[UPDATE] python 3.7

[Remi Pointel]

Hi,

this is the diff to update python 3 to 3.7.7.

Could it be tested in a bulk build please?

Ok?

Cheers,

Remi.
Index: Makefile
===
RCS file: /cvs/ports/lang/python/3.7/Makefile,v
retrieving revision 1.15
diff -u -p -u -p -r1.15 Makefile
--- Makefile	11 Feb 2020 20:43:11 -	1.15
+++ Makefile	23 Mar 2020 06:44:29 -
@@ -6,13 +6,9 @@
 # Python itself.
 
 VERSION =		3.7
-PATCHLEVEL =		.6
+PATCHLEVEL =		.7
 SHARED_LIBS =		python3.7m 0.0
 VERSION_SPEC =		>=3.7,<3.8
-
-REVISION =		0
-REVISION-main =		1
-REVISION-idle =		1
 
 CONFIGURE_ARGS +=	--with-ensurepip=no
 CONFIGURE_ARGS +=	--enable-loadable-sqlite-extensions
Index: distinfo
===
RCS file: /cvs/ports/lang/python/3.7/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo	28 Dec 2019 18:35:39 -	1.4
+++ distinfo	23 Mar 2020 06:44:29 -
@@ -1,2 +1,2 @@
-SHA256 (Python-3.7.6.tgz) = ru5oHCNa0zavEW8Iq2VjNhoMgcU3BywbMJ1uQFCqIRQ=
-SIZE (Python-3.7.6.tgz) = 23148187
+SHA256 (Python-3.7.7.tgz) = jIvpHNJkihoMJR8E6gu0wqVXD+ucReqqIkHHhVhbR1o=
+SIZE (Python-3.7.7.tgz) = 23161893
Index: pkg/PLIST-idle
===
RCS file: /cvs/ports/lang/python/3.7/pkg/PLIST-idle,v
retrieving revision 1.7
diff -u -p -u -p -r1.7 PLIST-idle
--- pkg/PLIST-idle	11 Feb 2020 20:43:12 -	1.7
+++ pkg/PLIST-idle	23 Mar 2020 06:44:29 -
@@ -3,12 +3,12 @@
 @option is-branch
 @conflict python-idle-${VERSION_SPEC}
 @conflict python-idle->=3.2,<3.7
+@conflict python->=3.6,<3.6.8p0
 @pkgpath lang/python/3.2,-idle
 @pkgpath lang/python/3.3,-idle
 @pkgpath lang/python/3.4,-idle
 @pkgpath lang/python/3.5,-idle
 @pkgpath lang/python/3.6,-idle
-@conflict python->=3.6,<3.6.8p0
 bin/idle3
 bin/idle3.7
 lib/python3.7/idlelib/
Index: pkg/PLIST-main
===
RCS file: /cvs/ports/lang/python/3.7/pkg/PLIST-main,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 PLIST-main
--- pkg/PLIST-main	11 Feb 2020 20:43:12 -	1.11
+++ pkg/PLIST-main	23 Mar 2020 06:44:29 -
@@ -2378,9 +2378,9 @@ lib/python3.7/lib-dynload/
 @so lib/python3.7/lib-dynload/zlib.so
 lib/python3.7/lib2to3/
 lib/python3.7/lib2to3/Grammar.txt
-lib/python3.7/lib2to3/Grammar3.7.6.final.0.pickle
+lib/python3.7/lib2to3/Grammar3.7.7.final.0.pickle
 lib/python3.7/lib2to3/PatternGrammar.txt
-lib/python3.7/lib2to3/PatternGrammar3.7.6.final.0.pickle
+lib/python3.7/lib2to3/PatternGrammar3.7.7.final.0.pickle
 lib/python3.7/lib2to3/__init__.py
 lib/python3.7/lib2to3/__main__.py
 lib/python3.7/lib2to3/__pycache__/

Re: update: devel/py-wcwidth (cascading devel/{py-prompt_toolkit,ipython} dropping py2 support)

[Bjorn Ketelaars]

On Sun 22/03/2020 18:39, Paco Esteban wrote:
> Ping ?
> 
> (This needs bket's diff discussed on another thread:
> https://marc.info/?l=openbsd-ports&m=158464110714289&w=2)
> 
> Find attached the (still relevant) parts for convenience.

Output of 'show-reverse-deps devel/py-wcwidth' shows that there are
other consumers that still need the python2-flavor. At least:
- devel/bpython
- devel/py-curtsies

devel/kf5/ktextigwidgets: remove python module usage

[Klemens Nanni]

Same as kconfwidgets: configures and builds without any python available,
runtime shows how traces of python usage.

Feedback? OK?


Index: Makefile
===
RCS file: /cvs/ports/devel/kf5/ktextwidgets/Makefile,v
retrieving revision 1.10
diff -u -p -r1.10 Makefile
--- Makefile20 May 2019 22:15:06 -  1.10
+++ Makefile23 Mar 2020 01:20:44 -
@@ -2,6 +2,7 @@
 
 COMMENT =  text editing widgets for KDE
 DISTNAME = ktextwidgets-${VERSION}
+REVISION = 0
 
 SHARED_LIBS =  KF5TextWidgets  4.2
 
@@ -10,8 +11,6 @@ WANTLIB += KF5ConfigCore KF5ConfigGui KF
 WANTLIB += KF5I18n KF5IconThemes KF5Service KF5SonnetCore KF5SonnetUi
 WANTLIB += KF5WidgetsAddons KF5WindowSystem Qt5Core Qt5Gui Qt5Widgets
 WANTLIB += Qt5Xml m
-
-MODULES += lang/python
 
 BUILD_DEPENDS =devel/gettext,-tools
 

devel/kf5/kconfigwidgets: remove python module usage

[Klemens Nanni]

No python is tested for during configure, build passes without python
installed and PLIST has no traces of any usage;  tests fail regardless
of my changes.

Feedback? OK?


Index: Makefile
===
RCS file: /cvs/ports/devel/kf5/kconfigwidgets/Makefile,v
retrieving revision 1.10
diff -u -p -r1.10 Makefile
--- Makefile20 May 2019 22:15:04 -  1.10
+++ Makefile23 Mar 2020 01:10:14 -
@@ -2,14 +2,13 @@
 
 COMMENT =  widgets for KDE configuration dialogs
 DISTNAME = kconfigwidgets-${VERSION}
+REVISION = 0
 
 SHARED_LIBS =  KF5ConfigWidgets3.3
 
 WANTLIB += ${COMPILER_LIBCXX} KF5Auth KF5Codecs KF5ConfigCore
 WANTLIB += KF5ConfigGui KF5CoreAddons KF5GuiAddons KF5I18n KF5WidgetsAddons
 WANTLIB += Qt5Core Qt5DBus Qt5Gui Qt5Widgets Qt5Xml m
-
-MODULES =  lang/python
 
 BUILD_DEPENDS =devel/gettext,-tools \
devel/kf5/kdoctools

devel/kf5/kfilemetadata: switch to python 3 (TDEP only)

[Klemens Nanni]

`make configure' says

-- Found PythonInterp: /usr/local/bin/python3.7 (found version "3.7.6") 

-- The following OPTIONAL packages have been found:
   ...
 * PythonInterp, Python Interpreter,    

   Python interpreter is needed to execute test for external extractors or 
writers


The port builds fine without any python interpreter available, I ran
`make test' with and without python 3 available and only the two
mentioned test cases would pass or fail respectively (all else passed).

Feedback? OK?



Index: Makefile
===
RCS file: /cvs/ports/devel/kf5/kfilemetadata/Makefile,v
retrieving revision 1.16
diff -u -p -r1.16 Makefile
--- Makefile19 Jun 2019 19:45:59 -  1.16
+++ Makefile23 Mar 2020 01:04:24 -
@@ -5,6 +5,8 @@ COMMENT-locale =localization files of kf
 DISTNAME = kfilemetadata-${VERSION}
 PKGNAME-main = kf5-kfilemetadata-${VERSION}
 PKGNAME-locale =kf5-kfilemetadata-l10n-${VERSION}
+REVISION-locale =  0
+REVISION-main =0
 
 SHARED_LIBS =  KF5FileMetaData 5.3# 5.20
 
@@ -19,6 +21,10 @@ WANTLIB += avcodec avformat avutil epub 
 WANTLIB += tag
 
 MODULES += lang/python
+MODPY_VERSION =${MODPY_DEFAULT_VERSION_3}
+# required only for externalextractortest and externalwritertest
+MODPY_BUILDDEP =No
+MODPY_RUNDEP = No
 
 BUILD_DEPENDS =devel/gettext,-tools
 

devel/kf5/kservice: switch to python 3

[Klemens Nanni]

kservice builds fine with python installed but all /usr/local/bin/python*
executables removed, there is nothing python related in PLIST and I fail
to spot any further use of python.

`make configure' also says this at the end:
-- Generating done
CMake Warning:
  Manually-specified variables were not used by the project:

  CMAKE_DISABLE_FIND_PACKAGE_PythonModuleGeneration

I fail to see where this is passed/defined, however.

So perhaps the lang/python usage in this port is accidential since its
import?  No python related changes afterwards.


Feedback? OK?


Index: Makefile
===
RCS file: /cvs/ports/devel/kf5/kservice/Makefile,v
retrieving revision 1.13
diff -u -p -r1.13 Makefile
--- Makefile20 May 2019 22:15:06 -  1.13
+++ Makefile23 Mar 2020 00:32:03 -
@@ -2,13 +2,12 @@
 
 COMMENT =  plugin framework for KDE desktop services
 DISTNAME = kservice-${VERSION}
+REVISION = 0
 
 SHARED_LIBS =  KF5Service  4.3
 
 WANTLIB += ${COMPILER_LIBCXX} KF5ConfigCore KF5CoreAddons KF5Crash
 WANTLIB += KF5DBusAddons KF5I18n Qt5Core Qt5DBus Qt5Xml c m
-
-MODULES += lang/python
 
 BUILD_DEPENDS =devel/bison \
devel/gettext,-tools \

Re: devel/kf5/ki18n: switch to python 3

[Klemens Nanni]

On Mon, Mar 23, 2020 at 01:13:44AM +0100, Klemens Nanni wrote:
> Step one in removing Python 2 from those KDE ports I happen to use:
> 
> Exactly one python script in this port gets installed, it is used as the
> comment in Makefile says.
> 
> cmake/build-tsfiles.cmake has this:
> 
>   69  list(APPEND commands
>   70  COMMAND ${PYTHON_EXECUTABLE}
>   71  -B
>   72  ${_ki18n_pmap_compile_script}
>   73  ${PO_DIR}/${pmap_file}
>   74  ${pmapc_file}
>   75  )
> 
> where _ki18n_pmap_compile_script is the absolute path to this script.
> 
> Feedback? OK?
Wrong diff, sorry.  While python is not used during build, cmake requires
it to detect the used version, so MODPY_BUILDDEP must be "Yes".


Index: Makefile
===
RCS file: /cvs/ports/devel/kf5/ki18n/Makefile,v
retrieving revision 1.11
diff -u -p -r1.11 Makefile
--- Makefile20 May 2019 22:15:05 -  1.11
+++ Makefile23 Mar 2020 00:25:52 -
@@ -2,13 +2,14 @@
 
 COMMENT =  KDE Gettext-based UI text internationalization
 DISTNAME = ki18n-${VERSION}
-REVISION = 0
+REVISION = 1
 
 SHARED_LIBS =  KF5I18n 4.3
 
 WANTLIB += ${COMPILER_LIBCXX} Qt5Core Qt5Network Qt5Qml intl m
 
 MODULES += lang/python
+MODPY_VERSION =${MODPY_DEFAULT_VERSION_3}
 
 BUILD_DEPENDS =devel/gettext,-tools \
x11/qt5/qtscript
@@ -20,6 +21,6 @@ LIB_DEPENDS = devel/gettext,-runtime \
 PORTHOME = ${WRKDIR}
 
 # Calling compileall.py isn't needed, the ts-pmap-compile script
-# gets called with "${PYTHON} -B ...".
+# gets called with "${PYTHON} -B ..." in cmake/build-tsfiles.cmake.
 
 .include 
Index: pkg/PLIST
===
RCS file: /cvs/ports/devel/kf5/ki18n/pkg/PLIST,v
retrieving revision 1.6
diff -u -p -r1.6 PLIST
--- pkg/PLIST   30 Mar 2019 09:13:52 -  1.6
+++ pkg/PLIST   22 Mar 2020 21:20:18 -
@@ -25,7 +25,7 @@ lib/cmake/KF5I18n/kf5i18nuic.cmake
 lib/cmake/KF5I18n/ts-pmap-compile.py
 @lib lib/libKF5I18n.so.${LIBKF5I18n_VERSION}
 lib/qt5/plugins/kf5/
-lib/qt5/plugins/kf5/ktranscript.so
+@so lib/qt5/plugins/kf5/ktranscript.so
 share/kf5/
 share/kf5/mkspecs/
 share/kf5/mkspecs/qt_KI18n.pri

devel/kf5/ki18n: switch to python 3

[Klemens Nanni]

Step one in removing Python 2 from those KDE ports I happen to use:

Exactly one python script in this port gets installed, it is used as the
comment in Makefile says.

cmake/build-tsfiles.cmake has this:

69  list(APPEND commands
70  COMMAND ${PYTHON_EXECUTABLE}
71  -B
72  ${_ki18n_pmap_compile_script}
73  ${PO_DIR}/${pmap_file}
74  ${pmapc_file}
75  )

where _ki18n_pmap_compile_script is the absolute path to this script.

Feedback? OK?


Index: Makefile
===
RCS file: /cvs/ports/devel/kf5/ki18n/Makefile,v
retrieving revision 1.11
diff -u -p -r1.11 Makefile
--- Makefile20 May 2019 22:15:05 -  1.11
+++ Makefile22 Mar 2020 22:25:59 -
@@ -2,13 +2,15 @@
 
 COMMENT =  KDE Gettext-based UI text internationalization
 DISTNAME = ki18n-${VERSION}
-REVISION = 0
+REVISION = 1
 
 SHARED_LIBS =  KF5I18n 4.3
 
 WANTLIB += ${COMPILER_LIBCXX} Qt5Core Qt5Network Qt5Qml intl m
 
 MODULES += lang/python
+MODPY_VERSION =${MODPY_DEFAULT_VERSION_3}
+MODPY_BUILDDEP =No
 
 BUILD_DEPENDS =devel/gettext,-tools \
x11/qt5/qtscript
@@ -20,6 +22,6 @@ LIB_DEPENDS = devel/gettext,-runtime \
 PORTHOME = ${WRKDIR}
 
 # Calling compileall.py isn't needed, the ts-pmap-compile script
-# gets called with "${PYTHON} -B ...".
+# gets called with "${PYTHON} -B ..." in cmake/build-tsfiles.cmake.
 
 .include 
Index: pkg/PLIST
===
RCS file: /cvs/ports/devel/kf5/ki18n/pkg/PLIST,v
retrieving revision 1.6
diff -u -p -r1.6 PLIST
--- pkg/PLIST   30 Mar 2019 09:13:52 -  1.6
+++ pkg/PLIST   22 Mar 2020 21:20:18 -
@@ -25,7 +25,7 @@ lib/cmake/KF5I18n/kf5i18nuic.cmake
 lib/cmake/KF5I18n/ts-pmap-compile.py
 @lib lib/libKF5I18n.so.${LIBKF5I18n_VERSION}
 lib/qt5/plugins/kf5/
-lib/qt5/plugins/kf5/ktranscript.so
+@so lib/qt5/plugins/kf5/ktranscript.so
 share/kf5/
 share/kf5/mkspecs/
 share/kf5/mkspecs/qt_KI18n.pri

update: net/swirc 3.2.1 -> 3.2.3

[Markus Uhlin]

Hi,

An update for net/swirc (3.2.1 -> 3.2.3).

I didn't post 3.2.2 so refer to both regarding version history:
https://github.com/uhlin/swirc/blob/master/CHANGELOG.md#323---2020-03-22
https://github.com/uhlin/swirc/blob/master/CHANGELOG.md#322---2020-03-17

As mentioned before, code review and feedback welcome.

I attach a diff for the port.

Yours faithfully
Markus
diff -ru /tmp/ports/net/swirc/Makefile /usr/ports/net/swirc/Makefile
--- /tmp/ports/net/swirc/Makefile   Sun Mar 22 19:11:02 2020
+++ /usr/ports/net/swirc/Makefile   Sun Mar 22 19:17:09 2020
@@ -1,7 +1,7 @@
 # $OpenBSD: Makefile,v 1.5 2020/03/09 19:25:33 solene Exp $
 
 COMMENT =  curses icb and irc client
-DISTNAME = swirc-3.2.1
+DISTNAME = swirc-3.2.3
 CATEGORIES =   net
 HOMEPAGE = https://www.nifty-networks.net/swirc/
 
diff -ru /tmp/ports/net/swirc/distinfo /usr/ports/net/swirc/distinfo
--- /tmp/ports/net/swirc/distinfo   Sun Mar 22 19:11:02 2020
+++ /usr/ports/net/swirc/distinfo   Sun Mar 22 19:18:28 2020
@@ -1,2 +1,2 @@
-SHA256 (swirc-3.2.1.tgz) = RHCpbI4FcSwcKGJqUH6a+rJfVmcICX6bRorHLbu4tc4=
-SIZE (swirc-3.2.1.tgz) = 24
+SHA256 (swirc-3.2.3.tgz) = UuJzu8nqY4SEGUeFjWDLdAplvYDSJXx8qznZpOuRnwY=
+SIZE (swirc-3.2.3.tgz) = 230048

Update textproc/ruby-nokogiri

[Sebastian Reitenbach]

Hi,

this updates nokogiri to 1.10.8, an update of security/ruby-cms_scanner, and
ultimately security/wpscan would require it.
Tested all dependencies that they still build with updated version,
the only one that has tests, www/ruby-sanitize, has the same 5 tests failing
with the update.

OK?

cheers,
Sebastian

? nokogiri-1.10.8.diff
Index: Makefile
===
RCS file: /cvs/ports/textproc/ruby-nokogiri/Makefile,v
retrieving revision 1.35
diff -u -r1.35 Makefile
--- Makefile26 Sep 2019 06:35:09 -  1.35
+++ Makefile22 Mar 2020 22:37:21 -
@@ -2,7 +2,7 @@

 COMMENT=   HTML, XML, SAX and Reader parser

-DISTNAME=  nokogiri-1.10.4
+DISTNAME=  nokogiri-1.10.8
 CATEGORIES=textproc

 HOMEPAGE=  http://nokogiri.org/
Index: distinfo
===
RCS file: /cvs/ports/textproc/ruby-nokogiri/distinfo,v
retrieving revision 1.15
diff -u -r1.15 distinfo
--- distinfo26 Sep 2019 06:35:09 -  1.15
+++ distinfo22 Mar 2020 22:37:21 -
@@ -1,2 +1,2 @@
-SHA256 (nokogiri-1.10.4.gem) = +/kcSH186ysFkzhPF1hD+ZFSMidU9v8md0C/8DHOrVo=
-SIZE (nokogiri-1.10.4.gem) = 8983040
+SHA256 (nokogiri-1.10.8.gem) = CAa4sFQYULWcT1iMU8/0LvLabH+i8LHLy4PRzCGSKPo=
+SIZE (nokogiri-1.10.8.gem) = 9251328
Index: pkg/PLIST
===
RCS file: /cvs/ports/textproc/ruby-nokogiri/pkg/PLIST,v
retrieving revision 1.16
diff -u -r1.16 PLIST
--- pkg/PLIST   26 Sep 2019 06:35:09 -  1.16
+++ pkg/PLIST   22 Mar 2020 22:37:21 -
@@ -36,7 +36,7 @@
 ${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/html/sax/parser.rb
 ${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/html/sax/parser_context.rb
 ${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/html/sax/push_parser.rb
-${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/nokogiri.so
+@so ${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/nokogiri.so
 ${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/syntax_error.rb
 ${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/version.rb
 ${GEM_LIB}/gems/${DISTNAME}/lib/nokogiri/xml/
@@ -92,7 +92,7 @@
 
${GEM_LIB}/gems/${DISTNAME}/patches/libxml2/0001-Revert-Do-not-URI-escape-in-server-side-includes.patch
 
${GEM_LIB}/gems/${DISTNAME}/patches/libxml2/0002-Remove-script-macro-support.patch
 
${GEM_LIB}/gems/${DISTNAME}/patches/libxml2/0003-Update-entities-to-remove-handling-of-ssi.patch
-${GEM_LIB}/gems/${DISTNAME}/patches/libxslt/
-${GEM_LIB}/gems/${DISTNAME}/patches/libxslt/0001-Fix-security-framework-bypass.patch
+${GEM_LIB}/gems/${DISTNAME}/patches/libxml2/0004-libxml2.la-is-in-top_builddir.patch
+${GEM_LIB}/gems/${DISTNAME}/patches/libxml2/0005-Fix-infinite-loop-in-xmlStringLenDecodeEntities.patch
 ${GEM_LIB}/gems/${DISTNAME}/ports/
 ${GEM_LIB}/specifications/${DISTNAME}.gemspec

update net/dnscrypt-proxy 2.0.40

[Nam Nguyen]

Here is an update for net/dnscrypt-proxy 2.0.40, released on March 21,
2020. I tested basic usage on amd64. `make test' passes.

Changelog:
https://github.com/DNSCrypt/dnscrypt-proxy/blob/2.0.40/ChangeLog

Feedback and tests are welcome.

Index: Makefile
===
RCS file: /cvs/ports/net/dnscrypt-proxy/Makefile,v
retrieving revision 1.51
diff -u -p -r1.51 Makefile
--- Makefile1 Feb 2020 20:45:26 -   1.51
+++ Makefile22 Mar 2020 20:47:34 -
@@ -4,7 +4,7 @@ COMMENT =   flexible DNS proxy with suppor
 
 GH_ACCOUNT =   jedisct1
 GH_PROJECT =   dnscrypt-proxy
-GH_TAGNAME =   2.0.39
+GH_TAGNAME =   2.0.40
 
 CATEGORIES =   net
 
Index: distinfo
===
RCS file: /cvs/ports/net/dnscrypt-proxy/distinfo,v
retrieving revision 1.27
diff -u -p -r1.27 distinfo
--- distinfo1 Feb 2020 20:45:26 -   1.27
+++ distinfo22 Mar 2020 20:47:34 -
@@ -1,2 +1,2 @@
-SHA256 (dnscrypt-proxy-2.0.39.tar.gz) = 
yUPHTAiUu1EzZSnnM8o4Ed/9uRSlm5cHxjoyfyyP+DU=
-SIZE (dnscrypt-proxy-2.0.39.tar.gz) = 2814424
+SHA256 (dnscrypt-proxy-2.0.40.tar.gz) = 
F/lc+eOM6GSonc76MKTC4j14+zUUzTVnpg96DfwTPXg=
+SIZE (dnscrypt-proxy-2.0.40.tar.gz) = 2322600
Index: patches/patch-dnscrypt-proxy_example-dnscrypt-proxy_toml
===
RCS file: 
/cvs/ports/net/dnscrypt-proxy/patches/patch-dnscrypt-proxy_example-dnscrypt-proxy_toml,v
retrieving revision 1.12
diff -u -p -r1.12 patch-dnscrypt-proxy_example-dnscrypt-proxy_toml
--- patches/patch-dnscrypt-proxy_example-dnscrypt-proxy_toml1 Feb 2020 
20:45:26 -   1.12
+++ patches/patch-dnscrypt-proxy_example-dnscrypt-proxy_toml22 Mar 2020 
20:47:34 -
@@ -3,7 +3,7 @@ $OpenBSD: patch-dnscrypt-proxy_example-d
 Index: dnscrypt-proxy/example-dnscrypt-proxy.toml
 --- dnscrypt-proxy/example-dnscrypt-proxy.toml.orig
 +++ dnscrypt-proxy/example-dnscrypt-proxy.toml
-@@ -47,7 +47,7 @@ max_clients = 250
+@@ -49,7 +49,7 @@ max_clients = 250
  ## Note (2): this feature is not compatible with systemd socket activation.
  ## Note (3): when using -pidfile, the PID file directory must be writable by 
the new user
  
@@ -12,7 +12,7 @@ Index: dnscrypt-proxy/example-dnscrypt-p
  
  
  ## Require servers (from static + remote sources) to satisfy specific 
properties
-@@ -586,7 +586,7 @@ cache_neg_max_ttl = 600
+@@ -585,7 +585,7 @@ cache_neg_max_ttl = 600
  
[sources.'public-resolvers']
urls = 
['https://raw.githubusercontent.com/DNSCrypt/dnscrypt-resolvers/master/v2/public-resolvers.md',
 'https://download.dnscrypt.info/resolvers-list/v2/public-resolvers.md']
@@ -21,7 +21,7 @@ Index: dnscrypt-proxy/example-dnscrypt-p
minisign_key = 'RWQf6LRCGA9i53mlYecO4IzT51TGPpvWucNSCh1CBM0QTaLn73Y7GFO3'
prefix = ''
  
-@@ -594,7 +594,7 @@ cache_neg_max_ttl = 600
+@@ -593,7 +593,7 @@ cache_neg_max_ttl = 600
  
[sources.'relays']
urls = 
['https://raw.githubusercontent.com/DNSCrypt/dnscrypt-resolvers/master/v2/relays.md',
 'https://download.dnscrypt.info/resolvers-list/v2/relays.md']

[UPDATE] x11/py-xcbgen 1.14

[Matthieu Herrb]

Hi,

the patch below updates x11/py-xcbgen to version 1.14.

This is the ports part of the XCB 1.14 update I just sent to tech@.

Comments, ok ?

Index: Makefile
===
RCS file: /cvs/OpenBSD/ports/x11/py-xcbgen/Makefile,v
retrieving revision 1.5
diff -u -p -u -r1.5 Makefile
--- Makefile3 Nov 2019 11:42:11 -   1.5
+++ Makefile22 Mar 2020 19:08:05 -
@@ -3,10 +3,9 @@
 COMMENT =  X C Binding - protocol binding generator
 
 # please keep in sync with the version in proto/xcb-proto in Xenocara
-VERSION=   1.13
+VERSION=   1.14
 DISTNAME = xcb-proto-${VERSION}
 PKGNAME =  py-xcbgen-${VERSION}
-REVISION = 0
 
 CATEGORIES =   x11
 
@@ -17,9 +16,9 @@ MAINTAINER =  Matthieu Herrb https://xcb.freedesktop.org/dist/
+MASTER_SITES = ${MASTER_SITE_XORG:=proto/}
 
-EXTRACT_SUFX=   .tar.bz2
+EXTRACT_SUFX=   .tar.xz
 
 MODULES=   lang/python
 
Index: distinfo
===
RCS file: /cvs/OpenBSD/ports/x11/py-xcbgen/distinfo,v
retrieving revision 1.2
diff -u -p -u -r1.2 distinfo
--- distinfo11 Sep 2018 21:01:53 -  1.2
+++ distinfo22 Mar 2020 19:07:53 -
@@ -1,2 +1,2 @@
-SHA256 (xcb-proto-1.13.tar.bz2) = e5hyHmab6AKE6bv+qwLS0NVM0RFytyJx5Hov6HXiveE=
-SIZE (xcb-proto-1.13.tar.bz2) = 151981
+SHA256 (xcb-proto-1.14.tar.xz) = GGo86yb5tKAV9aRNzIFMkwM6X8OWhPNvHsx5g0QWpgU=
+SIZE (xcb-proto-1.14.tar.xz) = 143384

-- 
Matthieu Herrb

[new] archivers/laszip 3.4.3

[Landry Breuil]

Hi,

here's a port for https://laszip.org/, a library to compress point cloud
files (known as LAS) - will be useful for a geo/pdal update i'm working
on. Its source is also in geo/lastools, but there's no way yet to build
against this lib. It might be used by other geo/* bits too at some
point.

ok to import ?

Landry


laszip.tgz
Description: application/tar-gz

Re: UPDATE: Nextcloud-18.0.2

[Bjorn Ketelaars]

On Sun 22/03/2020 22:16, Gonzalo Rodriguez wrote:
> ping

Works for me on amd64. OK bket@

Re: update devel/py-curtsies (triggers drop py2 support for devel/bpython)

[Paco Esteban]

Ping ?

Diffs attached again for convenience.

On Sun, 15 Mar 2020, Paco Esteban wrote:

> Hi ports@,
> 
> This complements an update sent previously for devel/py-wcwidth.
> Here's an update for devel/py-curtsies to 0.3.1.  You can see the
> changelog here:
> 
> https://github.com/bpython/curtsies/blob/master/CHANGELOG.md
> 
> All tests pass.
> 
> I've enabled pytest for tests and dropped py2 support for this port.
> This triggers drop py2 support for its only consumer devel/bpython (also
> attached).  Maintainer (bcallah@) is ok with it.  Four test fail, same
> as before the update.  Tested it lightly, so more eyes on it are
> welcome.
> 
> Cheers,
> 
> -- 
> Paco Esteban.
> 0x5818130B8A6DBC03

-- 
Paco Esteban.
0x5818130B8A6DBC03
Index: Makefile
===
RCS file: /home/cvs/ports/devel/py-curtsies/Makefile,v
retrieving revision 1.9
diff -u -p -r1.9 Makefile
--- Makefile12 Jul 2019 20:45:47 -  1.9
+++ Makefile13 Mar 2020 14:32:06 -
@@ -2,10 +2,9 @@
 
 COMMENT =  curses-like terminal wrapper
 
-MODPY_EGG_VERSION =0.2.11
+MODPY_EGG_VERSION =0.3.1
 DISTNAME = curtsies-${MODPY_EGG_VERSION}
 PKGNAME =  py-${DISTNAME}
-REVISION = 1
 CATEGORIES =   devel
 
 # MIT
@@ -15,11 +14,16 @@ MODULES =   lang/python
 
 MODPY_SETUPTOOLS = Yes
 MODPY_PI = Yes
+MODPY_PYTEST = Yes
 
 FLAVORS =  python3
-FLAVOR ?=
+FLAVOR =   python3
 
 RUN_DEPENDS =  devel/py-blessings${MODPY_FLAVOR} \
devel/py-wcwidth${MODPY_FLAVOR}
+
+TEST_DEPENDS = devel/py-nose${MODPY_FLAVOR} \
+   devel/py-mock${MODPY_FLAVOR} \
+   devel/py-pyte${MODPY_FLAVOR}
 
 .include 
Index: distinfo
===
RCS file: /home/cvs/ports/devel/py-curtsies/distinfo,v
retrieving revision 1.2
diff -u -p -r1.2 distinfo
--- distinfo2 Nov 2016 22:38:09 -   1.2
+++ distinfo13 Mar 2020 14:08:27 -
@@ -1,2 +1,2 @@
-SHA256 (curtsies-0.2.11.tar.gz) = 7EY532VDV5RMIBq9DXkdnf06wnzk+Quhwpkarwevku4=
-SIZE (curtsies-0.2.11.tar.gz) = 46740
+SHA256 (curtsies-0.3.1.tar.gz) = sskTqBE8Q4LhoiFnnyM4E5sRKDnesWwA7oc+V6SzO9Q=
+SIZE (curtsies-0.3.1.tar.gz) = 47944
Index: pkg/PLIST
===
RCS file: /home/cvs/ports/devel/py-curtsies/pkg/PLIST,v
retrieving revision 1.2
diff -u -p -r1.2 PLIST
--- pkg/PLIST   17 Jun 2017 07:07:56 -  1.2
+++ pkg/PLIST   13 Mar 2020 14:34:07 -
@@ -1,4 +1,6 @@
 @comment $OpenBSD: PLIST,v 1.2 2017/06/17 07:07:56 jasper Exp $
+@conflict py-curtsies-*
+@pkgpath devel/py-curtsies
 lib/python${MODPY_VERSION}/site-packages/curtsies/
 
lib/python${MODPY_VERSION}/site-packages/curtsies-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/
 
lib/python${MODPY_VERSION}/site-packages/curtsies-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/PKG-INFO
Index: Makefile
===
RCS file: /home/cvs/ports/devel/bpython/Makefile,v
retrieving revision 1.26
diff -u -p -r1.26 Makefile
--- Makefile22 Sep 2019 21:46:07 -  1.26
+++ Makefile15 Mar 2020 15:28:30 -
@@ -5,6 +5,7 @@ COMMENT=fancy interface to the Python i
 MODPY_EGG_VERSION=0.18
 DISTNAME=  bpython-${MODPY_EGG_VERSION}
 CATEGORIES=devel
+REVISION = 0
 
 HOMEPAGE=  https://bpython-interpreter.org/
 MAINTAINER=Brian Callahan 
@@ -19,11 +20,7 @@ MODULES= lang/python
 MODPY_SETUPTOOLS=Yes
 
 FLAVORS =  python3
-FLAVOR ?=
-
-.if ${FLAVOR:Mpython3}
-PKGNAME =  bpython${MODPY_MAJOR_VERSION}-${MODPY_EGG_VERSION}
-.endif
+FLAVOR =   python3
 
 RUN_DEPENDS=   devel/desktop-file-utils \
devel/py-curtsies${MODPY_FLAVOR} \
@@ -37,17 +34,5 @@ BUILD_DEPENDS=   devel/py-babel${MODPY_FLA
 
 TEST_DEPENDS=  devel/py-mock${MODPY_FLAVOR} \
devel/py-twisted${MODPY_FLAVOR}
-
-# de-conflict flavors
-post-install:
-.for b in bpdb bpython bpython-curses bpython-urwid
-   mv ${PREFIX}/bin/$b{,${MODPY_BIN_SUFFIX}}
-.endfor
-   mv ${PREFIX}/man/man1/bpython{,${MODPY_BIN_SUFFIX}}.1
-   mv ${PREFIX}/man/man5/bpython-config{,${MODPY_BIN_SUFFIX}}.5
-   mv 
${PREFIX}/share/appinfo/org.bpython-interpreter.bpython{,${MODPY_BIN_SUFFIX}}.appdata.xml
-   mv 
${PREFIX}/share/applications/org.bpython-interpreter.bpython{,${MODPY_BIN_SUFFIX}}.desktop
-   mv ${PREFIX}/share/pixmaps/bpython{,${MODPY_BIN_SUFFIX}}.png
-
 
 .include 
Index: pkg/PLIST
===
RCS file: /home/cvs/ports/devel/bpython/pkg/PLIST,v
retrieving revision 1.9
diff -u -p -r1.9 PLIST
--- pkg/PLIST   9 Aug 2019 00:00:40 -   1.9
+++ pkg/PLIST   15 Mar 2020 16:07:07 -
@@ -1,8 +1,10 @@
 @comment $OpenBSD: PLIST,v 1.9 2019/08/09 00:00:40 bcalla

Re: update: devel/py-wcwidth (cascading devel/{py-prompt_toolkit,ipython} dropping py2 support)

[Paco Esteban]

Ping ?

(This needs bket's diff discussed on another thread:
https://marc.info/?l=openbsd-ports&m=158464110714289&w=2)

Find attached the (still relevant) parts for convenience.

On Sun, 15 Mar 2020, Bjorn Ketelaars wrote:

> On Sun 15/03/2020 17:52, Paco Esteban wrote:
> > On Sun, 15 Mar 2020, Paco Esteban wrote:
> > 
> > > * devel/py-prompt_toolkit
> > > * also attached to this email.  Just made py3 only and bumped
> > >   revison.  All tests pass.
> > >   This also triggers the conversion of devel/ipython to py3 only.
> > >   Four tests fail, just as before.
> > >   Maintainer (edd@) has some doubts about this.  Mainly if users
> > >   will be upset for this.  Comments on this are appreciated.
> > 
> > I just realized that devel/ipython has A METRIC TON of consumers ... not
> > sure how they slept under my radar, sorry.
> 
> I'm trying to push several tools from jupyter forward, several of which
> depend on ipython. If I'm not mistaken there are 3 more pieces needed
> before it is possible to migrate ipython to py3-only:
> 
> - rpointel@ sent a diff to ports@ for plaso and all of its dependencies,
>   which still needs testing:
>   https://marc.info/?l=openbsd-ports&m=158368639815369&w=2
> - I sent a diff to port@ for devel/py-nbconvert (update and migration to
>   py3-only): https://marc.info/?l=openbsd-ports&m=158351871906886&w=2
> 
> After the above has been committed we need to migrate several ports to
> py3-only at the same time. I have a diff sitting in my tree for this:
> 
> PORT  CONSUMER(S)
> py-traitlets  ipython py-ipykernel py-jupyter_client py-jupyter_core 
> py-nbformat
> ipython   py-ipykernel py-jupyter_client py-nbformat 
> py-traitlets
> py-ipykernel  ipython py-jupyter_client py-jupyter_core
> py-jupyter_core   py-jupyter_client py-nbformat
> py-jupyter_client py-ipykernel
> py-nbformat   ipython
> 

-- 
Paco Esteban.
0x5818130B8A6DBC03
Index: Makefile
===
RCS file: /home/cvs/ports/devel/py-wcwidth/Makefile,v
retrieving revision 1.7
diff -u -p -r1.7 Makefile
--- Makefile12 Jul 2019 20:45:55 -  1.7
+++ Makefile13 Mar 2020 18:29:24 -
@@ -2,10 +2,9 @@
 
 COMMENT=   measures number of terminal column cells of wide-char 
code
 
-MODPY_EGG_VERSION= 0.1.7
+MODPY_EGG_VERSION= 0.1.8
 DISTNAME=  wcwidth-${MODPY_EGG_VERSION}
 PKGNAME=   py-${DISTNAME}
-REVISION=  1
 
 CATEGORIES=devel
 
@@ -22,6 +21,6 @@ MODPY_PYTEST =Yes
 MODPY_PYTEST_ARGS =wcwidth
 
 FLAVORS=   python3
-FLAVOR ?=
+FLAVOR =   python3
 
 .include 
Index: distinfo
===
RCS file: /home/cvs/ports/devel/py-wcwidth/distinfo,v
retrieving revision 1.3
diff -u -p -r1.3 distinfo
--- distinfo29 Sep 2016 15:05:13 -  1.3
+++ distinfo13 Mar 2020 18:30:46 -
@@ -1,2 +1,2 @@
-SHA256 (wcwidth-0.1.7.tar.gz) = PfNzciJtbmPhseHtoVxZS8qYoi0zojgyqQmY+qlrxl4=
-SIZE (wcwidth-0.1.7.tar.gz) = 22884
+SHA256 (wcwidth-0.1.8.tar.gz) = 8os+imSD5dSef4lJrBp4MU50AzOuMFtLpd79PnT7N6g=
+SIZE (wcwidth-0.1.8.tar.gz) = 19957
Index: pkg/PLIST
===
RCS file: /home/cvs/ports/devel/py-wcwidth/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST   7 Jan 2016 08:13:29 -   1.1.1.1
+++ pkg/PLIST   13 Mar 2020 18:30:19 -
@@ -1,4 +1,6 @@
 @comment $OpenBSD: PLIST,v 1.1.1.1 2016/01/07 08:13:29 ajacoutot Exp $
+@conflict py-wcwidth-*
+@pkgpath devel/py-wcwidth
 lib/python${MODPY_VERSION}/site-packages/wcwidth/
 
lib/python${MODPY_VERSION}/site-packages/wcwidth-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/
 
lib/python${MODPY_VERSION}/site-packages/wcwidth-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/PKG-INFO
Index: Makefile
===
RCS file: /home/cvs/ports/devel/py-prompt_toolkit/Makefile,v
retrieving revision 1.14
diff -u -p -r1.14 Makefile
--- Makefile2 Dec 2019 14:16:45 -   1.14
+++ Makefile13 Mar 2020 18:45:31 -
@@ -9,6 +9,7 @@ MODPY_EGG_VERSION=  1.0.15
 DISTNAME=  prompt_toolkit-${MODPY_EGG_VERSION}
 PKGNAME=   py-${DISTNAME}
 EPOCH= 1
+REVISION=  0
 
 CATEGORIES=devel
 
@@ -29,6 +30,6 @@ RUN_DEPENDS=  devel/py-six${MODPY_FLAVOR
textproc/py-pygments${MODPY_FLAVOR}
 
 FLAVORS=   python3
-FLAVOR ?=
+FLAVOR =   python3
 
 .include 
Index: pkg/PLIST
===
RCS file: /home/cvs/ports/devel/py-prompt_toolkit/pkg/PLIST,v
retrieving revision 1.8
diff -u -p -r1.8 PLIST
--- pkg/PLIST   2 Dec 2019 14:16:45 -   1.8
+++ pkg/PLIST   13 Mar 

Re: Migrate ipython/py-ipykernel/py-jupyter_*/py-nbformat/py-traitlets to py3-only

[Paco Esteban]

On Sun, 22 Mar 2020, Bjorn Ketelaars wrote:

> Unless someone objects, I will commit my diff tomorrow morning (CET).

Looking forward to it.

I'll send a ping with my two related diffs then :-)

-- 
Paco Esteban.
0x5818130B8A6DBC03

Re: remove historic misc/rpm

[Marc Espie]

On Sun, Mar 22, 2020 at 06:04:30PM +0100, Klemens Nanni wrote:
> We package RPM 3.0.6 released (Sep 13 2000 ??), see 
> https://rpm.org/timeline.html
> 
> Does anyone use it?  No ports in-tree does except patchuils for tests
> (see my other mail on ports@).
> 
> This port seems useless, conflicts the actually useful rpm2cpio port and
> seems to be dead weight.
> 
> OK to remove?
> 
> 
> Index: Makefile
> ===
> RCS file: /cvs/ports/devel/quirks/Makefile,v
> retrieving revision 1.892
> diff -u -p -r1.892 Makefile
> --- Makefile  20 Mar 2020 14:46:47 -  1.892
> +++ Makefile  22 Mar 2020 17:03:13 -
> @@ -5,7 +5,7 @@ CATEGORIES =  devel databases
>  DISTFILES =
>  
>  # API.rev
> -PKGNAME =quirks-3.283
> +PKGNAME =quirks-3.284
>  PKG_ARCH =   *
>  MAINTAINER = Marc Espie 
>  
> Index: files/Quirks.pm
> ===
> RCS file: /cvs/ports/devel/quirks/files/Quirks.pm,v
> retrieving revision 1.910
> diff -u -p -r1.910 Quirks.pm
> --- files/Quirks.pm   20 Mar 2020 14:46:47 -  1.910
> +++ files/Quirks.pm   22 Mar 2020 17:04:11 -
> @@ -1441,6 +1441,7 @@ my $obsolete_reason = {
>   'spectrum' => 3,
>   'py-xmpp' => 3,
>   'mirage' => 0,
> + 'rpm' => 0,
>  };
>  
>  # reasons for obsolete packages
> 

Sure.

It was marginally useful a very long time ago back when we had
linux emulation.

remove historic misc/rpm

[Klemens Nanni]

We package RPM 3.0.6 released (Sep 13 2000 ??), see 
https://rpm.org/timeline.html

Does anyone use it?  No ports in-tree does except patchuils for tests
(see my other mail on ports@).

This port seems useless, conflicts the actually useful rpm2cpio port and
seems to be dead weight.

OK to remove?


Index: Makefile
===
RCS file: /cvs/ports/devel/quirks/Makefile,v
retrieving revision 1.892
diff -u -p -r1.892 Makefile
--- Makefile20 Mar 2020 14:46:47 -  1.892
+++ Makefile22 Mar 2020 17:03:13 -
@@ -5,7 +5,7 @@ CATEGORIES =devel databases
 DISTFILES =
 
 # API.rev
-PKGNAME =  quirks-3.283
+PKGNAME =  quirks-3.284
 PKG_ARCH = *
 MAINTAINER =   Marc Espie 
 
Index: files/Quirks.pm
===
RCS file: /cvs/ports/devel/quirks/files/Quirks.pm,v
retrieving revision 1.910
diff -u -p -r1.910 Quirks.pm
--- files/Quirks.pm 20 Mar 2020 14:46:47 -  1.910
+++ files/Quirks.pm 22 Mar 2020 17:04:11 -
@@ -1441,6 +1441,7 @@ my $obsolete_reason = {
'spectrum' => 3,
'py-xmpp' => 3,
'mirage' => 0,
+   'rpm' => 0,
 };
 
 # reasons for obsolete packages

patchutils: remove misc/rpm TDEP

[Klemens Nanni]

We package RPM from 2000 and I want to remove it: this is the only
consumer of it in-tree... and it only uses it for tests.

with rpm installed:

# TOTAL: 137
# PASS:  20
# SKIP:  0
# XFAIL: 4
# FAIL:  113
# XPASS: 0
# ERROR: 0

with this diff and rpm removed:

# TOTAL: 137
# PASS:  30
# SKIP:  0
# XFAIL: 4
# FAIL:  103
# XPASS: 0
# ERROR: 0

OK?


Index: Makefile
===
RCS file: /cvs/ports/textproc/patchutils/Makefile,v
retrieving revision 1.5
diff -u -p -r1.5 Makefile
--- Makefile3 Jun 2019 16:06:58 -   1.5
+++ Makefile22 Mar 2020 16:57:06 -
@@ -22,8 +22,7 @@ MASTER_SITES= http://cyberelk.net/tim/da
 BUILD_DEPENDS= ${RUN_DEPENDS} \
textproc/xmlto
 RUN_DEPENDS=   devel/gpatch
-TEST_DEPENDS=  misc/rpm \
-   shells/bash
+TEST_DEPENDS=  shells/bash
 TEST_FLAGS=PATCH=${LOCALBASE}/bin/gpatch
 
 # only needed for tests

Re: Migrate ipython/py-ipykernel/py-jupyter_*/py-nbformat/py-traitlets to py3-only

[Bjorn Ketelaars]

On Sun 22/03/2020 16:31, Edd Barrett wrote:
> On Sun, Mar 22, 2020 at 03:57:50PM +, Edd Barrett wrote:
> > On Sun, Mar 22, 2020 at 03:34:07PM +, Edd Barrett wrote:
> > > Either way I will write an entry for current.html
> > > ...
> > >  * For `ipython-3` to be renamed `ipython` (as it is in the diff).
> > 
> > ^ Assuming we do this:
> > 
> > Index: current.html
> > ===
> > RCS file: /cvs/www/faq/current.html,v
> > retrieving revision 1.1032
> > diff -u -p -r1.1032 current.html
> > --- current.html17 Mar 2020 04:56:14 -  1.1032
> > +++ current.html22 Mar 2020 15:55:52 -
> > @@ -265,3 +265,11 @@ Please note that tools supplied by this 
> >  
> >  
> >  $OpenBSD: current.html,v 1.1032 2020/03/17 04:56:14 tb Exp $
> > +
> > +2020/02/23 - [packages] Retiring Python 2 support in 
> > IPython
> > +
> > +The IPython port now only supports Python 3.
> > +
> > +
> > +/usr/local/bin/ipython-3 has been renamed to
> > +/usr/local/bin/ipython.
> 
> Oops! Let's try again.
> 
> Index: current.html
> ===
> RCS file: /cvs/www/faq/current.html,v
> retrieving revision 1.1032
> diff -u -p -r1.1032 current.html
> --- current.html  17 Mar 2020 04:56:14 -  1.1032
> +++ current.html  22 Mar 2020 16:30:52 -
> @@ -256,6 +256,13 @@ Please note that tools supplied by this 
>  jupyter-notebook-3 has been renamed to
>  jupyter-notebook.
>  
> +2020/03/23 - [packages] Retiring Python 2 support in 
> IPython
> +
> +The IPython port now only supports Python 3.
> +
> +
> +/usr/local/bin/ipython-3 has been renamed to
> +/usr/local/bin/ipython.
>  
>  

clang-tools-extra: switch to python 3

[Klemens Nanni]

rsadowski noted that a simple version bump is not enough.

Using 2to3 for a few simple conversions is all it takes to fully switch.

Regardless of our python version, this port also requires shebang
adjustments, which are handled now as well.

Built and packaged on sparc64 without Pyhon 2 installed.

FWIW, attached is the output of `make fake' to show what 2to3 does.

OK?


Index: Makefile
===
RCS file: /cvs/ports/devel/clang-tools-extra/Makefile,v
retrieving revision 1.2
diff -u -p -r1.2 Makefile
--- Makefile22 Mar 2020 15:23:38 -  1.2
+++ Makefile22 Mar 2020 16:42:36 -
@@ -17,7 +17,7 @@ COMMENT=  Clang extra tools
 LLVM_V =   8.0.1
 DISTNAME = llvm-${LLVM_V}.src
 PKGNAME=   clang-tools-extra-${LLVM_V}
-REVISION=  0
+REVISION=  1
 
 CATEGORIES =   devel
 
@@ -41,6 +41,7 @@ COMPILER =base-clang ports-gcc
 
 MODULES =  devel/cmake \
lang/python
+MODPY_VERSION =${MODPY_DEFAULT_VERSION_3}
 
 RUN_DEPENDS =  devel/llvm
 
@@ -88,6 +89,8 @@ pre-configure:
-@ln -s ${MODPY_BIN} ${WRKDIR}/bin/python
 
 post-install:
+   find ${PREFIX}/share/clang -type f -name \*.py -exec ${MODPY_BIN_ADJ} 
{} +
+   2to3 -w -n ${PREFIX}/share/clang
${MODPY_BIN} ${MODPY_LIBDIR}/compileall.py \
${PREFIX}/share/clang
 
Index: pkg/PLIST
===
RCS file: /cvs/ports/devel/clang-tools-extra/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST   6 Nov 2019 10:07:56 -   1.1.1.1
+++ pkg/PLIST   22 Mar 2020 16:24:04 -
@@ -8,43 +8,44 @@
 @bin bin/clangd
 @bin bin/find-all-symbols
 @bin bin/modularize
-lib/libclangApplyReplacements.a
-lib/libclangChangeNamespace.a
-lib/libclangDaemon.a
-lib/libclangDoc.a
-lib/libclangIncludeFixer.a
-lib/libclangIncludeFixerPlugin.a
-lib/libclangMove.a
-lib/libclangQuery.a
-lib/libclangReorderFields.a
-lib/libclangTidy.a
-lib/libclangTidyAbseilModule.a
-lib/libclangTidyAndroidModule.a
-lib/libclangTidyBoostModule.a
-lib/libclangTidyBugproneModule.a
-lib/libclangTidyCERTModule.a
-lib/libclangTidyCppCoreGuidelinesModule.a
-lib/libclangTidyFuchsiaModule.a
-lib/libclangTidyGoogleModule.a
-lib/libclangTidyHICPPModule.a
-lib/libclangTidyLLVMModule.a
-lib/libclangTidyMPIModule.a
-lib/libclangTidyMiscModule.a
-lib/libclangTidyModernizeModule.a
-lib/libclangTidyObjCModule.a
-lib/libclangTidyPerformanceModule.a
-lib/libclangTidyPlugin.a
-lib/libclangTidyPortabilityModule.a
-lib/libclangTidyReadabilityModule.a
-lib/libclangTidyUtils.a
-lib/libclangTidyZirconModule.a
-lib/libfindAllSymbols.a
+@static-lib lib/libclangApplyReplacements.a
+@static-lib lib/libclangChangeNamespace.a
+@static-lib lib/libclangDaemon.a
+@static-lib lib/libclangDoc.a
+@static-lib lib/libclangIncludeFixer.a
+@static-lib lib/libclangIncludeFixerPlugin.a
+@static-lib lib/libclangMove.a
+@static-lib lib/libclangQuery.a
+@static-lib lib/libclangReorderFields.a
+@static-lib lib/libclangTidy.a
+@static-lib lib/libclangTidyAbseilModule.a
+@static-lib lib/libclangTidyAndroidModule.a
+@static-lib lib/libclangTidyBoostModule.a
+@static-lib lib/libclangTidyBugproneModule.a
+@static-lib lib/libclangTidyCERTModule.a
+@static-lib lib/libclangTidyCppCoreGuidelinesModule.a
+@static-lib lib/libclangTidyFuchsiaModule.a
+@static-lib lib/libclangTidyGoogleModule.a
+@static-lib lib/libclangTidyHICPPModule.a
+@static-lib lib/libclangTidyLLVMModule.a
+@static-lib lib/libclangTidyMPIModule.a
+@static-lib lib/libclangTidyMiscModule.a
+@static-lib lib/libclangTidyModernizeModule.a
+@static-lib lib/libclangTidyObjCModule.a
+@static-lib lib/libclangTidyPerformanceModule.a
+@static-lib lib/libclangTidyPlugin.a
+@static-lib lib/libclangTidyPortabilityModule.a
+@static-lib lib/libclangTidyReadabilityModule.a
+@static-lib lib/libclangTidyUtils.a
+@static-lib lib/libclangTidyZirconModule.a
+@static-lib lib/libfindAllSymbols.a
+share/clang/${MODPY_PYCACHE}/
+share/clang/${MODPY_PYCACHE}clang-include-fixer.${MODPY_PYC_MAGIC_TAG}pyc
+share/clang/${MODPY_PYCACHE}clang-tidy-diff.${MODPY_PYC_MAGIC_TAG}pyc
+share/clang/${MODPY_PYCACHE}run-clang-tidy.${MODPY_PYC_MAGIC_TAG}pyc
+share/clang/${MODPY_PYCACHE}run-find-all-symbols.${MODPY_PYC_MAGIC_TAG}pyc
 share/clang/clang-include-fixer.el
 share/clang/clang-include-fixer.py
-share/clang/clang-include-fixer.pyc
 share/clang/clang-tidy-diff.py
-share/clang/clang-tidy-diff.pyc
 share/clang/run-clang-tidy.py
-share/clang/run-clang-tidy.pyc
 share/clang/run-find-all-symbols.py
-share/clang/run-find-all-symbols.pyc
===>  Faking installation for clang-tools-extra-8.0.1p1
[0/1] cd 
/usr/ports/pobj/clang-tools-extra-8.0.1/build-sparc64/tools/clang/tools/extra 
&& /usr/local/bin/cmake -P cmake_install.cmake
-- Install configuration: "Release"
-- Installing: 
/usr/ports/pobj/clang-tools-extra-8.0.1/fake-sparc64/usr/local/lib/libclangApplyRep

Re: Migrate ipython/py-ipykernel/py-jupyter_*/py-nbformat/py-traitlets to py3-only

[Edd Barrett]

On Sun, Mar 22, 2020 at 03:57:50PM +, Edd Barrett wrote:
> On Sun, Mar 22, 2020 at 03:34:07PM +, Edd Barrett wrote:
> > Either way I will write an entry for current.html
> > ...
> >  * For `ipython-3` to be renamed `ipython` (as it is in the diff).
> 
> ^ Assuming we do this:
> 
> Index: current.html
> ===
> RCS file: /cvs/www/faq/current.html,v
> retrieving revision 1.1032
> diff -u -p -r1.1032 current.html
> --- current.html  17 Mar 2020 04:56:14 -  1.1032
> +++ current.html  22 Mar 2020 15:55:52 -
> @@ -265,3 +265,11 @@ Please note that tools supplied by this 
>  
>  
>  $OpenBSD: current.html,v 1.1032 2020/03/17 04:56:14 tb Exp $
> +
> +2020/02/23 - [packages] Retiring Python 2 support in 
> IPython
> +
> +The IPython port now only supports Python 3.
> +
> +
> +/usr/local/bin/ipython-3 has been renamed to
> +/usr/local/bin/ipython.

Oops! Let's try again.

Index: current.html
===
RCS file: /cvs/www/faq/current.html,v
retrieving revision 1.1032
diff -u -p -r1.1032 current.html
--- current.html17 Mar 2020 04:56:14 -  1.1032
+++ current.html22 Mar 2020 16:30:52 -
@@ -256,6 +256,13 @@ Please note that tools supplied by this 
 jupyter-notebook-3 has been renamed to
 jupyter-notebook.
 
+2020/03/23 - [packages] Retiring Python 2 support in 
IPython
+
+The IPython port now only supports Python 3.
+
+
+/usr/local/bin/ipython-3 has been renamed to
+/usr/local/bin/ipython.
 
 

Re: pkg/PLIST automation

[Martin]

Latest update for comms/uhd attached. uhd-3.15.0.0p0 package builds from ports 
successfully right after fix all stupid mistakes. Thanks for help.

Yes, USRP devices require async transfers. In what stage async USB transfers 
codebase now?
I'm looking for a project, but I'm not sure that I'm experienced enough to 
write kernel/libusb code. I have to check.
FYI, I have one USRP device on hand with USB interface to test the driver.

Martin

‐‐‐ Original Message ‐‐‐
On Sunday, March 22, 2020 2:37 PM, Stuart Henderson  
wrote:

> On 2020/03/22 14:29, Stuart Henderson wrote:
>
> > On 2020/03/22 14:17, Marc Espie wrote:
> >
> > > On Sun, Mar 22, 2020 at 01:59:36PM +0100, Marc Espie wrote:
> > >
> > > > On Sun, Mar 22, 2020 at 08:20:59AM +, Martin wrote:
> > > >
> > > > > It looks strange, but
> > > > > make show=PKGNAME
> > > > > uhd-3.15.0.0
> > > > > make show=FULLPKGNAME-
> > > > > uhd-3.15.0.0p0
> > > > > make show=FULLPKGNAME
> > > > > uhd-3.15.0.0p0
> > > > > I attached the new port I do which builds successfully, but make 
> > > > > package is broken.
> > > >
> > > > No it doesn't.
> > > >
> > > > -DCMAKE_C_COMPILER=/usr/local/bin/clang \\
> > > > -DCMAKE_CXX_COMPILER=/usr/local/bin/clang++ \\
> > > > -DPYTHON_EXECUTABLE=/usr/local/bin/python3.7 \\
> > > > -DCMAKE_INSTALL_PREFIX=/usr/local ../
> > > >
> > >
> > > I was talking too fast, your problem is waaay more stupid, the spaces
> > > you left at the end of the line with REVISION ARE significant, and
> > > you end up with a fullpkgname with spaces, which isn't so great ;)
> > > You still need to clean up all the rest though... starting with a clean
> > > system would help.
> > > Also your comment is wy too long
> >
> > With this port, getting it built is likely only the start - for the
> > USB-based devices it looks like they need async transfers, in which case
> > it will need some work on the USB stack (kernel/libusb) as these aren't
> > implemented yet (same problem with rtl-sdr which is why the port of that
> > is barely functional).
> > If that doesn't sound like something you want to work on then OpenBSD is
> > probably not the right platform to use with this (rather nice looking)
> > hardware.
>
> Here's a slightly cleaned up port FWIW.




uhd.tgz
Description: application/gzip

Re: Migrate ipython/py-ipykernel/py-jupyter_*/py-nbformat/py-traitlets to py3-only

[Edd Barrett]

On Sun, Mar 22, 2020 at 03:34:07PM +, Edd Barrett wrote:
> Either way I will write an entry for current.html
> ...
>  * For `ipython-3` to be renamed `ipython` (as it is in the diff).

^ Assuming we do this:

Index: current.html
===
RCS file: /cvs/www/faq/current.html,v
retrieving revision 1.1032
diff -u -p -r1.1032 current.html
--- current.html17 Mar 2020 04:56:14 -  1.1032
+++ current.html22 Mar 2020 15:55:52 -
@@ -265,3 +265,11 @@ Please note that tools supplied by this 
 
 
 $OpenBSD: current.html,v 1.1032 2020/03/17 04:56:14 tb Exp $
+
+2020/02/23 - [packages] Retiring Python 2 support in 
IPython
+
+The IPython port now only supports Python 3.
+
+
+/usr/local/bin/ipython-3 has been renamed to
+/usr/local/bin/ipython.

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

Re: Migrate ipython/py-ipykernel/py-jupyter_*/py-nbformat/py-traitlets to py3-only

[Edd Barrett]

Hi,

Sorry for the delay.

On Sat, Mar 21, 2020 at 05:14:57PM +0100, Paco Esteban wrote:
> Waiting for Edd's (maintainer) decision on the binary name change
> notification for users regarding ipython.

Currently we have install either `ipython-2` and `ipython-3`. Nothing
installs any `ipython` binary at present.

So I'd be happy for either of the following things to happen:

 * For `ipython-2` to just disappear.
 * For `ipython-3` to be renamed `ipython` (as it is in the diff).

Either way I will write an entry for current.html

Latest diff reads fine for me.

Thanks for working on this!

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

Re: clang-tools-extra: prevent picking up z3

[Klemens Nanni]

On Sun, Mar 22, 2020 at 02:42:39PM +0100, Klemens Nanni wrote:
> If devel/z3 is installed, cmake picks it up at configure time:
> 
>   -- Found Z3: /usr/local/lib/libz3.so.2.0 (found suitable version "4.8.6 
> - 64 bit", minimum required is "4.7.1") 
>   -- Configuring done
> 
> Bump REVISION to be safe in case build machines have picked it up.
jca noted how devel/llvm turns off the detection already;  above log does
not appear in the first place.

> Feedback? OK?
Now Cc'ing maintainer as well.


Index: Makefile
===
RCS file: /cvs/ports/devel/clang-tools-extra/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 Makefile
--- Makefile6 Nov 2019 10:07:55 -   1.1.1.1
+++ Makefile22 Mar 2020 15:14:32 -
@@ -17,6 +17,7 @@ COMMENT=  Clang extra tools
 LLVM_V =   8.0.1
 DISTNAME = llvm-${LLVM_V}.src
 PKGNAME=   clang-tools-extra-${LLVM_V}
+REVISION=  0
 
 CATEGORIES =   devel
 
@@ -50,6 +51,7 @@ CONFIGURE_ARGS =  -DLLVM_ENABLE_FFI=OFF \
-DLLVM_ENABLE_TERMINFO=ON \
-DLLVM_ENABLE_RTTI=ON \
-DCMAKE_DISABLE_FIND_PACKAGE_LibXml2=ON \
+   -DCMAKE_DISABLE_FIND_PACKAGE_Z3=ON \
-DLLVM_LINK_LLVM_DYLIB=OFF \
-DLLVM_BUILD_LLVM_DYLIB=OFF \
-DGO_EXECUTABLE=GO_EXECUTABLE-NOTFOUND \

Re: pkg/PLIST automation

[Stuart Henderson]

On 2020/03/22 14:29, Stuart Henderson wrote:
> On 2020/03/22 14:17, Marc Espie wrote:
> > On Sun, Mar 22, 2020 at 01:59:36PM +0100, Marc Espie wrote:
> > > On Sun, Mar 22, 2020 at 08:20:59AM +, Martin wrote:
> > > > It looks strange, but
> > > > 
> > > > make show=PKGNAME
> > > > uhd-3.15.0.0
> > > > make show=FULLPKGNAME-
> > > > uhd-3.15.0.0p0
> > > > make show=FULLPKGNAME
> > > > uhd-3.15.0.0p0
> > > > 
> > > > I attached the new port I do which builds successfully, but make 
> > > > package is broken.
> > > 
> > > No it doesn't.
> > > 
> > > 
> > > -DCMAKE_C_COMPILER=/usr/local/bin/clang \
> > > -DCMAKE_CXX_COMPILER=/usr/local/bin/clang++ \
> > > -DPYTHON_EXECUTABLE=/usr/local/bin/python3.7 \
> > > -DCMAKE_INSTALL_PREFIX=/usr/local ../
> > > 
> > 
> > I was talking too fast, your problem is waaay more stupid, the spaces
> > you left at the end of the line with REVISION *ARE* significant, and
> > you end up with a fullpkgname with spaces, which isn't so great ;)
> > 
> > You still need to clean up all the rest though... starting with a clean
> > system would help.
> > 
> > Also your comment is wy too long
> > 
> 
> With this port, getting it built is likely only the start - for the
> USB-based devices it looks like they need async transfers, in which case
> it will need some work on the USB stack (kernel/libusb) as these aren't
> implemented yet (same problem with rtl-sdr which is why the port of that
> is barely functional).
> 
> If that doesn't sound like something you want to work on then OpenBSD is
> probably not the right platform to use with this (rather nice looking)
> hardware.
> 

Here's a slightly cleaned up port FWIW.



uhd.tgz
Description: application/tar-gz

Re: pkg/PLIST automation

[Stuart Henderson]

On 2020/03/22 14:17, Marc Espie wrote:
> On Sun, Mar 22, 2020 at 01:59:36PM +0100, Marc Espie wrote:
> > On Sun, Mar 22, 2020 at 08:20:59AM +, Martin wrote:
> > > It looks strange, but
> > > 
> > > make show=PKGNAME
> > > uhd-3.15.0.0
> > > make show=FULLPKGNAME-
> > > uhd-3.15.0.0p0
> > > make show=FULLPKGNAME
> > > uhd-3.15.0.0p0
> > > 
> > > I attached the new port I do which builds successfully, but make package 
> > > is broken.
> > 
> > No it doesn't.
> > 
> > 
> > -DCMAKE_C_COMPILER=/usr/local/bin/clang \
> > -DCMAKE_CXX_COMPILER=/usr/local/bin/clang++ \
> > -DPYTHON_EXECUTABLE=/usr/local/bin/python3.7 \
> > -DCMAKE_INSTALL_PREFIX=/usr/local ../
> > 
> 
> I was talking too fast, your problem is waaay more stupid, the spaces
> you left at the end of the line with REVISION *ARE* significant, and
> you end up with a fullpkgname with spaces, which isn't so great ;)
> 
> You still need to clean up all the rest though... starting with a clean
> system would help.
> 
> Also your comment is wy too long
> 

With this port, getting it built is likely only the start - for the
USB-based devices it looks like they need async transfers, in which case
it will need some work on the USB stack (kernel/libusb) as these aren't
implemented yet (same problem with rtl-sdr which is why the port of that
is barely functional).

If that doesn't sound like something you want to work on then OpenBSD is
probably not the right platform to use with this (rather nice looking)
hardware.

z3: conflict with devel/angr/py-z3-solver

[Klemens Nanni]

The other port has `@conflict z3-*', devel/z3 fails to install on first
try:

# pkg_add z3
quirks-3.273->3.283: ok
Collision in z3-4.8.6: the following files already exist
/usr/local/bin/z3 (py3-z3-solver-4.8.7.0 and z3-4.8.6)
...
Read shared items: ok
Couldn't install z3-4.8.6

Strangely enough, successive pkg_add attempts behave as if the conflict
marker was installed:

# pkg_add z3
quirks-3.283 signed on 2020-03-21T17:14:41Z
Can't install z3-4.8.6 because of conflicts (py3-z3-solver-4.8.7.0)
--- z3-4.8.6 ---
Can't install z3-4.8.6: conflicts
Couldn't install z3-4.8.6

I don't know that pkg_add(1) does under the hood and pkg_create(1) is
not specific enough about whether conflicts need to be registered in
all effected ports:

@conflict pkgspec
Declare a conflict with packages matching pkgspec (see
packages-specs(7)).  The pkgname package can not be installed if
a package matching pkgspec has been installed because they
install the same files and thus conflict.

But I suppose having @conflict in devel/z3 would handle the first
pkg_add attempt properly.

Feedback? OK?

Index: Makefile
===
RCS file: /cvs/ports/math/z3/Makefile,v
retrieving revision 1.18
diff -u -p -r1.18 Makefile
--- Makefile19 Oct 2019 09:42:20 -  1.18
+++ Makefile22 Mar 2020 13:09:20 -
@@ -10,6 +10,7 @@ GH_TAGNAME =  ${GH_PROJECT}-${VERSION}
 
 DISTNAME = ${GH_TAGNAME}
 PKGNAME =  ${DISTNAME:L}
+REVISION = 0
 
 SHARED_LIBS =  z3  2.0 # 4.8
 
Index: pkg/PLIST
===
RCS file: /cvs/ports/math/z3/pkg/PLIST,v
retrieving revision 1.6
diff -u -p -r1.6 PLIST
--- pkg/PLIST   19 Oct 2019 09:42:20 -  1.6
+++ pkg/PLIST   22 Mar 2020 13:09:41 -
@@ -1,4 +1,5 @@
 @comment $OpenBSD: PLIST,v 1.6 2019/10/19 09:42:20 rsadowski Exp $
+@conflict py3-z3-solver-*
 @bin bin/z3
 include/z3++.h
 include/z3.h

clang-tools-extra: prevent picking up z3

[Klemens Nanni]

If devel/z3 is installed, cmake picks it up at configure time:

-- Found Z3: /usr/local/lib/libz3.so.2.0 (found suitable version "4.8.6 
- 64 bit", minimum required is "4.7.1") 
-- Configuring done

Bump REVISION to be safe in case build machines have picked it up.

Feedback? OK?

Index: Makefile
===
RCS file: /cvs/ports/devel/clang-tools-extra/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 Makefile
--- Makefile6 Nov 2019 10:07:55 -   1.1.1.1
+++ Makefile22 Mar 2020 13:36:40 -
@@ -17,6 +17,7 @@ COMMENT=  Clang extra tools
 LLVM_V =   8.0.1
 DISTNAME = llvm-${LLVM_V}.src
 PKGNAME=   clang-tools-extra-${LLVM_V}
+REVISION=  0
 
 CATEGORIES =   devel
 
@@ -57,6 +58,7 @@ CONFIGURE_ARGS =  -DLLVM_ENABLE_FFI=OFF \
 
 CONFIGURE_ARGS +=  -DCLANG_ENABLE_STATIC_ANALYZER=ON \
-DCLANG_INCLUDE_TESTS=OFF \
+   -DCLANG_ANALYZER_ENABLE_Z3_SOLVER=OFF \
-DLLVM_BUILD_TESTS=OFF \
-DLLVM_BUILD_BENCHMARKS=OFF \
-DLLVM_ENABLE_DOXYGEN=OFF \

Re: pkg/PLIST automation

[Marc Espie]

On Sun, Mar 22, 2020 at 01:59:36PM +0100, Marc Espie wrote:
> On Sun, Mar 22, 2020 at 08:20:59AM +, Martin wrote:
> > It looks strange, but
> > 
> > make show=PKGNAME
> > uhd-3.15.0.0
> > make show=FULLPKGNAME-
> > uhd-3.15.0.0p0
> > make show=FULLPKGNAME
> > uhd-3.15.0.0p0
> > 
> > I attached the new port I do which builds successfully, but make package is 
> > broken.
> 
> No it doesn't.
> 
> 
> -DCMAKE_C_COMPILER=/usr/local/bin/clang \
> -DCMAKE_CXX_COMPILER=/usr/local/bin/clang++ \
> -DPYTHON_EXECUTABLE=/usr/local/bin/python3.7 \
> -DCMAKE_INSTALL_PREFIX=/usr/local ../
> 

I was talking too fast, your problem is waaay more stupid, the spaces
you left at the end of the line with REVISION *ARE* significant, and
you end up with a fullpkgname with spaces, which isn't so great ;)

You still need to clean up all the rest though... starting with a clean
system would help.

Also your comment is wy too long

Re: UPDATE: Nextcloud-18.0.2

[Gonzalo Rodriguez]

ping

— gonzalo

> On 13. Mar 2020, at 21:47, Gonzalo L. Rodriguez  wrote:
> 
> Hello,
> 
> Update for Nextcloud to 18.0.2:
> 
> https://nextcloud.com/changelog/#latest18
> 
> OK? Comments?
> 
> Cheers.-
> 
> -- 
> 
>- gonzalo
> 

Re: Kernel relink and disable ulpt in kernel - cups printing

[François Chambaud]

Mikolaj Kucharski  writes:

> Hi,
>
> On Thu, Mar 19, 2020 at 08:08:10AM +0100, Antoine Jacoutot wrote:
>> > - How do you handle kernel modifications via config(8) with kernel
>> >   relinking in a more automated way?
>> 
>> $ cat /etc/rc.shutdown 
>> printf 'disable ulpt\nq\n' | config -ef /bsd
>> sha256 /bsd >/var/db/kernel.SHA256
>
> Neat, simple and straightforward. I like it, thanks Antoine.

Or with -h option:

 -h hashfile
 Place the checksum into hashfile instead of stdout.

sha256 -h /var/db/kernel.SHA256 /bsd

-- 
François Chambaud
www.chambaud.org

Re: pkg/PLIST automation

[Marc Espie]

On Sun, Mar 22, 2020 at 08:20:59AM +, Martin wrote:
> It looks strange, but
> 
> make show=PKGNAME
> uhd-3.15.0.0
> make show=FULLPKGNAME-
> uhd-3.15.0.0p0
> make show=FULLPKGNAME
> uhd-3.15.0.0p0
> 
> I attached the new port I do which builds successfully, but make package is 
> broken.

No it doesn't.


-DCMAKE_C_COMPILER=/usr/local/bin/clang \
-DCMAKE_CXX_COMPILER=/usr/local/bin/clang++ \
-DPYTHON_EXECUTABLE=/usr/local/bin/python3.7 \
-DCMAKE_INSTALL_PREFIX=/usr/local ../


Now your actual problem is rather funny.
GH_* creates a PKGNAME that's incompatible with the OpenBSD naming
rules, and bsd.port.mk does its best, BUT gets all confused with the v*
that looks like an EPOCH already applied.

Re: Firefox -current: hangouts screen sharing cause pledge(2) to kill the browser

[Andrea Biscuola]

> Il giorno 22 mar 2020, alle ore 08:45, Landry Breuil  ha 
> scritto:
> 
>> On Sat, Mar 21, 2020 at 06:49:42PM +0100, Andrea Biscuola wrote:
>> Hi ports@
>> 
>> I switched to firefox recently, because with chromium, hangouts screen 
>> sharing
>> doesn't work at all and I need it for work.
>> 
>> While testing Firefox screen sharing in hangouts, pledge kills it with:
>>"/bsd: firefox[43476]: pledge "", syscall 289"
>> 
>> The syscall itself is shmget(2). Running hangouts with pledge disabled in 
>> firefox
>> works fine (I can share the screen without problems).
> 
> Honestly, i never really tried screen sharing, not expecting it to work
> at all - good to know it can work in some conditions !

The most stupid thing is chromium not coping well with rhe goole services

>> 
>> I tried to see how to solve the situation without disabling pledge. I 
>> modified the
>> kernel to include shmget(2), shmat(2), shmctl(2) and shmdt(2) in the stdio 
>> set
>> and now, screen sharing works.
> 
> Good luck trying to get an shm pledge class :)
> 
> In the meantime, i'd just advise you to disable pledge in firefox.

As semarie@ wrote, I'll just disable pledge for firefox. The shm functions are 
not well accepted on OpenBSD.

In the end, what I dis was just an experiment, but as I said, I already sent 
the message when I fully understood the implications.

Thank you and regards.

Andrea

Re: Firefox -current: hangouts screen sharing cause pledge(2) to kill the browser

[Andrea Biscuola]

> Il giorno 22 mar 2020, alle ore 09:12, Sebastien Marie  ha 
> scritto:
> 
>> On Sat, Mar 21, 2020 at 06:49:42PM +0100, Andrea Biscuola wrote:
>> 
>> I tried to see how to solve the situation without disabling pledge. I 
>> modified the
>> kernel to include shmget(2), shmat(2), shmctl(2) and shmdt(2) in the stdio 
>> set
>> and now, screen sharing works.
> 
> Well. So in order to permit firefox to use such syscalls, you also allowed
> mostly any program using pledge(2) to use them too ?
> 
> Do you think sshd(8) sandbox needs them ? or doas(1) ? or su(1)
> Allowing new syscalls in an existing promise could be a hard exercice as it
> require to considere all the programs already using this promise and see
> implication.
> 

I was thinking the same, and I already sent the message when this came to mind.

Also, looking back, I understood the shm functions are accepted and implemented 
reluctantly in OpenBSD, that's why, probably, they are not allowed at all under 
pledge(2).

Sorry for the noise.

> I would advice you to follow landry@ advice to disable pledge(2) for firefox. 
> It
> is could be less dangerous than extending "stdio" and increasing the attack
> surface of all pledged programs.

Will do

> Thanks.
> -- 
> Sebastien Marie
> 

Thank you!

Andrea

powerpc bulk build report

[cwen]

Bulk build on macppc-0.ports.openbsd.org

Started : Mon Mar  2 16:05:15 MST 2020
Finished: Sun Mar 22 03:42:03 MDT 2020
Duration: 19 Days 10 hours 37 minutes

Built using OpenBSD 6.6-current (GENERIC) #676: Sun Mar  1 13:32:42 MST 2020

Built 9859 packages

Number of packages built each day:
Mar 2: 4340
Mar 3: 1109
Mar 4: 450
Mar 5: 274
Mar 6: 140
Mar 7: 115
Mar 8: 322
Mar 9: 173
Mar 10: 247
Mar 11: 241
Mar 12: 132
Mar 13: 240
Mar 14: 286
Mar 15: 295
Mar 16: 313
Mar 17: 360
Mar 18: 349
Mar 19: 473
Mar 20: 588
Mar 21: 1921
Mar 22: 1010


Critical path missing pkgs: 
http://build-failures.rhaalovely.net/powerpc/2020-03-02/summary.log

Build failures: 35
http://build-failures.rhaalovely.net/powerpc/2020-03-02/benchmarks/wrk.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/cad/magic.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/cad/openscad.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/cad/qucs.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/databases/sqlitebrowser.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/devel/geany.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/devel/libpeas.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/devel/xtensa-elf/gcc.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/games/pokerth.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/games/postal.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/games/stone-soup.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/games/valyriatear.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/games/xevil.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/graphics/colord-gtk.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/lang/gprolog.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/lang/hashlink.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/lang/parrot.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/lang/squeak/vm.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/mail/kopano/core.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/math/py-scikit-learn.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/misc/dtcltiny.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/net/dleyna/renderer.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/net/dleyna/server.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/productivity/gnucash.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/sysutils/usmb.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/telephony/asterisk,imap,-calendar.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/telephony/iaxclient.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/www/sassc.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/x11/gnome/libgweather.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/x11/gtk-vnc.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/x11/gtksourceview4.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/x11/kde4/libs,,-en_US.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/x11/libdbus-c++.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/x11/libhandy.log
http://build-failures.rhaalovely.net/powerpc/2020-03-02/x11/mate/caja.log

Recurrent failures:
 failures/benchmarks/wrk.log
 failures/cad/magic.log
 failures/cad/openscad.log
 failures/cad/qucs.log
 failures/databases/sqlitebrowser.log
 failures/devel/geany.log
 failures/devel/libpeas.log
 failures/devel/xtensa-elf/gcc.log
 failures/games/stone-soup.log
 failures/games/valyriatear.log
 failures/games/xevil.log
 failures/graphics/colord-gtk.log
 failures/lang/gprolog.log
 failures/lang/parrot.log
 failures/lang/squeak/vm.log
 failures/mail/kopano/core.log
 failures/productivity/gnucash.log
 failures/summary.log
 failures/sysutils/usmb.log
 failures/telephony/iaxclient.log
 failures/www/sassc.log
 failures/x11/gnome/libgweather.log
 failures/x11/gtk-vnc.log
 failures/x11/gtksourceview4.log
 failures/x11/kde4/libs,,-en_US.log
 failures/x11/libdbus-c++.log
 failures/x11/libhandy.log
 failures/x11/mate/caja.log

New failures:
+failures/lang/hashlink.log
+failures/telephony/asterisk,imap,-calendar.log

Resolved failures:
-failures/cad/netgen.log
-failures/databases/sqliteman.log
-failures/geo/spatialite/gis.log
-failures/x11/kde4/smokeqt.log
-failures/x11/ogre.log

Packages newly built:
+cad/netgen
+databases/py-minidb,python3
+devel/git-crypt
+devel/got,-main
+devel/got,-web
+devel/http-parser
+devel/p5-Conf-Libconfig
+devel/py-esptool,python3
+devel/py-kitchen,python3
+editors/py-qscintilla,-common
+editors/py-qscintilla,-main
+editors/py-qscintilla,python3
+editors/py-qscintilla,python3,-common
+editors/py-qscintilla,python3,-main
+geo/spatialite/gis
+misc/p5-OPCUA-Open62541
+multimedia/libde265
+multimedia/libde265,-main
+multimedia/libde265,-tools
+multimedia/libheif
+net/purple-matrix
+net/py-tlslite-ng,python3

Re: pkg/PLIST automation

[Martin]

It looks strange, but

make show=PKGNAME
uhd-3.15.0.0
make show=FULLPKGNAME-
uhd-3.15.0.0p0
make show=FULLPKGNAME
uhd-3.15.0.0p0

I attached the new port I do which builds successfully, but make package is 
broken.

This is my first port, any advice would be highly appreciated.

Martin

‐‐‐ Original Message ‐‐‐
On Saturday, March 21, 2020 2:11 PM, Ingo Schwarze  wrote:

> Hi Martin,
>
> Martin wrote on Sat, Mar 21, 2020 at 06:07:57AM +:
>
> > make fake -> make update-plist successfully generates PLIST but
> > make package and make install produce an error:
> > make: don't know how to make /usr/ports/packages/.tgz
> > (prerequisite of: /usr/ports/packages/amd64/all/uhd-3.15.0.0p0)
>
> It is hard to judge what exactly went wrong without seeing your
> Makefile.
>
> You are supposed to follow thw documentation (e.g. bsd.port.mk(5))
> and also use it for debugging. If you can't find your mistake by
> looking at the documentation, you can try looking at the code -
> in this case /usr/ports/infrastructure/mk/bsd.port.mk - to find
> out where exactly things break, then deduce from that what your
> error might be and go back to the documentation how the specific
> thing(s) that broke for you are supposed to be used correctly.
>
> If you can't find your own mistake and ask for the help of others
> to find them, you are supposed to provide complete information.
> It is hard to find your error without seeing your Makefile.
>
> Here is an example of how i might proceed to debug the message above
> if the documentation wouldn't help me (assuming you did not
> set MULTI_PACKAGES):
>
> -   search for "\.tgz" -> line 965
> so ${FULLPKGNAME${_S}} seems to be empty
>
> -   without MULTI_PACKAGES, _S is "-"
> (leaving it as an exercise why that is so)
>
> -   line 636: FULLPKGNAME- is set to ${FULLPKGNAME}
>
> -   line 611: FULLPKGNAME is set to ${PKGNAME}${FLAVOR_EXT}
>
> So try to find out whether in your Makefile, you fiddled with
> variables like PKGNAME or FULLPKGNAME in an invalid manner.
>
> While reading bsd.port.mk, you can also use commands like
> make show=FULLPKGNAME-
> make show=FULLPKGNAME
> make show=PKGNAME
> to inspect various variables that file is using, to debug your Makefile.
>
> I find it hard to be more specific given the amount of information
> you provided.
>
> If you did try to use MULTI_PACKAGES, first try to get it to work
> without that, then add MULTI_PACKAGES once everything works without it.
>
> Yours,
> Ingo
>




port-uhd.tgz
Description: application/gzip

Re: Firefox -current: hangouts screen sharing cause pledge(2) to kill the browser

[Sebastien Marie]

On Sat, Mar 21, 2020 at 06:49:42PM +0100, Andrea Biscuola wrote:
> 
> I tried to see how to solve the situation without disabling pledge. I 
> modified the
> kernel to include shmget(2), shmat(2), shmctl(2) and shmdt(2) in the stdio set
> and now, screen sharing works.

Well. So in order to permit firefox to use such syscalls, you also allowed
mostly any program using pledge(2) to use them too ?

Do you think sshd(8) sandbox needs them ? or doas(1) ? or su(1) ?

Allowing new syscalls in an existing promise could be a hard exercice as it
require to considere all the programs already using this promise and see
implication.

I would advice you to follow landry@ advice to disable pledge(2) for firefox. It
is could be less dangerous than extending "stdio" and increasing the attack
surface of all pledged programs.

Thanks.
-- 
Sebastien Marie

Re: Firefox -current: hangouts screen sharing cause pledge(2) to kill the browser

[Landry Breuil]

On Sat, Mar 21, 2020 at 06:49:42PM +0100, Andrea Biscuola wrote:
> Hi ports@
> 
> I switched to firefox recently, because with chromium, hangouts screen sharing
> doesn't work at all and I need it for work.
> 
> While testing Firefox screen sharing in hangouts, pledge kills it with:
>   "/bsd: firefox[43476]: pledge "", syscall 289"
> 
> The syscall itself is shmget(2). Running hangouts with pledge disabled in 
> firefox
> works fine (I can share the screen without problems).

Honestly, i never really tried screen sharing, not expecting it to work
at all - good to know it can work in some conditions !

> 
> I tried to see how to solve the situation without disabling pledge. I 
> modified the
> kernel to include shmget(2), shmat(2), shmctl(2) and shmdt(2) in the stdio set
> and now, screen sharing works.

Good luck trying to get an shm pledge class :)

In the meantime, i'd just advise you to disable pledge in firefox.