

2009-02-04 Thread Paul Wise
Package: lintian Severity: wishlist [I sent this to lintian-maint, filing a bug so it isn't forgotten] I would like it if lintian warned about the following issues with package version numbers: 1.2.3.dfsg1 (info/warning): It should warn about this because of this: 1.2.3 1.2.3+dfsg1 1.2.3.4

Bug#515689: duplicate-font-file X also in ttf-aenigma, bad advice?

2009-02-17 Thread Paul Wise
On Tue, 2009-02-17 at 14:39 -0800, Russ Allbery wrote: Could you comment on this from the Debian font team perspective? What's the right thing to do in a case like this? I can comment from my perspective (best suggestion at the top): Firstly figure out what the font is doing in the package.

Bug#516250: lintian: warn about some embedded flash applications

2009-02-19 Thread Paul Wise
Package: lintian Version: 2.2.5 Severity: wishlist I was perusing the list of .swf (flash executables) files in the archive: http://packages.debian.org/search?suite=sidsearchon=contentskeywords=.swf I note the following files that should probably not be there and the corresponding software

Bug#516250: lintian: warn about some embedded flash applications

2009-02-19 Thread Paul Wise
On Fri, 2009-02-20 at 11:51 +0900, Paul Wise wrote: - http://weblogs.macromedia.com/flashjavascript/readme.html http://www.macromedia.com/go/flashjavascript BSD-like: http://weblogs.macromedia.com/flashjavascript

Bug#516250: lintian: warn about some embedded flash applications

2009-02-19 Thread Paul Wise
On Thu, 2009-02-19 at 19:34 -0800, Russ Allbery wrote: In this particular case, I think the best path forward would be to form a Flash packaging group that takes as their mission packaging the common code -- in other words, being proactive at developing the solution instead of trying to push

Bug#516534: lintian: *-patch-missing-description should mention more things

2009-02-21 Thread Paul Wise
Package: lintian Version: 2.2.5 Severity: wishlist I think *-patch-missing-description should mention more things and I propose the following replacement text. The things I mention are very useful for reducing divergence in cases where the Debian maintainer hasn't done a very good job of

binary-from-other-architecture multi-arch bits

2009-02-23 Thread Paul Wise
Hi all, Should binary-from-other-architecture be emitted for i386 binaries in amd64 packages where the package depends on the 32-bit versions of libc/etc. For example, nsis isn't portable to !i386, upstream builds with -m32 by default and I use that to give it to amd64 folks. Once Debian has

Re: binary-from-other-architecture multi-arch bits

2009-02-23 Thread Paul Wise
On Mon, 2009-02-23 at 10:05 -0800, Russ Allbery wrote: You should file a bug about this. I was expecting some problems along those lines since it's a brand-new check. Looking for a dependency on 32-bit libc is a good way of detecting this case and avoiding it. Thanks! Will do. Also,

Bug#516856: lintian: don't emit binary-from-other-architecture for bi-arch packages

2009-02-23 Thread Paul Wise
Package: lintian Version: 2.2.6 Severity: normal nsis on amd64 is built for i386 for portability issues. I can do this because of libc6-i386, which is for running 32-bit apps on a 64-bit system. Until Debian gets multi-arch support (hopefully for squeeze), lintian should not warn about i386

Bug#516858: lintian: false negative arch-dependent-file-in-usr-share (nsis)

2009-02-23 Thread Paul Wise
Package: lintian Version: 2.2.6 Severity: normal NSIS contains the following architecture-specific files in /usr/share. These are all binaries for Microsoft Windows built with the mingw32 cross-compiler we have in Debian. nsis is a tool to create installers for Windows applications and so it

Bug#516858: lintian: false negative arch-dependent-file-in-usr-share (nsis)

2009-02-23 Thread Paul Wise
On Mon, 2009-02-23 at 18:16 -0800, Russ Allbery wrote: It's not warning because they're not ELF, and those are the only types of binaries that it warns about. I figured as much. Hm. I'm somewhat torn on this -- my guess was that whenever there are non-ELF binaries in an

Bug#517637: lintian: warn about watch files that don't detect tarball type changes

2009-02-28 Thread Paul Wise
Package: lintian Version: 2.2.6 Severity: wishlist I recently found a package where upstream switched from tar.gz to tar.bz2 and released a new upstream version. The debian/watch file for that package didn't detect the new upstream version because it was looking for tar.gz and upstream only

Bug#517637: lintian: warn about watch files that don't detect tarball type changes

2009-02-28 Thread Paul Wise
On Sun, 2009-03-01 at 12:23 +0900, Paul Wise wrote: I think lintian should warn about this situation (and the reverse) and suggest that both tar.gz and tar.bzip2 be checked for. I forgot to mention that it should also check for tgz maybe tbz2/tbz. The canonical watch regex for stuff

Bug#517637: lintian: warn about watch files that don't detect tarball type changes

2009-02-28 Thread Paul Wise
On Sun, 2009-03-01 at 12:39 +0900, Paul Wise wrote: The canonical watch regex for stuff that currently detects only one or two variants should probably be increased to something like this: (tar|tar\.gz|tar\.bz2|tgz|tbz|tbz2) Make that (?:tar|tar\.gz|tar\.bz2|tgz|tbz|tbz2) This ensures

Bug#517637: lintian: warn about watch files that don't detect tarball type changes

2009-02-28 Thread Paul Wise
On Sat, 2009-02-28 at 20:14 -0800, Russ Allbery wrote: Hm, I've seen upstreams switch from .tar.gz to .tar.bz2, but I've not seen one switch from .tar.gz to .tgz or vice versa. Maybe my experience is just limited, though. I'd be very surprised to see any upstream go back to a plain .tar

Bug#517637: uscan: 517637: add an option to auto-generalise the search regex

2009-03-01 Thread Paul Wise
retitle 517637 [uscan] add an option to auto-generalise the search regex reassign 517637 devscripts thanks On Sun, 2009-03-01 at 13:35 +0900, Paul Wise wrote: Hmm, perhaps this should be a job for dehs or debexpo, which could run uscan and be a bit smarter about checking the results

lintian NMU check and teams

2009-04-05 Thread Paul Wise
Hi all, I've been thinking that it might be a good idea for lintian to not emit the NMU warnings when the changelog has a Foo team upload message at the beginning. I realise this would be a fairly major change for Debian. It would make me much more comfortable doing more QA and more uploads

Re: lintian NMU check and teams

2009-04-05 Thread Paul Wise
On Sun, 2009-04-05 at 12:54 -0700, Russ Allbery wrote: I'm not sure what the drawback is to adding yourself to Uploaders. Could you elaborate? Well, it means that in various places I'll be marked as maintaining the package rather than having done a once-off upload to bring the packaging up to

Re: lintian NMU check and teams

2009-04-05 Thread Paul Wise
On Sun, 2009-04-05 at 20:36 -0700, Russ Allbery wrote: Hm. So it's sort of a team NMU -- you're not planning on taking ongoing responsibility for the package except as just another member of the team. Yeah, a team upload is a good way of putting it. You're uploading it as a member of the

Bug#525782: lintian: warn about #MISSING: ... in symbols files

2009-04-26 Thread Paul Wise
Package: lintian Version: 2.2.9 Severity: wishlist Please warn about symbols files containing #MISSING: ... These lines indicate some symbols have disappeared in newer versions of a library. The removed symbols might have been private symbols that have been hidden or they were public symbols

Bug#545781: lintian.d.o: issues with reports of maintainers with lintian-clean packages

2009-09-09 Thread Paul Wise
Package: lintian Severity: minor mat...@eisbox.net has no lintian complaints on his single package and his lintian pages say that. http://lintian.debian.org/maintainer/mat...@eisbox.net.html#libicns http://lintian.debian.org/full/mat...@eisbox.net.html#libicns Both pages link to the directory

Bug#528002: lintian: 528002: warn on packages with menu files but no desktop files

2010-05-12 Thread Paul Wise
I'd like to suggest that lintian should also warn about the reverse situation (menu file but no desktop file). If the lintian maintainers agree, please retitle this bug. -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part

Bug#592379: lintian: python packages with “except” overwriting builtins

2010-08-09 Thread Paul Wise
Package: lintian Severity: wishlist Some python-based applications use incorrect constructs in their exception handling, which was reported in the following thread and results in some exceptions not being handled properly. http://lists.debian.org/debian-python/2010/08/msg1.html

Re: Untrusted search path vulnerabilities

2010-11-18 Thread Paul Wise
On Wed, Nov 17, 2010 at 22:58, Jakub Wilk jw...@debian.org wrote: A number of packages in the archive sets the PYTHONPATH environment variable in an insecure way. They do something like:      PYTHONPATH=/spam/eggs:$PYTHONPATH This is wrong, because if PYTHONPATH were originally unset or

Bug#622149: [Pkg-fonts-devel] Bug#622149: lintian: false positive: font-in-non-font-package in fonts-* package

2011-04-11 Thread Paul Wise
In addition, I don't think the naming policy is yet ready for general use. For example no-one commented on my question about foundries, which I now think are a irrelevant in the package name, but might be relevant in the package description. -- bye, pabs http://wiki.debian.org/PaulWise --

[PATCH] Check for GLee, GLEW, libtheora embedded code copies

2011-05-01 Thread Paul Wise
--- checks/binaries |3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/checks/binaries b/checks/binaries index 8c46be2..ebfb859 100644 --- a/checks/binaries +++ b/checks/binaries @@ -113,6 +113,9 @@ our %EMBEDDED_LIBRARIES = ( 'openjpeg' = qr'tcd_decode:

Bug#628189: lintian: check for icons with a different size to the name of the directory they are in

2011-05-28 Thread Paul Wise
Package: lintian Severity: wishlist In #628146 it was reported that desktop-base installs a 36x36 into /usr/share/icons/hicolor/48x48/emblems/ instead of /usr/share/icons/hicolor/36x36/emblems/. It would be good if lintian could check for these issues: * images in the wrong

Bug#629288: lintian: Check for GLee, GLEW, libtheora embedded code copies

2011-06-06 Thread Paul Wise
On Sun, 2011-06-05 at 13:11 +0200, Niels Thykier wrote: I am not really sure what to say about the detection strings, except I feel they seem very generic (but I could be wrong)[1]. With embedded-library being an auto-reject tag now (about 45 minutes ago), I think we have to be more careful

lintian and Debian derivatives

2011-07-06 Thread Paul Wise
Hi all, As a member of the Debian derivatives front desk (CCed) and initiator of the derivatives census, which aims to make derivatives more visible to Debian, I figured I should update lintian folks on the state of lintian development and usage in Debian derivatives.

Re: lintian and Debian derivatives

2011-07-07 Thread Paul Wise
On Thu, Jul 7, 2011 at 12:49 AM, Geoffrey Thomas wrote: We are the only one? I'm proud :-) We set that up about a year ago because why not, but in practice, we've more been using Lintian output from the package build process (debuild/sbuild) than the page, and making sure there are no

Re: lintian and Debian derivatives

2011-07-07 Thread Paul Wise
On Thu, Jul 7, 2011 at 11:18 AM, Neil Williams wrote: Lintian requires a lot of resources if it's to be run automatically against an entire distribution. Yeah. Also there is quite a bit of scope for expanding the range of tests lintian performs, by running external checkers like cppcheck,

Re: Lintian as a static analysis framework

2011-07-07 Thread Paul Wise
On Thu, Jul 7, 2011 at 6:24 PM, Stefano Zacchiroli wrote: - the reborn of a sources.debian.org service (sort of browsable /  highlighted/ searchable Debian sources at your fingertips on the web) Is there any info about this? Last year Noel was working on such a project also:

Re: Lintian as a static analysis framework

2011-07-08 Thread Paul Wise
On Fri, Jul 8, 2011 at 9:25 AM, Stefano Zacchiroli wrote: I haven't yet chosen the technology for code indexing, but if anyone has experience with *multi-language* code indexers, I'd be happy to hear from you. I've looked around a bit, but I've found good technologies only for specific

Re: Lintian as a static analysis framework

2011-07-08 Thread Paul Wise
On Fri, Jul 8, 2011 at 12:30 PM, Niels Thykier wrote: We could do it like that, though the vendor profiles specification actually deliberately did not answer the question of how to add third-party checks.  I know some people already do this, so we have made Lintian behave sanely to it.  The

[lintian] [PATCH 1/2] Also search for fonts in packages with names starting with fonts-

2011-08-25 Thread Paul Wise
This is the new font naming policy for Debian. --- data/files/fonts |2 +- private/refresh-fonts-data |4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/data/files/fonts b/data/files/fonts index 69631a0..03da667 100644 --- a/data/files/fonts +++

[lintian] [PATCH 2/2] Refresh fonts data against unstable

2011-08-25 Thread Paul Wise
--- data/files/fonts | 463 ++ 1 files changed, 324 insertions(+), 139 deletions(-) diff --git a/data/files/fonts b/data/files/fonts index 03da667..e211f7e 100644 --- a/data/files/fonts +++ b/data/files/fonts @@ -2,7 +2,7 @@ # package that

Bug#365213: lintian: 365213: non-free RFC check code

2011-10-19 Thread Paul Wise
Apparently there is some code for checking for non-free RFCs here: http://wiki.debian.org/NonFreeIETFDocuments http://git.josefsson.org/cgi-bin/gitweb.cgi?p=tools.git;a=tree;f=rfc-search;hb=HEAD -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally

[PATCH] Also detect COFF object files as Windows binaries.

2011-12-19 Thread Paul Wise
--- checks/cruft |2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/checks/cruft b/checks/cruft index 25ee03b..520e155 100644 --- a/checks/cruft +++ b/checks/cruft @@ -149,6 +149,8 @@ for my $file (keys(%$file_info)) { tag 'source-contains-prebuilt-binary', $file;

Bug#653910: lintian: please detect and warn about *.chm (MS Windows HtmlHelp Data) in source/binary packages

2012-01-01 Thread Paul Wise
Package: lintian Severity: wishlist Please detect and warn about *.chm files. The file program says these are MS Windows HtmlHelp Data and they all begin with the 4 bytes ITSF, which is the magic of the container format. CHM files are mainly produced by proprietary, Windows-specific software.

increasing default lintian verbosity?

2012-01-21 Thread Paul Wise
Hi all, One thing I see on debian-mentors a lot is a mismatch between the default configuration of lintian (used by most new people) and the configuration of people reviewing packages (much more verbose). This comes up a lot, resulting in people needing to explain the difference a lot and people

Re: increasing default lintian verbosity?

2012-01-21 Thread Paul Wise
On Sat, 2012-01-21 at 22:02 -0800, Russ Allbery wrote: I don't think this is a good idea. The reason why Lintian is conservative about what it reports by default is because we know from past history that people will not use it if they get too many tags that they think are pointless or are

Re: increasing default lintian verbosity?

2012-01-21 Thread Paul Wise
On Sat, 2012-01-21 at 23:02 -0800, Russ Allbery wrote: Yeah, but that's in the middle of a rather long document. Given how integral Lintian is to the mentoring process, I wonder if it would be good to push it at people more directly on the pages they have to read to learn how to upload. For

Re: increasing default lintian verbosity?

2012-01-23 Thread Paul Wise
On Mon, 2012-01-23 at 13:04 +0100, Niels Thykier wrote: On a related note, I plotted (part of) the data[1] we got on debian-rules-missing-recommended-target tag[2]. The huge increase of affected packages in the start can most likely be attributed to this full-run[3]. The second spike

Bug#657110: lintian: please check for commented-out Vcs-Git/Vcs-Browser pointing at collab-maint

2012-01-23 Thread Paul Wise
Package: lintian Severity: wishlist dh_make creates lines like this in debian/control by default: #Vcs-Git: git://git.debian.org/collab-maint/pkg.git #Vcs-Browser: http://git.debian.org/?p=collab-maint/pkg.git;a=summary I think lintian should check for these lines and warn if they are still

Bug#675459: lintian: please add a general framework for detecting missing source files

2012-06-01 Thread Paul Wise
Package: lintian Severity: wishlist It would be very nice if lintian could have detected this situation: http://lwn.net/Articles/453970/ This would need a mechanism to check for generated files that are missing source files. This bug isn't about any specific type of missing source, but I have

Bug#685583: lintian: check for gsettings schemas with incorrect paths

2012-08-21 Thread Paul Wise
Package: lintian Version: 2.5.10.1 Severity: wishlist glib from experimental removed the patch that disabled warnings about deprecated gsettings schema paths, leading to warnings like these in during upgrade: Processing triggers for libglib2.0-0:amd64 ... warning: Schema

Bug#687230: lintian: check for duplicated translation data in binary packages

2012-09-10 Thread Paul Wise
Package: lintian Severity: wishlist Please add a check for duplicated translation data. By duplicated translation data I mean the same translation data in multiple formats. This happens when both the source and binary forms of translation data are shipped in binary packages. The source forms of

Re: [proposal] change color for version number if upstream uses github and just check master

2012-09-14 Thread Paul Wise
On Sat, Sep 15, 2012 at 5:03 AM, Hideki Yamane wrote: It has watch results and show version number with pink color if upstream released newer version than Debian package. However, if upstream uses github and sometimes watch file is check just its master branch in git (without enough

Bug#690014: lintian: please encourage updating to config.guess and config.sub that support arm64

2012-10-08 Thread Paul Wise
Package: lintian Version: 2.5.10.2 Severity: wishlist User: debian-...@lists.debian.org Usertags: arm64 X-Debbugs-CC: debian-...@lists.debian.org Please encourage updating to config.guess and config.sub that support arm64. The GNU name for arm64 is aarch64. aarch64 was added to

Re: Intend to change lintian.d.o/qa-list.txt

2012-12-29 Thread Paul Wise
On Sun, Dec 30, 2012 at 4:09 AM, Niels Thykier wrote: As far as I know, Lintian exports a file called qa-list.txt for consumption by the QA infrastructure. Particularly I suspect that the PTS and the DDPO consumes this file to display the relevant TODO item/column. If anyone is aware of any

Re: Intend to change lintian.d.o/qa-list.txt

2012-12-30 Thread Paul Wise
On Sun, Dec 30, 2012 at 5:16 PM, Stefano Zacchiroli wrote: Looks like the two formats are easily mechanically distinguishable: the That part is easy but I was not thinking about it, but about how the PTS should present the lintian data. -- bye, pabs http://wiki.debian.org/PaulWise -- To

Bug#697161: [new check] check for patches that apply changes twice

2013-01-01 Thread Paul Wise
Package: lintian Version: 2.5.11 Severity: wishlist dpkg-source on squeeze and earlier errors out when one of the dpkg-source 3.0 (quilt) patches modifies one file twice. With dpkg-source on wheezy and later this becomes a warning instead. I've encountered this issue while unpacking source

Bug#697164: [new check] check for source packages with symlinks pointing outside

2013-01-01 Thread Paul Wise
Package: lintian Version: 2.5.11 Severity: wishlist dpkg-source is not able (#645157) to properly handle source packages with relative/absolute symlinks that point outside of the package. It would be good if lintian could detect this situation and give an error. lintian should check both the

Re: Intend to change lintian.d.o/qa-list.txt

2013-01-13 Thread Paul Wise
On Sun, Jan 13, 2013 at 5:14 PM, Niels Thykier wrote: Are there any news on the PTS side? DDPO has been fixed already, so to my knowledge we are currently just waiting for the PTS. Still discussing how the data should be presented AFAICS. What do the lintian maintainers think? Should the

Re: Intend to change lintian.d.o/qa-list.txt

2013-01-14 Thread Paul Wise
On Sun, Jan 13, 2013 at 6:15 PM, Paul Wise wrote: Still discussing how the data should be presented AFAICS. What do the lintian maintainers think? Should the links remain as they are and point at data for unstable only? Bart suggested off-list to just keep things as-is for now and deal

Bug#702545: lintian: check compressed files

2013-03-08 Thread Paul Wise
In addition, the various other compression formats include test options: find -type f -iname \*.bz2 -print0 | xargs --no-run-if-empty --null bzip2 --test find -type f -iname \*.xz -print0 | xargs --no-run-if-empty --null xz --test find -type f -iname \*.zip -print0 | xargs --no-run-if-empty

[lintian] [PATCH] Also warn about incorrect case with OpenStreetMap

2013-04-08 Thread Paul Wise
--- data/spelling/corrections-case | 3 +++ 1 file changed, 3 insertions(+) diff --git a/data/spelling/corrections-case b/data/spelling/corrections-case index 20507f0..03a2700 100644 --- a/data/spelling/corrections-case +++ b/data/spelling/corrections-case @@ -54,6 +54,9 @@ ocaml||OCaml

Bug#705441: lintian: drop extended-description-is-probably-too-short for -dbg packages

2013-04-14 Thread Paul Wise
Package: lintian Severity: wishlist Please drop extended-description-is-probably-too-short for -dbg packages since most of them are probably just one-liners like this: This package contains debug symbols for foo. -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This

Bug#707700: lintian: warn about broken Vcs-Browser URLs

2013-05-10 Thread Paul Wise
Package: lintian Severity: wishlist Vcs-Browser URLs of this form give HTTP 400 Bad Request, due to the ?op=log at the end. lintian should warn about them and not output the problematic part in the output of the other vcs-* tags. This is another unhandled part of the alioth transition a while

Bug#707906: lintian: examples for third-party checks

2013-05-11 Thread Paul Wise
Package: lintian Severity: wishlist It would be nice to have some examples for third-party checks. As an example, for the last few uploads I forgot to add debian/upstream files and I would like to remind myself for future uploads that they are missing. This is obviously not yet suitable for an

Bug#708482: lintian: warn about files named configure.in

2013-05-15 Thread Paul Wise
Package: lintian Severity: wishlist From [1]: - Automake 2.0 will drop support for the long-deprecated 'configure.in' name for the Autoconf input file. You are advised to start using the recommended name 'configure.ac' instead, ASAP. I would suggest the following plan for Debian

Bug#711069: lintian: detect self-containing headers in debian/control, DEBIAN/control etc

2013-06-04 Thread Paul Wise
Package: lintian Severity: wishlist We have some packages in the archive that have a Maintainer: field that starts with 'Maintainer: '. I think it would be good if lintian could detect this and similar situations where the field value starts with the field name and a colon. Here are some examples

  1   2   3   4   5   6   7   8   >