Re: Apparent missing dep in print/hplip
On Mon, Jun 13, 2016 at 03:53:52PM -0400, Michael McConville wrote: > Twenty seconds of search queries suggests that it's a missing Qt4 Python > library. Here's an apparently similar issue from Debian: > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=525556 If you need gui support, then install hplip-gui. Otherwise, run it in console mode (hp-setup -i). > mike:/home/mike$ hp-setup > > HP Linux Imaging and Printing System (ver. 3.16.5) > Printer/Fax Setup Utility ver. 9.0 > > Copyright (c) 2001-15 HP Development Company, LP > This software comes with ABSOLUTELY NO WARRANTY. > This is free software, and you are welcome to distribute it > under certain conditions. See COPYING file for more details. > > Traceback (most recent call last): > File "/usr/local/bin/hp-setup", line 313, in > ui = import_module(ui_package + ".setupdialog") > File "/usr/local/lib/python2.7/importlib/__init__.py", line 37, in > import_module > __import__(name) > ImportError: No module named ui4.setupdialog > -- Antoine
Re: pledge ffmpegthumbnailer
Hi, First, a bump of library version would be required as you added new exported functions. But I think your changes are too invasive for been patches for OpenBSD port tree. You should first deal with upstream for these changes, else it will be a shame for us to deal with future upgrade. Thanks. -- Sebastien Marie
pledge ffmpegthumbnailer
This thing has been used in the past as a vector to exploit ffmpeg library vulneraiblities, and it's also unpledgeable without some changes. It opens the only file it's going to write early on, so if we could pledge immediately after that it wouldn't need wpath or cpath, but the public methods to generate the thumbnail automatically open the file for writing and then begin decoding the video. We can't add pledge calls inside those methods as they're part of libffmpegthumbnailer and other programs might call them. The input source can use any protocol that libavformat supports (and one that it doesn't), so depending on the input it can need inet sockets or unix sockets, or both. This patch adds a public method to call the private method we need, so the image can be opened for writing and then pledge happens and then it goes on to decode the video. It also adds a method to determine the protocol that libav is going to use to access the input so an appropriate pledge can be picked. This method uses libavformat's avio_find_protocol_name() and compares it against known protocols that were obtained with avio_enum_protocols(). A call to av_register_all() is added early on. This sets up libav's protocol associations (which is needed for avio_find_protocol_name() to work) and also loads the codec libraries. This also prevents later calls to av_register_all() from trying to read in the codec libraries. This and a fix to a bogus stat(2) call, which would happen even when remote protocols are being used, prevents rpath being needed for non-file protocols. The pledges: Local files need rpath but no sockets. Network protocols need dns and inet. Unix sockets need unix. Neither of those need rpath. This means, in normal use, if it's reading a file it doesn't get sockets and if it's accessing sockets it doesn't get to read files. The unfortunate exceptions being some non-standard protocols that libav can use such as concat and crypto, which can use any and multiple protocols. Without these changes the best pledge would be: stdio rpath wpath cpath dns inet unix -- Carlin Index: graphics/ffmpegthumbnailer/Makefile === RCS file: /cvs/ports/graphics/ffmpegthumbnailer/Makefile,v retrieving revision 1.27 diff -u -p -u -r1.27 Makefile --- graphics/ffmpegthumbnailer/Makefile 11 Mar 2016 19:59:14 - 1.27 +++ graphics/ffmpegthumbnailer/Makefile 13 Jun 2016 16:03:49 - @@ -3,7 +3,7 @@ COMMENT= lightweight video thumbnailer for file managers DISTNAME= ffmpegthumbnailer-2.0.8 -REVISION= 3 +REVISION= 4 CATEGORIES=graphics multimedia MASTER_SITES= https://ffmpegthumbnailer.googlecode.com/files/ @@ -14,6 +14,7 @@ HOMEPAGE= https://github.com/dirkvdb/ffm # GPLv2+ PERMIT_PACKAGE_CDROM= Yes +# uses pledge() WANTLIB += avcodec avformat avutil c jpeg m png pthread stdc++ WANTLIB += swscale x264 Index: graphics/ffmpegthumbnailer/patches/patch-libffmpegthumbnailer_videothumbnailer_cpp === RCS file: graphics/ffmpegthumbnailer/patches/patch-libffmpegthumbnailer_videothumbnailer_cpp diff -N graphics/ffmpegthumbnailer/patches/patch-libffmpegthumbnailer_videothumbnailer_cpp --- /dev/null 1 Jan 1970 00:00:00 - +++ graphics/ffmpegthumbnailer/patches/patch-libffmpegthumbnailer_videothumbnailer_cpp 13 Jun 2016 16:03:49 - @@ -0,0 +1,114 @@ +$OpenBSD$ + +_generateThumbnail() provides a way to call the private +generateThumbnail() method with a manually created ImageWriter, this is +needed so pledge() can be called immediately after creating the +ImageWriter but before the first frame of the video is decoded. + +_getProtocolType() provides a method to determine the pledge() +requirements of the protocol that is being used. + +Avoiding the stat() call on non-file protocols helps allow for dropping +of the rpath promise when using network protocols. + +--- libffmpegthumbnailer/videothumbnailer.cpp.orig Sun Aug 26 00:07:44 2012 libffmpegthumbnailer/videothumbnailer.cpp Tue Jun 14 03:55:54 2016 +@@ -37,6 +37,10 @@ + #include + #include + ++extern "C" { ++#include ++} ++ + using namespace std; + + namespace ffmpegthumbnailer +@@ -177,6 +181,11 @@ void VideoThumbnailer::generateSmartThumbnail(MovieDec + videoFrame = videoFrames[bestFrame]; + } + ++void VideoThumbnailer::_generateThumbnail(const string& videoFile, ImageWriter& imageWriter) ++{ ++generateThumbnail(videoFile, imageWriter, NULL); ++} ++ + void VideoThumbnailer::generateThumbnail(const string& videoFile, ThumbnailerImageType type, const string& outputFile, AVFormatContext* pAvContext) + { + ImageWriter* imageWriter = ImageWriterFactory::createImageWriter(type, outputFile); +@@ -194,7 +203,7 @@ void VideoThumbnailer::generateThumbnail(const string& + + void VideoThumbnailer::writeImage(const string& videoFile, ImageWriter& imageWriter, const VideoFrame& vid
Re: pledge par2cmdline
On Mon, Jun 13, 2016 at 08:42:34PM +0100, Mikolaj Kucharski wrote: > On Mon, Jun 13, 2016 at 07:35:23PM +0200, Sebastien Marie wrote: > > On Sun, Jun 12, 2016 at 09:17:27PM +0100, Mikolaj Kucharski wrote: > > > Will you be driving this to get merged upstream Carlin? > > > > Trying to push it upstream would be a good thing. > > I can fill pull request if you want. However if Carlin wrote the diffs I > think it would be better if he drives this, but I don't mind doing it. > Just let me know. > If you think upstream will take this I'm happy for you to try and get it commited. -- Carlin
File::LibMagic dumping core non-deterministically
Hello, on my mail server running postfix + amavisd, amavisd was dying occasionally when using File::LibMagic to analyze mail messages. I can reproduce it on the server (OpenBSD the-server 5.9 GENERIC.MP#3 amd64) with this Perl script: > #!/usr/bin/perl > > use File::LibMagic; > > my $file = shift; > $file || die "usage: $0 \n"; > > my $magic = File::LibMagic->new(); > > for(my $i = 0; $i < 10; $i++) { > my $desc = $magic->describe_filename($file); > if ($i % 100 == 0) { print "."; } > } using the following file (see bottom for b64 encoded version): # cat -tv p005 > $OpenBSD$ > > Use C codegen for mlton-20130715. > See https://github.com/MLton/mlton/issues/148 for details. > > --- mlyacc/Makefile.orig^IThu May 26 13:14:56 2016 > +++ mlyacc/Makefile^IThu May 26 13:15:10 2016 > @@ -13,7 +13,7 @@ BIN := $(BUILD)/bin > LIB := $(BUILD)/lib > MLTON := mlton > TARGET := self > -FLAGS := -target $(TARGET) > +FLAGS := -codegen c -target $(TARGET) > NAME := mlyacc > PATH := $(BIN):$(shell echo $$PATH) > Sometimes it core dumps on the very first iteration, sometimes it takes 100. I've fixed the problem on my server by configuring amavisd to use /usr/bin/file instead, which is not a problem because the server does not handle a high volume. Maybe someone who is more skilled debugging perl core dumps than I would like to have a crack at fixing the problem? cheers, Rob Urban -- snip -- base-64 encoded version of "p005": > begin-base64 644 p005.gz > H4sICEQHXVcCA3AwMDUAbZDLbsJADEXX9Vd4kQUonQzDs4pUiaRQGimBqoQPSIKTjDokiBkW/H3z > aFUVdWPJ9rm+tq3dmSp/v7IADprwBbP6SAVVmNcXPClTV2w8EpPRQswc2BNhacxZu5wX0pTX1Mnq > E4/CBuMdzKXWV9JcTJ+6CUcyiVTaAWCMNfNuSZbxKPmkXCpy6ossHuLyilFyw/EcxcQVU3c2x8Zy > DrZt3yvu4ZkrRj28XCITk8cF2l1sUj/YovuM1sA/BOFqyFNZAYaB/6eoZAoYhfGuY7sTAGPvY7OO > 24ImlQN7Db3Nvk2ZSS4FmUbeI0Owf3s/j8v+wXDrReveob0H8N2L374XCbZD1xrokpRCysoaLavt > NiL4At4kOzWdAQAA >
Re: windowmaker diff
On Wednesday, June 8, 2016 16:49 CEST, David Coppa wrote: > On Tue, 07 Jun 2016, Reyk Floeter wrote: > > > On Tue, Jun 07, 2016 at 08:48:26PM +0200, David Coppa wrote: > > > On Tue, Jun 7, 2016 at 8:35 PM, Reyk Floeter wrote: > > > > On Tue, Jun 07, 2016 at 08:23:01PM +0200, Reyk Floeter wrote: > > > >> On Tue, Jun 07, 2016 at 05:27:51PM +0200, David Coppa wrote: > > > >> > Hi Reyk, > > > >> > > > > >> > Since I've just found (by reading your howto about the Lenovo > > > >> > Thinkpad > > > >> > X1 Carbon) that you're using Window Maker... > > > >> > Could you please test the attached diff that updates our windowmaker > > > >> > port to its latest release, 0.95.7? > > > >> > > > > >> > Ciao! > > > >> > David > > > >> > > > >> Sure! > > > >> > > > > > > > > It compiles, installed, but one annoying thing is that window-resize > > > > seems to be broken ('Maximize active window' gives strange results and > > > > covers the dock) > > > > > > > > There might be other issues. > > > > > > Can you try starting from scratch with a clean environment? > > > > > > I mean: > > > > > > $ mv ~/GNUstep ~/GNUstep.old > > > > > > > Yes, that helped. > > > > A few other things have changed (like the default menu, in a negative > > way hiding xterm and Firefox somewhere in a submenu). Maximization > > now includes the space behind the dock - I have to get used to this. > > > > I also don't like the new "jumping" icons. They should focus on > > modernizing wmaker (eg. by adding support for DPI scaling) instead > > bloating it with gimmicks (11 different ways of maximization?!). > > But there is a chance that something will happen now ;) > > > > After all, I think the ports update is OK. > > Widening the audience to ports@ > > Here's a new version of the diff. > I've stolen some useful patches from FreeBSD, and fixed a time_t > issue. > > Maybe it's the right time to update our prehistoric port of Window > Maker... > > If you're a wmaker user, please test! finally got around to it, I use it all day with GNUstep GWorkspace as desktop. So far, works well for me on amd64. OK sebastia@ > > Ciao, > David > > Index: Makefile > === > RCS file: /cvs/ports/x11/windowmaker/Makefile,v > retrieving revision 1.93 > diff -u -p -u -p -r1.93 Makefile > --- Makefile 11 Mar 2016 20:28:33 - 1.93 > +++ Makefile 8 Jun 2016 14:16:50 - > @@ -3,14 +3,15 @@ > COMMENT-main=window manager that emulates NEXTSTEP(tm) > COMMENT-lang=language subpackage for Window Maker > > -V= 0.92.0 > +V = 0.95.7 > DISTNAME=WindowMaker-${V} > PKGNAME-main=${DISTNAME:L} > -REVISION-main= 56 > FULLPKGNAME-lang=windowmaker-lang-${V} > -REVISION-lang= 5 > > -SHARED_LIBS += wraster 5.0 # 4.0 > +SHARED_LIBS += WINGs 0.0 # 2.1 > +SHARED_LIBS += WMaker0.0 # 1.1 > +SHARED_LIBS += WUtil 0.0 # 3.0 > +SHARED_LIBS += wraster 5.1 # 4.1 > > CATEGORIES= x11 x11/windowmaker > > @@ -19,22 +20,27 @@ HOMEPAGE= http://windowmaker.org/ > # GPLv2+ > PERMIT_PACKAGE_CDROM=Yes > > -WANTLIB += X11 Xext Xft Xinerama Xpm Xrender c expat fontconfig > -WANTLIB += freetype jpeg m png pthread-stubs tiff gif xcb z > +WANTLIB += ICE SM X11 Xmu Xext Xft Xinerama Xpm Xrandr Xrender > +WANTLIB += Xt c expat fontconfig freetype gif jpeg kvm m png > +WANTLIB += pthread pthread-stubs tiff xcb z > > MASTER_SITES=${MASTER_SITE_WINDOWMAKER:=source/release/} > -EXTRACT_SUFX=.tar.bz2 > > NO_TEST= Yes > SEPARATE_BUILD= Yes > CONFIGURE_STYLE= gnu > -CONFIGURE_ARGS+= --enable-static \ > +CONFIGURE_ARGS+= --disable-magick \ > + --disable-webp \ > + --enable-randr \ > + --enable-static \ > --enable-xinerama \ > - --with-datadir="${LOCALBASE}/share/WindowMaker" \ > - --with-nlsdir="${LOCALBASE}/share/locale" > + --localedir="${LOCALBASE}/share/locale" > > -CONFIGURE_ENV= CPPFLAGS="-I${LOCALBASE}/include" > LIBS="-L${LOCALBASE}/lib" \ > - LINGUAS='be bg bs ca cs da de el es et fi fr gl hr hu > it ja ko ms nl no pl pt ro ru sk sv tr zh_CN zh_TW' > +CONFIGURE_ENV= CPPFLAGS="-I${X11BASE}/include > -I${LOCALBASE}/include" \ > + LIBS="-L${X11BASE}/lib -L${LOCALBASE}/lib" \ > + LINGUAS='*' \ > + ac_cv_lib_exif_exif_data_new_from_file=no \ > + wm_cv_func_secure_getenv=no > > MULTI_PACKAGES= -main -lang > > @@ -43,27 +49,24 @@ LIB_DEPENDS= graphics/tiff \ > graphics/png \ > g
Apparent missing dep in print/hplip
Twenty seconds of search queries suggests that it's a missing Qt4 Python library. Here's an apparently similar issue from Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=525556 Thanks, Mike mike:/home/mike$ hp-setup HP Linux Imaging and Printing System (ver. 3.16.5) Printer/Fax Setup Utility ver. 9.0 Copyright (c) 2001-15 HP Development Company, LP This software comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to distribute it under certain conditions. See COPYING file for more details. Traceback (most recent call last): File "/usr/local/bin/hp-setup", line 313, in ui = import_module(ui_package + ".setupdialog") File "/usr/local/lib/python2.7/importlib/__init__.py", line 37, in import_module __import__(name) ImportError: No module named ui4.setupdialog
Re: pledge par2cmdline
On Mon, Jun 13, 2016 at 07:35:23PM +0200, Sebastien Marie wrote: > On Sun, Jun 12, 2016 at 09:17:27PM +0100, Mikolaj Kucharski wrote: > > Will you be driving this to get merged upstream Carlin? > > Trying to push it upstream would be a good thing. I can fill pull request if you want. However if Carlin wrote the diffs I think it would be better if he drives this, but I don't mind doing it. Just let me know. -- best regards q#
Re: NEW: devel/py-py-backports-shutil-get-terminal-size and py-backports
On Mon, 13 Jun 2016 18:12:17 +0500, Alexandr Shadchin wrote: > > Is there any reason to hardcode > > # Only Python 2 > > MODPY_VERSION = ${MODPY_DEFAULT_VERSION_2} > > because as there is no FLAVOR it will be automatically py2? > > > > Copy-paste from py-backports-ssl-match-hostname. > For me it does not make sense. I think that it can be removed. I think they do no harm but removing them would bring consistency. Anyway, you're the maintainer, do as you prefer :) ok danj@
Re: add python 3 flavor for www/py-django/stable
On Sun, 12 Jun 2016 12:10:20 +0200, Remi Pointel wrote: > I understand the feedbacks, so this is a new diff to have the > possibility to install django/stable with python 3. > > It will be possible to install: > - py3-django-lts with py-django > - py3-django with py-django > > I added the conflict between py3-django-lts and py3-django. > > Ok? ok danj@ Can you add a TEST_DEPENDS on devel/py-mock on py-django (it's only needed for py2), thanks. Cheers, Daniel
Re: UPDATE: devel/py-pexpect 4.1.0
On Mon, 13 Jun 2016 18:33:42 +0500, Alexandr Shadchin wrote: > Hi, > > This diff updates py-pexpect to the latest release. > Tested on amd64. Take maintainership. > > Comments ? OK ? ok danj@ > -MAKE_ENV += PYTHONPATH=${WRKSRC} > SPHINXBUILD=${LOCALBASE}/bin/sphinx-build${MODPY_BIN_SUFFIX} > +MAKE_ENV += PYTHONPATH=${WRKSRC} > > post-build: > - cd ${WRKSRC}/doc && ${SETENV} ${MAKE_ENV} ${MAKE_PROGRAM} html > + cd ${WRKSRC}/doc && ${SETENV} ${MAKE_ENV} ${MAKE_PROGRAM} html \ > + SPHINXBUILD=${LOCALBASE}/bin/sphinx-build${MODPY_BIN_SUFFIX} Out of curiosity, what does that change?
Re: pledge par2cmdline
On Sun, Jun 12, 2016 at 09:17:27PM +0100, Mikolaj Kucharski wrote: > Hi, > > This looks good to me. I did make test on what is currently in the ports > tree (below diff) and tested Carlin's changes agains what is on GitHub > and patches apply cleanly (as there ware no a lot of changes between > 0.6.14 release and current master). Thanks to both, diff commited. > Will you be driving this to get merged upstream Carlin? Trying to push it upstream would be a good thing. Regards. -- Sebastien Marie
UPDATE: devel/py-pexpect 4.1.0
Hi, This diff updates py-pexpect to the latest release. Tested on amd64. Take maintainership. Comments ? OK ? -- Alexandr Shadchin Index: Makefile === RCS file: /cvs/ports/devel/py-pexpect/Makefile,v retrieving revision 1.23 diff -u -p -w -r1.23 Makefile --- Makefile17 Apr 2016 07:37:29 - 1.23 +++ Makefile13 Jun 2016 13:31:35 - @@ -2,14 +2,15 @@ COMMENT= pure Python Expect-like module -MODPY_EGG_VERSION= 4.0.1 +MODPY_EGG_VERSION= 4.1.0 DISTNAME= pexpect-${MODPY_EGG_VERSION} PKGNAME= py-${DISTNAME} CATEGORIES=devel -REVISION= 2 HOMEPAGE= http://pexpect.sourceforge.net/ +MAINTAINER = Alexandr Shadchin + # ISC PERMIT_PACKAGE_CDROM= Yes @@ -32,7 +33,7 @@ MODPY_ADJ_FILES = tests/fakessh/ssh EXAMPLESDIR= ${PREFIX}/share/examples/${MODPY_PY_PREFIX}pexpect DOCDIR=${PREFIX}/share/doc/${MODPY_PY_PREFIX}pexpect -MAKE_ENV +=PYTHONPATH=${WRKSRC} SPHINXBUILD=${LOCALBASE}/bin/sphinx-build${MODPY_BIN_SUFFIX} +MAKE_ENV +=PYTHONPATH=${WRKSRC} pre-build: ${SUBST_CMD} ${WRKSRC}/pexpect/replwrap.py @@ -42,7 +43,8 @@ pre-build: ${SUBST_CMD} ${WRKSRC}/tests/test_run.py post-build: - cd ${WRKSRC}/doc && ${SETENV} ${MAKE_ENV} ${MAKE_PROGRAM} html + cd ${WRKSRC}/doc && ${SETENV} ${MAKE_ENV} ${MAKE_PROGRAM} html \ + SPHINXBUILD=${LOCALBASE}/bin/sphinx-build${MODPY_BIN_SUFFIX} post-install: ${INSTALL_DATA_DIR} ${EXAMPLESDIR} Index: distinfo === RCS file: /cvs/ports/devel/py-pexpect/distinfo,v retrieving revision 1.8 diff -u -p -w -r1.8 distinfo --- distinfo7 Jan 2016 17:16:20 - 1.8 +++ distinfo13 Jun 2016 13:31:35 - @@ -1,2 +1,2 @@ -SHA256 (pexpect-4.0.1.tar.gz) = IyeV68qvLhIDltu6o6Ep7aUXV+6q4ZEVWPTvjuQU/Gw= -SIZE (pexpect-4.0.1.tar.gz) = 143657 +SHA256 (pexpect-4.1.0.tar.gz) = w4HGDxmHNVtl348Ion9CiDGRTIqBCRvRd4rDNvovJ+c= +SIZE (pexpect-4.1.0.tar.gz) = 140294 Index: patches/patch-doc_Makefile === RCS file: patches/patch-doc_Makefile diff -N patches/patch-doc_Makefile --- patches/patch-doc_Makefile 16 Apr 2016 19:06:26 - 1.1 +++ /dev/null 1 Jan 1970 00:00:00 - @@ -1,12 +0,0 @@ -$OpenBSD: patch-doc_Makefile,v 1.1 2016/04/16 19:06:26 shadchin Exp $ doc/Makefile.orig Fri Apr 15 20:41:00 2016 -+++ doc/Makefile Fri Apr 15 20:41:10 2016 -@@ -3,7 +3,7 @@ - - # You can set these variables from the command line. - SPHINXOPTS= --SPHINXBUILD = sphinx-build -+SPHINXBUILD ?= sphinx-build - PAPER = - BUILDDIR = _build - Index: patches/patch-tests_test_maxcanon_py === RCS file: patches/patch-tests_test_maxcanon_py diff -N patches/patch-tests_test_maxcanon_py --- patches/patch-tests_test_maxcanon_py7 Jan 2016 17:16:21 - 1.1 +++ /dev/null 1 Jan 1970 00:00:00 - @@ -1,13 +0,0 @@ -$OpenBSD: patch-tests_test_maxcanon_py,v 1.1 2016/01/07 17:16:21 shadchin Exp $ -https://github.com/pexpect/pexpect/issues/283 - tests/test_maxcanon.py.origThu Dec 31 22:44:43 2015 -+++ tests/test_maxcanon.py Thu Dec 31 22:45:00 2015 -@@ -97,6 +97,7 @@ class TestCaseCanon(PexpectTestCase.PexpectTestCase): - assert not child.isalive() - assert child.exitstatus == 0 - -+@pytest.mark.xfail() - @pytest.mark.skipif( - sys.platform.lower().startswith('freebsd'), - reason='os.write to BLOCK indefinitely on FreeBSD in this case' Index: pkg/PLIST === RCS file: /cvs/ports/devel/py-pexpect/pkg/PLIST,v retrieving revision 1.8 diff -u -p -w -r1.8 PLIST --- pkg/PLIST 15 Apr 2016 08:53:49 - 1.8 +++ pkg/PLIST 13 Jun 2016 13:31:35 - @@ -20,7 +20,8 @@ lib/python${MODPY_VERSION}/site-packages lib/python${MODPY_VERSION}/site-packages/pexpect/${MODPY_PYCACHE}screen.${MODPY_PYC_MAGIC_TAG}pyc lib/python${MODPY_VERSION}/site-packages/pexpect/${MODPY_PYCACHE}spawnbase.${MODPY_PYC_MAGIC_TAG}pyc lib/python${MODPY_VERSION}/site-packages/pexpect/${MODPY_PYCACHE}utils.${MODPY_PYC_MAGIC_TAG}pyc -${MODPY_COMMENT}lib/python${MODPY_VERSION}/site-packages/pexpect/async.py +lib/python${MODPY_VERSION}/site-packages/pexpect/async.py +lib/python${MODPY_VERSION}/site-packages/pexpect/bashrc.sh lib/python${MODPY_VERSION}/site-packages/pexpect/exceptions.py lib/python${MODPY_VERSION}/site-packages/pexpect/expect.py lib/python${MODPY_VERSION}/site-packages/pexpect/fdpexpect.py
Re: NEW: devel/py-py-backports-shutil-get-terminal-size and py-backports
On Tue, Jun 07, 2016 at 11:26:36PM +0200, Daniel Jakots wrote: > On Tue, 31 May 2016 15:11:32 +0500, Alexandr Shadchin > wrote: > > > Hi, > > > > ok to import? > > py-backports-shutil-get-terminal-size need for update ipython. > > > > py-backports: > > Namespace for backported Python features. > > > > py-backports-shutil-get-terminal-size: > > A backport of the get_terminal_size function from Python 3.3’s shutil. > > > > Unlike the original version it is written in pure Python rather than > > C, so it might be a tiny bit slower. > > > > Is there any reason to hardcode > # Only Python 2 > MODPY_VERSION = ${MODPY_DEFAULT_VERSION_2} > because as there is no FLAVOR it will be automatically py2? > Copy-paste from py-backports-ssl-match-hostname. For me it does not make sense. I think that it can be removed. -- Alexandr Shadchin
Re: add python 3 flavor for www/py-django/stable
On Sun, Jun 12, 2016 at 12:10:20PM +0200, Remi Pointel wrote: > Hi, > > I understand the feedbacks, so this is a new diff to have the possibility to > install django/stable with python 3. > > It will be possible to install: > - py3-django-lts with py-django > - py3-django with py-django > > I added the conflict between py3-django-lts and py3-django. > > Ok? > > Cheers, > > Remi. ok shadchin@ -- Alexandr Shadchin
Re: fix py-mock test_depends
On Sun, Jun 05, 2016 at 11:22:55PM +0200, Daniel Jakots wrote: > Hi, > > While reviewing shadchin@'s diff (for funcsigs): > > ===> Regression tests for py3-mock-1.3.0 > cd /usr/ports/pobj/py-mock-1.3.0-python3/mock-1.3.0 && > /usr/local/bin/python3.4 -m unittest discover > EEE > == > ERROR: mock.tests.testcallable (unittest.loader._FailedTest) > -- > Traceback (most recent call last): > File "/usr/local/lib/python3.4/unittest/case.py", line 58, in > testPartExecutor > yield > File "/usr/local/lib/python3.4/unittest/case.py", line 580, in run > testMethod() > File "/usr/local/lib/python3.4/unittest/loader.py", line 33, in testFailure > raise self._exception > ImportError: Failed to import test module: mock.tests.testcallable > Traceback (most recent call last): > File "/usr/local/lib/python3.4/unittest/loader.py", line 323, in _find_tests > module = self._get_module_from_name(name) > File "/usr/local/lib/python3.4/unittest/loader.py", line 301, in > _get_module_from_name > __import__(name) > File "/usr/ports/pobj/py-mock-1.3.0-python3/mock-1.3.0/mock/__init__.py", > line 2, in > import mock.mock as _mock > File "/usr/ports/pobj/py-mock-1.3.0-python3/mock-1.3.0/mock/mock.py", line > 67, in > import six > ImportError: No module named 'six' > > Quick fix: > Comments? OK? > ok shadchin@ > Index: Makefile > === > RCS file: /cvs/ports/devel/py-mock/Makefile,v > retrieving revision 1.8 > diff -u -p -r1.8 Makefile > --- Makefile 26 Dec 2015 20:48:11 - 1.8 > +++ Makefile 5 Jun 2016 21:19:00 - > @@ -18,6 +18,7 @@ MODULES=lang/python > BUILD_DEPENDS= devel/py-pbr${MODPY_FLAVOR} > RUN_DEPENDS= devel/py-six${MODPY_FLAVOR} \ > devel/py-pbr${MODPY_FLAVOR} > +TEST_DEPENDS=${RUN_DEPENDS} > > FLAVORS =python3 > FLAVOR ?= > -- Alexandr Shadchin