CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: j...@cvs.openbsd.org2022/12/29 23:31:52 Modified files: games/arx-libertatis: Makefile Log message: Re-enable LTO on riscv64
Re: CVS: cvs.openbsd.org: ports
On 2022/12/28 07:19, Antoine Jacoutot wrote: > CVSROOT: /cvs > Module name: ports > Changes by: ajacou...@cvs.openbsd.org 2022/12/28 07:19:10 > > Modified files: > audio/soundtouch: Makefile distinfo > audio/soundtouch/pkg: PLIST > > Log message: > Update to soundtouch-2.3.2. > The cmake build doesn't have a way to disable x86 optimizations, this fails on at least i386. Might be simplest to move back to autoconf, what do you think? >>> Building on localhost under audio/soundtouch BDEPENDS = [devel/cmake;devel/ninja] DIST = [audio/soundtouch:soundtouch-2.3.2.tar.gz] FULLPKGNAME = soundtouch-2.3.2 Avoided depends for cmake-3.25.1v0 ninja-1.11.1 distfiles size=583373 >>> Running build in audio/soundtouch at 1672330144.42 ===> audio/soundtouch ===> soundtouch-2.3.2 depends on: cmake-* -> cmake-3.25.1v0 Package database already locked... awaiting release... done! ===> soundtouch-2.3.2 depends on: ninja-* -> ninja-1.11.1 ===> Verifying specs: c++ c++abi pthread c m ===> found c++.9.0 c++abi.6.0 pthread.26.3 c.96.4 m.10.1 ===> Checking files for soundtouch-2.3.2 `/mnt/distfiles/soundtouch-2.3.2.tar.gz' is up to date. >> (SHA256) soundtouch-2.3.2.tar.gz: OK ===> Extracting for soundtouch-2.3.2 ===> Patching for soundtouch-2.3.2 ===> Compiler link: clang -> /usr/bin/clang ===> Compiler link: clang++ -> /usr/bin/clang++ ===> Compiler link: cc -> /usr/bin/cc ===> Compiler link: c++ -> /usr/bin/c++ ===> Generating configure for soundtouch-2.3.2 ===> Configuring for soundtouch-2.3.2 -- The CXX compiler identification is Clang 13.0.0 -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Check for working CXX compiler: /pobj/soundtouch-2.3.2/bin/c++ - skipped -- Detecting CXX compile features -- Detecting CXX compile features - done -- Could NOT find OpenMP_CXX (missing: OpenMP_CXX_FLAGS OpenMP_CXX_LIB_NAMES) -- Could NOT find OpenMP (missing: OpenMP_CXX_FOUND) -- Configuring done -- Generating done CMake Warning: Manually-specified variables were not used by the project: CMAKE_ASM_COMPILER_AR CMAKE_ASM_COMPILER_RANLIB CMAKE_C_COMPILER_AR CMAKE_C_COMPILER_RANLIB CMAKE_Fortran_COMPILER_AR CMAKE_Fortran_COMPILER_RANLIB CMAKE_OBJCXX_COMPILER_AR CMAKE_OBJCXX_COMPILER_RANLIB CMAKE_OBJC_COMPILER_AR CMAKE_OBJC_COMPILER_RANLIB -- Build files have been written to: /pobj/soundtouch-2.3.2/build-i386 ===> Building for soundtouch-2.3.2 [1/20] /pobj/soundtouch-2.3.2/bin/c++ -DSOUNDTOUCH_FLOAT_SAMPLES -DSoundTouch_EXPORTS -I/pobj/soundtouch-2.3.2/soundtouch/include -O2 -pipe -DNDEBUG -fPIC -Ofast -MD -MT CMakeFiles/SoundTouch.dir/source/SoundTouch/AAFilter.cpp.o -MF CMakeFiles/SoundTouch.dir/source/SoundTouch/AAFilter.cpp.o.d -o CMakeFiles/SoundTouch.dir/source/SoundTouch/AAFilter.cpp.o -c /pobj/soundtouch-2.3.2/soundtouch/source/SoundTouch/AAFilter.cpp [2/20] /pobj/soundtouch-2.3.2/bin/c++ -DSOUNDTOUCH_FLOAT_SAMPLES -DSoundTouch_EXPORTS -I/pobj/soundtouch-2.3.2/soundtouch/include -O2 -pipe -DNDEBUG -fPIC -Ofast -MD -MT CMakeFiles/SoundTouch.dir/source/SoundTouch/BPMDetect.cpp.o -MF CMakeFiles/SoundTouch.dir/source/SoundTouch/BPMDetect.cpp.o.d -o CMakeFiles/SoundTouch.dir/source/SoundTouch/BPMDetect.cpp.o -c /pobj/soundtouch-2.3.2/soundtouch/source/SoundTouch/BPMDetect.cpp [3/20] /pobj/soundtouch-2.3.2/bin/c++ -DSOUNDTOUCH_FLOAT_SAMPLES -DSoundTouch_EXPORTS -I/pobj/soundtouch-2.3.2/soundtouch/include -O2 -pipe -DNDEBUG -fPIC -Ofast -MD -MT CMakeFiles/SoundTouch.dir/source/SoundTouch/cpu_detect_x86.cpp.o -MF CMakeFiles/SoundTouch.dir/source/SoundTouch/cpu_detect_x86.cpp.o.d -o CMakeFiles/SoundTouch.dir/source/SoundTouch/cpu_detect_x86.cpp.o -c /pobj/soundtouch-2.3.2/soundtouch/source/SoundTouch/cpu_detect_x86.cpp /pobj/soundtouch-2.3.2/soundtouch/source/SoundTouch/cpu_detect_x86.cpp:49:12: warning: 'bit_MMX' macro redefined [-Wmacro-redefined] #define bit_MMX (1 << 23) ^ /usr/lib/clang/13.0.0/include/cpuid.h:133:9: note: previous definition is here #define bit_MMX 0x0080 ^ /pobj/soundtouch-2.3.2/soundtouch/source/SoundTouch/cpu_detect_x86.cpp:50:12: warning: 'bit_SSE' macro redefined [-Wmacro-redefined] #define bit_SSE (1 << 25) ^ /usr/lib/clang/13.0.0/include/cpuid.h:136:9: note: previous definition is here #define bit_SSE 0x0200 ^ /pobj/soundtouch-2.3.2/soundtouch/source/SoundTouch/cpu_detect_x86.cpp:51:12: warning: 'bit_SSE2' macro redefined [-Wmacro-redefined] #define bit_SSE2(1 << 26) ^ /usr/lib/clang/13.0.0/include/cpuid.h:137:9: note: previous definition is here #define bit_SSE20x0400 ^ 3 warnings generated. [4/20] /pobj/soundtouch-2.3.2/bin/c++ -DSOUNDTOUCH_FLOAT_SAMPLES -DSoundTouch_EXPORTS -I/pobj/soundtouch-2.3.2/soundtouch/include -O2 -pipe -DNDEBUG -fPIC -Ofast -MD -MT CMakeFiles/SoundTouch.dir/source/SoundTouch/FIFOSampleBuffer.cpp.o -MF
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 17:14:37 Modified files: mail/dovecot : Makefile distinfo mail/dovecot-pigeonhole: Makefile distinfo Log message: Update to Dovecot 2.3.20 / Dovecot-pigeonhole 0.5.20. From Brad.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 17:13:29 Modified files: games/renpy: Makefile graphics/ffmpeg: Makefile graphics/ffmpeg/pkg: PLIST graphics/openscenegraph: Makefile multimedia/handbrake: Makefile multimedia/qtav: Makefile x11/xfce4/tumbler: Makefile Log message: Disable the long deprecated and unmaintained libavresample in ffmpeg, and sync WANTLIBs. From Brad.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 17:08:18 Modified files: multimedia/transcode: Makefile multimedia/transcode/patches: patch-filter_filter_resample_c Log message: Switch transcode from avresample to swresample. From Brad.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 16:57:24 Modified files: graphics/rawtherapee: Makefile Log message: bump REVISION, missed in previous patch addition
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tra...@cvs.openbsd.org 2022/12/29 16:55:55 Modified files: devel/arduino-esp32: Makefile devel/arduino-esp32/patches: patch-platform_txt Log message: fix a missed compiler flag in devel/arduino-esp32. ok sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 16:45:12 Modified files: infrastructure/bin: modgo-gen-modules-helper Log message: add a comment on the MODGO_VERSION line output by "make modgo-gen-modules" that it shouldn't go into modules.inc; low-tech approach but hopefully people will notice it if they read their diff before sending it out ;) tb@ thinks it would help too.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 16:41:58 Modified files: net: Makefile infrastructure/db: user.list Log message: +endlessh
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 16:41:22 Log message: import ports/net/endlessh, from michi+openbsd at dataswamp/org, ok pascal@ Endlessh is an SSH tarpit that very slowly sends an endless, random SSH banner. It keeps SSH clients locked up for hours or even days at a time. The purpose is to put your real SSH server on another port and then let the script kiddies get stuck in this tarpit instead of bothering a real server. Status: Vendor Tag: sthen Release Tags: sthen_20221229 N ports/net/endlessh/Makefile N ports/net/endlessh/distinfo N ports/net/endlessh/pkg/DESCR N ports/net/endlessh/pkg/PLIST N ports/net/endlessh/pkg/endlessh.rc N ports/net/endlessh/pkg/endlessh6.rc N ports/net/endlessh/pkg/endlessh.login N ports/net/endlessh/pkg/endlessh6.login N ports/net/endlessh/patches/patch-Makefile N ports/net/endlessh/patches/patch-endlessh_1 No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 16:34:14 Modified files: lang/node : Makefile distinfo lang/node/patches: patch-Makefile patch-common_gypi patch-deps_npm_node_modules_node-gyp_gyp_pylib_gyp_generator_make_py patch-deps_v8_include_v8config_h patch-deps_v8_src_api_api_cc patch-deps_v8_src_base_atomicops_h patch-deps_v8_src_base_platform-openbsd_cc patch-deps_v8_src_base_platform-posix_cc patch-deps_v8_src_codegen_riscv64_assembler-riscv64_cc patch-deps_v8_src_codegen_riscv64_macro-assembler-riscv64_cc patch-deps_v8_src_trap-handler_h patch-deps_v8_src_trap-handler_handler-inside-posix_h patch-deps_v8_src_wasm_baseline_ia32_liftoff-assembler-ia32_h patch-lib_internal_modules_cjs_loader_js patch-lib_net_js patch-node_gyp patch-src_cares_wrap_h patch-src_env_cc patch-tools_v8_gypfiles_v8_gyp lang/node/pkg : PLIST Added files: lang/node/patches: patch-deps_base64_base64_Makefile patch-deps_llhttp_common_gypi patch-deps_v8_src_base_cpu_cc patch-deps_v8_tools_gcmole_Makefile patch-include_node_common_gypi patch-tools_v8_gypfiles_toolchain_gypi Log message: update to node-18.12.1, from Volker Schlecht, various feedback/tweaks/testing from jca@ tb@ thfr@ aisha@ and Barry K. Nathan
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 16:24:30 Modified files: security : Makefile Log message: +age-plugin-yubikey
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 16:24:06 Log message: import ports/security/age-plugin-yubikey, from Volker Schlecht, ok op@ age-plugin-yubikey is a plugin for age clients like age and rage, which enables files to be encrypted to age identities stored on YubiKeys. Status: Vendor Tag: sthen Release Tags: sthen_20221229 N ports/security/age-plugin-yubikey/Makefile N ports/security/age-plugin-yubikey/crates.inc N ports/security/age-plugin-yubikey/distinfo N ports/security/age-plugin-yubikey/pkg/DESCR N ports/security/age-plugin-yubikey/pkg/PLIST N ports/security/age-plugin-yubikey/pkg/MESSAGE No conflicts created by this import
Re: [UPDATE] lang/node 18.12.1
It took half an eternity, but my raspberry pi just successfully built a working node 18.12.1 with the attached diff. It's the same as the last one, I'm just attaching it again for convenience. So far we had successful tests on: * amd64 * i386 * aarch64 * riscv64 On 12/29/22 00:36, A Tammy wrote: On 12/27/22 11:10, Volker Schlecht wrote: bump On 12/20/22 19:15, Volker Schlecht wrote: bump once more in the hope of finding someone to eventually commit ;-) working fine for me. ok aisha going to commit this tomorrow if no one has any objections. I don't know if we have testers for this on the more esoteric archs and if should wait for them?Index: Makefile === RCS file: /cvs/ports/lang/node/Makefile,v retrieving revision 1.111 diff -u -p -r1.111 Makefile --- Makefile 13 Nov 2022 15:28:44 - 1.111 +++ Makefile 29 Dec 2022 09:17:31 - @@ -5,12 +5,11 @@ USE_WXNEEDED = Yes COMMENT = JavaScript runtime built on Chrome's V8 JavaScript engine -NODE_VERSION = v16.17.1 -PLEDGE_VER = 1.1.2 +NODE_VERSION = v18.12.1 +PLEDGE_VER = 1.1.3 DISTFILES = node-pledge-{}${PLEDGE_VER}.tar.gz:0 \ ${DISTNAME}-headers.tar.gz \ ${DISTNAME}.tar.xz -REVISION = 2 DISTNAME = node-${NODE_VERSION} PKGNAME = ${DISTNAME:S/v//g} @@ -34,7 +33,7 @@ MODULES = lang/python WANTLIB += c execinfo m pthread ${COMPILER_LIBCXX} WANTLIB += z brotlienc brotlidec WANTLIB += icudata icui18n icuuc cares nghttp2 uv -WANTLIB += lib/eopenssl11/ssl lib/eopenssl11/crypto +WANTLIB += lib/eopenssl30/ssl lib/eopenssl30/crypto COMPILER = base-clang ports-gcc @@ -71,16 +70,16 @@ SUBST_VARS += WRKDIST SUBST_VARS += NODE_VERSION SUBST_VARS += EOPENSSL_LIB -# OpenSSL used: {X,Ed}25519 via EVP, SSL_CIPHER_standard_name and 5-10 other missing symbols/defines. +# uses a wide range of OpenSSL API and only really supports boring/openssl LIB_DEPENDS += archivers/brotli \ devel/libuv \ net/libcares \ textproc/icu4c \ www/nghttp2 \ - security/openssl/1.1 + security/openssl/3.0 -EOPENSSL_LIB = ${LOCALBASE}/lib/eopenssl11 -EOPENSSL_INC = ${LOCALBASE}/include/eopenssl11 +EOPENSSL_LIB = ${LOCALBASE}/lib/eopenssl30 +EOPENSSL_INC = ${LOCALBASE}/include/eopenssl30 post-extract: mv ${WRKDIR}/node-pledge-${PLEDGE_VER} \ @@ -100,8 +99,9 @@ post-install: ${INSTALL_DATA_DIR} ${PREFIX}/share/node ${INSTALL_DATA} ${FULLDISTDIR}/${DISTNAME}-headers.tar.gz \ ${PREFIX}/share/node/${DISTNAME}-headers.tar.gz - ln -s ${TRUEPREFIX}/lib/node_modules/npm/bin/node-gyp-bin/node-gyp \ + ln -fs ${TRUEPREFIX}/lib/node_modules/npm/bin/node-gyp-bin/node-gyp \ ${PREFIX}/bin/node-gyp; + rm -r ${PREFIX}/lib/node_modules/npm/node_modules/node-pledge/build/node_gyp_bins cd ${PREFIX}/lib/node_modules/npm/node_modules && for x in *; do \ mv $$x ../..; \ ln -s ../../$$x $$x; \ Index: distinfo === RCS file: /cvs/ports/lang/node/distinfo,v retrieving revision 1.65 diff -u -p -r1.65 distinfo --- distinfo 1 Nov 2022 12:01:49 - 1.65 +++ distinfo 29 Dec 2022 09:17:31 - @@ -1,6 +1,6 @@ -SHA256 (node-pledge-1.1.2.tar.gz) = zY/JcbZ32mmtqWXXNn3/9aTh7Y3F6fAAaADDA8SYwEk= -SHA256 (node-v16.17.1-headers.tar.gz) = Ncy5XK8CzaO9aA2kNQqK5dZmp6nq46/lwqGz7ymu8Qg= -SHA256 (node-v16.17.1.tar.xz) = ZyH+tBUtVtLGs1jOOXq9Wn8drwnuLiXFAhubTT+GozA= -SIZE (node-pledge-1.1.2.tar.gz) = 3155 -SIZE (node-v16.17.1-headers.tar.gz) = 568068 -SIZE (node-v16.17.1.tar.xz) = 35661452 +SHA256 (node-pledge-1.1.3.tar.gz) = fEaXvLg6hYEJ69K+mgQFizf8DiJY2/DtyFJB/pEanVU= +SHA256 (node-v18.12.1-headers.tar.gz) = nVXuByum1aFB2wks7xoPcV99P8k4KFptknodCgx0Qvc= +SHA256 (node-v18.12.1.tar.xz) = T6QGRRvFJlmikOUs/bIWKnYL1UnaS4u+vmop8pbZON8= +SIZE (node-pledge-1.1.3.tar.gz) = 3167 +SIZE (node-v18.12.1-headers.tar.gz) = 8563785 +SIZE (node-v18.12.1.tar.xz) = 38454588 Index: patches/patch-Makefile === RCS file: /cvs/ports/lang/node/patches/patch-Makefile,v retrieving revision 1.15 diff -u -p -r1.15 patch-Makefile --- patches/patch-Makefile 18 Mar 2022 19:35:16 - 1.15 +++ patches/patch-Makefile 29 Dec 2022 09:17:31 - @@ -1,7 +1,7 @@ Index: Makefile --- Makefile.orig +++ Makefile -@@ -163,7 +163,7 @@ config.gypi: configure configure.py src/node_version.h +@@ -185,7 +185,7 @@ config.gypi: configure configure.py src/node_version.h fi .PHONY: install @@ -10,7 +10,7 @@ Index: Makefile $(PYTHON) tools/install.py $@ '$(DESTDIR)' '$(PREFIX)' .PHONY: uninstall -@@ -394,6 +394,12 @@ test/addons/.buildstamp: $(ADDONS_PREREQS) \ +@@ -416,6 +416,12 @@ test/addons/.buildstamp: $(ADDONS_PREREQS) \ # Just goes to show that recursive make really is harmful... # TODO(bnoordhuis) Force rebuild after gyp update. build-addons: | $(NODE_EXE) test/addons/.buildstamp Index: patches/patch-common_gypi
[maintainer update] sysutils/ruby-servengine 2.3.0 -> 2.3.1
https://github.com/treasure-data/serverengine/blob/v2.3.1/Changelog 2022-12-22 version 2.3.1 * Don't treat as error when worker shuts down with exit status 0 Minor tweaks in code. No PLIST changes. ok? --Kurt Index: Makefile === RCS file: /cvs/ports/sysutils/ruby-serverengine/Makefile,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 Makefile --- Makefile20 Sep 2022 16:18:53 - 1.1.1.1 +++ Makefile29 Dec 2022 21:58:51 - @@ -1,7 +1,7 @@ COMMENT = multiprocess server framework for Ruby -DISTNAME = serverengine-2.3.0 +DISTNAME = serverengine-2.3.1 CATEGORIES = sysutils Index: distinfo === RCS file: /cvs/ports/sysutils/ruby-serverengine/distinfo,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 distinfo --- distinfo20 Sep 2022 16:18:53 - 1.1.1.1 +++ distinfo29 Dec 2022 21:58:51 - @@ -1,2 +1,2 @@ -SHA256 (serverengine-2.3.0.gem) = EmPnOYIPJOXgZuyg01lTuQ+JDW0vCYMUi+32Wmd13eo= -SIZE (serverengine-2.3.0.gem) = 42496 +SHA256 (serverengine-2.3.1.gem) = 3qWTfDxITn1nzFqfcQVcqm/gj6aWLt/FM/nb83Tqehc= +SIZE (serverengine-2.3.1.gem) = 43008
net/rsync: update to 3.2.7
Straightforward maintenance update of net/rsync to 3.2.7. I looked over the list of changes, but I didn't see anything that would require special attention. This should definitely be double-checked by somebody else. Index: Makefile === RCS file: /cvs/ports/net/rsync/Makefile,v retrieving revision 1.96 diff -u -p -r1.96 Makefile --- Makefile8 Nov 2022 11:17:00 - 1.96 +++ Makefile29 Dec 2022 21:48:53 - @@ -1,13 +1,6 @@ COMMENT = mirroring/synchronization over low bandwidth links -DISTNAME = rsync-3.2.5 - -# distfile was rerolled under the same name; remove PKGNAME/DISTFILES -# when updating -PKGNAME = rsync-3.2.5pl0 -DISTFILES =rsync-3.2.5pl0{rsync-3.2.5}${EXTRACT_SUFX} -REVISION = 0 - +DISTNAME = rsync-3.2.7 CATEGORIES = net HOMEPAGE = https://rsync.samba.org/ Index: distinfo === RCS file: /cvs/ports/net/rsync/distinfo,v retrieving revision 1.34 diff -u -p -r1.34 distinfo --- distinfo15 Aug 2022 14:51:17 - 1.34 +++ distinfo29 Dec 2022 21:48:53 - @@ -1,2 +1,2 @@ -SHA256 (rsync-3.2.5pl0.tar.gz) = KsTSFjXN95GGe8N3w1ym3af1DZGaWL5FBX/VFgDGmro= -SIZE (rsync-3.2.5pl0.tar.gz) = 1129957 +SHA256 (rsync-3.2.7.tar.gz) = Tn2dP27RCHjFjF+3JKZ9rPS2qsc0CxPkiPstxBNG8rs= +SIZE (rsync-3.2.7.tar.gz) = 1149787 Index: patches/patch-authenticate_c === RCS file: /cvs/ports/net/rsync/patches/patch-authenticate_c,v retrieving revision 1.9 diff -u -p -r1.9 patch-authenticate_c --- patches/patch-authenticate_c11 Mar 2022 19:47:20 - 1.9 +++ patches/patch-authenticate_c29 Dec 2022 21:48:53 - @@ -1,12 +1,12 @@ Index: authenticate.c --- authenticate.c.orig +++ authenticate.c -@@ -349,7 +349,7 @@ void auth_client(int fd, const char *user, const char +@@ -351,7 +351,7 @@ void auth_client(int fd, const char *user, const char char pass2[MAX_DIGEST_LEN*2]; if (!user || !*user) - user = "nobody"; + user = "_rsync"; + negotiate_daemon_auth(-1, 1); if (!(pass = getpassf(password_file)) -&& !(pass = getenv("RSYNC_PASSWORD"))) { Index: patches/patch-rsync_1_md === RCS file: /cvs/ports/net/rsync/patches/patch-rsync_1_md,v retrieving revision 1.2 diff -u -p -r1.2 patch-rsync_1_md --- patches/patch-rsync_1_md15 Aug 2022 13:45:23 - 1.2 +++ patches/patch-rsync_1_md29 Dec 2022 21:48:53 - @@ -1,7 +1,7 @@ Index: rsync.1.md --- rsync.1.md.orig +++ rsync.1.md -@@ -4628,7 +4628,7 @@ file is included or excluded. +@@ -4723,7 +4723,7 @@ file is included or excluded. The USER or LOGNAME environment variables are used to determine the default username sent to an rsync daemon. If neither is set, the username defaults Index: patches/patch-rsyncd_conf_5_md === RCS file: /cvs/ports/net/rsync/patches/patch-rsyncd_conf_5_md,v retrieving revision 1.1 diff -u -p -r1.1 patch-rsyncd_conf_5_md --- patches/patch-rsyncd_conf_5_md 23 May 2022 00:24:58 - 1.1 +++ patches/patch-rsyncd_conf_5_md 29 Dec 2022 21:48:53 - @@ -1,7 +1,7 @@ Index: rsyncd.conf.5.md --- rsyncd.conf.5.md.orig +++ rsyncd.conf.5.md -@@ -474,7 +474,7 @@ the values of parameters. See the GLOBAL PARAMETERS s +@@ -502,7 +502,7 @@ the values of parameters. See the GLOBAL PARAMETERS s and from that module should take place as when the daemon was run as root. In combination with the "[gid](#)" parameter this determines what file permissions are available. The default when run by a super-user is to @@ -10,7 +10,7 @@ Index: rsyncd.conf.5.md to not try to change the user. See also the "[gid](#)" parameter. The RSYNC_USER_NAME environment variable may be used to request that rsync -@@ -492,7 +492,7 @@ the values of parameters. See the GLOBAL PARAMETERS s +@@ -520,7 +520,7 @@ the values of parameters. See the GLOBAL PARAMETERS s extra ones be set as supplemental groups. You may also specify a "`*`" as the first gid in the list, which will be replaced by all the normal groups for the transfer's user (see "[uid](#)"). The default when run by a super-user @@ -19,7 +19,7 @@ Index: rsyncd.conf.5.md other supplementary groups. The default for a non-super-user is to not change any group attributes (and indeed, your OS may not allow a non-super-user to try to change their group settings). -@@ -1170,8 +1170,8 @@ A simple rsyncd.conf file that allow anonymous rsync t +@@ -1198,8 +1198,8 @@ A simple rsyncd.conf file that allow anonymous rsync t A more sophisticated example would be: > ``` -- Christian "naddy" Weisgerber na...@mips.inka.de
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: gkoeh...@cvs.openbsd.org2022/12/29 14:20:25 Modified files: lang/ruby : Makefile.inc Log message: Configure lang/ruby --without-baseruby This prevents picking /usr/local/bin/ruby from devel/ruby-shims as baseruby, and erroring in "make fake" when baseruby can't require tool/downloader.rb. We don't need baseruby, because we are building ruby-*.tar.gz, not a git checkout of ruby. ok jeremy@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: pas...@cvs.openbsd.org 2022/12/29 14:07:28 Added files: graphics/rawtherapee/patches: patch-rtengine_dcraw_c patch-rtengine_procparams_cc patch-rtengine_rtthumbnail_cc patch-rtexif_rtexif_cc patch-rtgui_cacheimagedata_cc patch-rtgui_multilangmgr_cc patch-rtgui_options_cc Log message: Add a workaround for rawtherapee crashing on startup due to threaded use of setlocale(), see https://github.com/Beep6581/RawTherapee/issues/6452. ok kirby@ (maintainer)
[update] devel/rebar3 3.20.0
Cc: Maintainer Updates rebar3 to version 3.20 https://github.com/erlang/rebar3/releases/tag/3.20.0 * Starting with this version they bundle all of their build dependencies * Drop all DISTFILES for the bundled dependencies * Keep DISTFILE for meck around, since it's needed to run make test * Adapt patches to their new paths * Adapt substitutions to their new paths * Only unpack meck hex module in do-test Rebuilds elixir, rebuilds rabbitmq, works on the command line on amd64 Index: Makefile === RCS file: /cvs/ports/devel/rebar3/Makefile,v retrieving revision 1.7 diff -u -p -r1.7 Makefile --- Makefile 20 Jul 2022 01:09:58 - 1.7 +++ Makefile 29 Dec 2022 20:29:28 - @@ -2,7 +2,7 @@ COMMENT = Erlang build tool GH_ACCOUNT = erlang GH_PROJECT = rebar3 -GH_TAGNAME = 3.19.0 +GH_TAGNAME = 3.20.0 CATEGORIES = devel HOMEPAGE = https://www.rebar3.org MAINTAINER = Jonathan Matthew @@ -26,22 +26,11 @@ ERRORS += "Invalid FLAVOR EXTRACT_ONLY = rebar3-${GH_TAGNAME}.tar.gz -HEX_MODULES = \ - bbmustache 1.12.2 \ - certifi 2.9.0 \ - cf 0.3.1 \ - cth_readable 1.5.1 \ - erlware_commons 1.5.0 \ - eunit_formatters 0.5.0 \ - getopt 1.0.1 \ - providers 1.9.0 \ - relx 4.7.0 \ - ssl_verify_fun 1.1.6 # for tests HEX_MODULES += \ meck 0.8.13 -DISTFILES = rebar3-{}${GH_TAGNAME}.tar.gz +DISTFILES = rebar3-{}${GH_TAGNAME}.tar.gz .for _m _v in ${HEX_MODULES} DISTFILES += ${_m}-${_v}.tar:1 .endfor @@ -50,23 +39,12 @@ BUILD_DEPENDS += ${RUN_DEPENDS} SUBST_VARS = ERL_VERSION -post-extract: - # extract hex modules into _build/default/lib for the bootstrap, - # and _checkouts for rebar3 itself, otherwise it will fetch from hex - mkdir -p ${WRKSRC}/_build/default/lib -.for _m _v in ${HEX_MODULES} - mkdir -p ${WRKDIR}/${_m} - tar xf ${FULLDISTDIR}/${_m}-${_v}.tar -C ${WRKDIR}/${_m} - - mkdir -p ${WRKSRC}/_checkouts/${_m} - tar xzf ${WRKDIR}/${_m}/contents.tar.gz -C ${WRKSRC}/_checkouts/${_m} - cp -r ${WRKSRC}/_checkouts/${_m} ${WRKSRC}/_build/default/lib/ -.endfor pre-configure: - ${SUBST_CMD} ${WRKSRC}/bootstrap ${WRKSRC}/src/rebar_prv_escriptize.erl \ - ${WRKSRC}/_build/default/lib/relx/priv/templates/bin \ - ${WRKSRC}/_build/default/lib/relx/priv/templates/extended_bin + ${SUBST_CMD} ${WRKSRC}/bootstrap \ + ${WRKSRC}/apps/rebar/src/rebar_prv_escriptize.erl \ + ${WRKSRC}/vendor/relx/priv/templates/bin \ + ${WRKSRC}/vendor/relx/priv/templates/extended_bin do-build: cd ${WRKBUILD} && ${SETENV} ${MAKE_ENV} ${WRKSRC}/bootstrap @@ -76,6 +54,14 @@ do-install: PORTHOME= ${WRKDIR} do-test: +.for _m _v in ${HEX_MODULES} + mkdir -p ${WRKDIR}/${_m} + tar xf ${FULLDISTDIR}/${_m}-${_v}.tar -C ${WRKDIR}/${_m} + + mkdir -p ${WRKSRC}/_checkouts/${_m} + tar xzf ${WRKDIR}/${_m}/contents.tar.gz -C ${WRKSRC}/_checkouts/${_m} + cp -r ${WRKSRC}/_checkouts/${_m} ${WRKSRC}/_build/default/lib/ +.endfor cd ${WRKSRC} && \ ${SETENV} ${ALL_TEST_ENV} ./rebar3 escriptize && \ ${SETENV} ${ALL_TEST_ENV} ./rebar3 ct Index: distinfo === RCS file: /cvs/ports/devel/rebar3/distinfo,v retrieving revision 1.3 diff -u -p -r1.3 distinfo --- distinfo 20 Jul 2022 01:09:58 - 1.3 +++ distinfo 29 Dec 2022 20:29:28 - @@ -1,24 +1,4 @@ -SHA256 (bbmustache-1.12.2.tar) = aIszpNXMLVH1da3ws2g/xAo4MUovFQkG7c/Hf1tXezs= -SHA256 (certifi-2.9.0.tar) = Jm2ka9sG1sbTX955m8so022YXUJK18CLW7SPW1zdRkE= -SHA256 (cf-0.3.1.tar) = MV6NRH06SwK82/o5etA7u5iKbgqm9E063Q9OPDv5dnI= -SHA256 (cth_readable-1.5.1.tar) = aGVBoi7+bKWkGgR7OVFsLdKPs8reXySi8ZFFs5Z/nYA= -SHA256 (erlware_commons-1.5.0.tar) = PnxvsrpMKbDdXf6dAxtmRJ4giOzsGoFGW9n94F7X0Ns= -SHA256 (eunit_formatters-0.5.0.tar) = 1si6ITQklE5uBbvAl8MgAc3Qq+OSXQJFTyKbINaHY8k= -SHA256 (getopt-1.0.1.tar) = U+Grg7nOtlyWctPno1uAkum9ybPugHIUcaFhwQxZlZw= SHA256 (meck-0.8.13.tar) = 008BPBVttRrVfMVWiRuXIOahwd9f4uFa+ZnITWzr6xo= -SHA256 (providers-1.9.0.tar) = 0ofodEBqFQVghkKwo9tbaNato/KrABrsh+f018efwBc= -SHA256 (rebar3-3.19.0.tar.gz) = /570LHN0gEd+vfDdnZXs5TShTJbwXtr78y6a+XMoC8M= -SHA256 (relx-4.7.0.tar) = oqv6+nC846C5im5RQKyYQpuZgPg1/7RzV/BkVCL8L6M= -SHA256 (ssl_verify_fun-1.1.6.tar) = vbDSRx9FPIj/OQjnaG+G+b4yfQZcwewW+kVAGX6gRoA= -SIZE (bbmustache-1.12.2.tar) = 17920 -SIZE (certifi-2.9.0.tar) = 156160 -SIZE (cf-0.3.1.tar) = 10240 -SIZE (cth_readable-1.5.1.tar) = 19968 -SIZE (erlware_commons-1.5.0.tar) = 53760 -SIZE (eunit_formatters-0.5.0.tar) = 14848 -SIZE (getopt-1.0.1.tar) = 19456 +SHA256 (rebar3-3.20.0.tar.gz) = U+1/KUqLj7TX11mIxpGUlDgxwQTTmDKh+jAwexqFk94= SIZE (meck-0.8.13.tar) = 34304 -SIZE (providers-1.9.0.tar) = 14848 -SIZE (rebar3-3.19.0.tar.gz) = 442286 -SIZE (relx-4.7.0.tar) = 76288 -SIZE (ssl_verify_fun-1.1.6.tar) = 14848 +SIZE (rebar3-3.20.0.tar.gz) = 785992 Index: patches/patch-_build_default_lib_relx_priv_templates_bin === RCS file:
Re: [update][proposal] sysutils/nnn add nerd font flavor
Am Thu, Dec 29, 2022 at 07:19:23PM +0100 schrieb Joel Carnat: > Le 29/12/2022 à 12:42, Stuart Henderson a écrit : > > For the fonts I think I'd probably pick a couple that already exist in > > ports/fonts. Simplest approach for ports layout is probably like > > ports/fonts/nerd-fonts/{terminus,profont,noto) or similar, using the > > release zips rather than downloading the full repo, factoring out as > > much as possible to Makefile.inc. > Attached is an archive to deploy such > ports/fonts/nerd-fonts/{terminus,profont,noto}. I'm sorry, I couldn't find > the proper cvs command to generate the diff file for those. > > Does this seem right? Just tested with terminus and codenewroman: it works. > > Regards, > Joel C.
Re: [update][proposal] sysutils/nnn add nerd font flavor
Le 29/12/2022 à 12:42, Stuart Henderson a écrit : For the fonts I think I'd probably pick a couple that already exist in ports/fonts. Simplest approach for ports layout is probably like ports/fonts/nerd-fonts/{terminus,profont,noto) or similar, using the release zips rather than downloading the full repo, factoring out as much as possible to Makefile.inc. Attached is an archive to deploy such ports/fonts/nerd-fonts/{terminus,profont,noto}. I'm sorry, I couldn't find the proper cvs command to generate the diff file for those. Does this seem right? Regards, Joel C. port-nerd-fonts.tar.gz Description: application/gzip
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2022/12/29 10:01:33 Modified files: net/tg_owt : Makefile distinfo Log message: bump bundled libyuv to latest HEAD no PLIST or symbol change
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2022/12/29 09:58:45 Modified files: www/nextcloud/24: Tag: OPENBSD_7_2 Makefile distinfo www/nextcloud/24/pkg: Tag: OPENBSD_7_2 PLIST Log message: www/nextcloud/24: MFC update to 24.0.8. ok gonzalo@ (maintainer)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2022/12/29 09:57:23 Modified files: net/tdesktop : Makefile distinfo Removed files: net/tdesktop/patches: patch-Telegram_SourceFiles_platform_linux_notifications_manager_linux_cpp patch-Telegram_SourceFiles_platform_linux_specific_linux_cpp patch-Telegram_lib_base_base_platform_linux_base_last_input_linux_cpp Log message: update to tdesktop 4.4.3 Drop local clang/glibmm workarounds, upstream fixed it. This is the first release (I can remember) that builds and runs without any pending PRs or local patches on OpenBSD.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2022/12/29 09:41:29 Modified files: net/tg_owt/pkg : PLIST Log message: regen, missed in previous "Update to latest master, tdesktop 4.4.2 requires newer WebRTC"
Re: [new] age-plugin-yubikey
On 2022/12/29 16:28, Omar Polo wrote: > On 2022/12/29 15:06:36 +, Stuart Henderson wrote: > > Thanks - this is OK sthen@ for someone to import > > can't runtest it but ok for me too. > > i would have preferred if the note was added in DESCR rather than > MESSAGE but *shrug*. (MESSAGE can be annoying because it's printed > every time the package is installed/updated.) it's not printed at update time (unless the message actually changes), only install > nitpicking: maybe who imports it wants to remove the double empty line > in the Makefile. ack. will do it later if nobody beats me to it.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2022/12/29 08:57:33 Modified files: net/tdesktop : Makefile distinfo Log message: update to tdesktop 4.4.2 All PRs got merged.
Re: [maintainer update] difftastic 0.40
On 2022/12/29 14:19:36 +0100, Volker Schlecht wrote: > Here's the diff with your suggestions and difftastic updated to 0.40 > https://github.com/Wilfred/difftastic/releases/tag/0.40.0 committed, thanks!
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: o...@cvs.openbsd.org2022/12/29 08:31:32 Modified files: devel/difftastic: Makefile crates.inc distinfo Added files: devel/difftastic/patches: patch-Cargo_lock patch-Cargo_toml Log message: update devel/difftastic to 0.40.0 diff from MAINTAINER Volker Schlecht, thanks! Some minor tweaks by me.
Re: [new] age-plugin-yubikey
On 2022/12/29 15:06:36 +, Stuart Henderson wrote: > Thanks - this is OK sthen@ for someone to import can't runtest it but ok for me too. i would have preferred if the note was added in DESCR rather than MESSAGE but *shrug*. (MESSAGE can be annoying because it's printed every time the package is installed/updated.) nitpicking: maybe who imports it wants to remove the double empty line in the Makefile.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: b...@cvs.openbsd.org2022/12/29 08:10:04 Modified files: audio/ncspot : Makefile crates.inc distinfo Log message: Update ncspot to 0.12.0 Changes: https://github.com/hrkfdn/ncspot/releases/tag/v0.12.0 >From Henrik Friedrichsen (maintainer). Thank you!
Re: [new] age-plugin-yubikey
Update with a brief MESSAGE is attached! On 12/29/22 15:53, Stuart Henderson wrote: On 2022/12/29 14:30, Volker Schlecht wrote: Works for me. Basically OK sthen for import, though it could probably do with a RUN_DEPENDS on security/age, and possibly also worth adding a one-line pkg/MESSAGE (or at least a quick note in DESCR) saying that pcscd need to be running to use it? I can add both, however age-plugin-yubikey *does* run here without age, that is you can provision identities on a Yubikey with it. It also integrates with rage and other implementations of the age spec, none of which are in ports yet, but might be in the future. Also, the error message when pcscd isn't running is pretty obvious: Error: Error while communicating with YubiKey: PC/SC error Cause: The Smart card resource manager is not running Does a message or note really make sense then? If the error said "pcscd isn't running" then I'd agree. But it's all a bit opaque and if someone isn't already familiar with the stack needed for running this (for example, someone might be used to SSH with U2F keys and could easily be forgiven for thinking it's a similar situation, and that doesn't need anything external) I think it's useful to give them a bit of a better nudge in the right direction. age-plugin-yubikey.tar.gz Description: application/gzip
Re: [new] age-plugin-yubikey
On 2022/12/29 14:30, Volker Schlecht wrote: > > > Works for me. Basically OK sthen for import, though it could probably do > > with a RUN_DEPENDS on security/age, and possibly also worth adding a > > one-line pkg/MESSAGE (or at least a quick note in DESCR) saying that > > pcscd need to be running to use it? > > I can add both, however age-plugin-yubikey *does* run here without age, that > is you can provision identities on a Yubikey with it. > > It also integrates with rage and other implementations of the age spec, none > of which are in ports yet, but might be in the future. > > Also, the error message when pcscd isn't running is pretty obvious: > > Error: Error while communicating with YubiKey: PC/SC error > Cause: The Smart card resource manager is not running > > Does a message or note really make sense then? If the error said "pcscd isn't running" then I'd agree. But it's all a bit opaque and if someone isn't already familiar with the stack needed for running this (for example, someone might be used to SSH with U2F keys and could easily be forgiven for thinking it's a similar situation, and that doesn't need anything external) I think it's useful to give them a bit of a better nudge in the right direction.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2022/12/29 07:53:25 Modified files: net/tg_owt/patches: patch-CMakeLists_txt Log message: regen, missed in previous
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bl...@cvs.openbsd.org 2022/12/29 07:52:51 Modified files: devel/p5-XS-Parse-Keyword: Makefile distinfo Log message: update XS-Parse-Keyword to 0.31
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2022/12/29 07:52:29 Modified files: net/tg_owt : Makefile distinfo Log message: Update to latest master, tdesktop 4.4.2 requires newer WebRTC All PRs got merged.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bl...@cvs.openbsd.org 2022/12/29 07:45:04 Modified files: devel/p5-Syntax-Keyword-Match: Makefile distinfo Log message: update p5-Syntax-Keyword-Match to 0.10
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: tra...@cvs.openbsd.org 2022/12/29 07:30:34 Modified files: cad/coil64 : Makefile distinfo Log message: Update cad/coil64 to v2.1.26. ok benoit@
Re: [update] www/py-bokeh to 3.0.3
On 2022/12/26 14:59, Lucas Raab wrote: > On Mon, Dec 26, 2022 at 11:33:25AM +, Stuart Henderson wrote: > > On 2022/12/24 15:53, Lucas Raab wrote: > > > Hello, > > > > > > Here's an update for bokeh to the latest, but needs more eyes and tests. > > > I've had it rolling around in my tree for a bit and works, but I'm not > > > going to pretend I extensively use its capabilities. > > > > > > Attached to this are two new RDEPs, the latter of which should be > > > applicable to a future matplotlib update: > > > * geo/py-xyzservices > > > * math/py-contourpy > > > > contourpy needs WANTLIB and COMPILER=base-clang ports-gcc, otherwise > > ok to import those. > > > Ah! Thanks for the reminder > > > > Tests are disabled as the pypi-hosted dist hasn't provided tests. > > > Setting GH_ does obviously get the tests, but then the python build > > > process wants to rebuild the css files and wants node to do so. There > > > are also many, many more TDEPs that would need to imported. > > > > Small nit below otherwise bokeh looks good to me too (I can't do > > extensive testing either, but I don't think that's really needed > > anyway, as long as the basics work if there are problems they're > > unlikely to be OpenBSD-specific). > > > > > RUN_DEPENDS= devel/py-dateutil${MODPY_FLAVOR} \ > > > > Looks like this can be dropped, I don't see it used any more > > Heh, I saw the removal in the release notes, but assumed it was a > used-if-available vs outright removal. I should have looked closer... > > > > > > devel/py-typing-extensions${MODPY_FLAVOR}>=3.10.0.0 \ > > > > (This is no longer a listed dependency but is still imported in some > > cases so makes sense to keep it listed. typing.extensions is part > > of core Python and as of 3.10 many ports can use that instead of > > typing_extensions which is an external module). > > > > Sounds good. Updated diff attached also fixing the order of numpy and > pandas in RDEPs. OK. fwiw I would add a "not in pypi distfile" comment next to NO_TEST (and maybe drop the "several TDEPs don't exist" in this case; when I tried running tests in older versions, we didn't have everything listed, but a decent number of tests ran anyway)
Re: [new] age-plugin-yubikey
Works for me. Basically OK sthen for import, though it could probably do with a RUN_DEPENDS on security/age, and possibly also worth adding a one-line pkg/MESSAGE (or at least a quick note in DESCR) saying that pcscd need to be running to use it? I can add both, however age-plugin-yubikey *does* run here without age, that is you can provision identities on a Yubikey with it. It also integrates with rage and other implementations of the age spec, none of which are in ports yet, but might be in the future. Also, the error message when pcscd isn't running is pretty obvious: Error: Error while communicating with YubiKey: PC/SC error Cause: The Smart card resource manager is not running Does a message or note really make sense then?
Re: [new] age-plugin-yubikey
On 2022/12/29 13:21, Stuart Henderson wrote: > On 2022/12/29 11:00, Volker Schlecht wrote: > > Hi Omar! > > > > On 12/29/22 10:15, Omar Polo wrote: > > > > > it fails to build if security/pcsc-lite is not installed: > > > > Fixed. I applied your other suggestions as well except for > > > > > +DISTFILES+= ${GH_DISTNAME} > > > > s/DISTNAME/DISTFILE/ ;-) > > > > As always, thanks for your feedback! > > > > cu, > > Volker > > Works for me. Basically OK sthen for import, though it could probably do > with a RUN_DEPENDS on security/age, and possibly also worth adding a ...or possibly even RUN_DEPENDS on security/age>=1.1.1 because the error when running it with 1.0 isn't very obvious. > one-line pkg/MESSAGE (or at least a quick note in DESCR) saying that > pcscd need to be running to use it? >
Re: [new] age-plugin-yubikey
On 2022/12/29 11:00, Volker Schlecht wrote: > Hi Omar! > > On 12/29/22 10:15, Omar Polo wrote: > > > it fails to build if security/pcsc-lite is not installed: > > Fixed. I applied your other suggestions as well except for > > > +DISTFILES+=${GH_DISTNAME} > > s/DISTNAME/DISTFILE/ ;-) > > As always, thanks for your feedback! > > cu, > Volker Works for me. Basically OK sthen for import, though it could probably do with a RUN_DEPENDS on security/age, and possibly also worth adding a one-line pkg/MESSAGE (or at least a quick note in DESCR) saying that pcscd need to be running to use it?
[maintainer update] difftastic 0.40
Here's the diff with your suggestions and difftastic updated to 0.40 https://github.com/Wilfred/difftastic/releases/tag/0.40.0 On 12/29/22 13:40, Omar Polo wrote: On 2022/12/27 17:11:39 +0100, Volker Schlecht wrote: bump quickly tested with some C files, works fine and is interesting. slightly improved diff. need to set COMPILER when using c++, rearranged WANTLIBs, used GH_DISTFILE and move the include crates.inc below. ok?Index: Makefile === RCS file: /cvs/ports/devel/difftastic/Makefile,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 Makefile --- Makefile 8 Nov 2022 14:54:43 - 1.1.1.1 +++ Makefile 29 Dec 2022 13:16:59 - @@ -2,7 +2,7 @@ COMMENT = diff tool that compares files GH_ACCOUNT = Wilfred GH_PROJECT = difftastic -GH_TAGNAME = 0.37.0 +GH_TAGNAME = 0.40.0 HOMEPAGE = https://difftastic.wilfred.me.uk/ @@ -13,14 +13,11 @@ CATEGORIES = devel # MIT PERMIT_PACKAGE = Yes -# as devel/cargo MODULES adds DISTFILES, GH_* didn't -DISTFILES += ${DISTNAME}${EXTRACT_SUFX} +DISTFILES += ${GH_DISTFILE} MODULES = devel/cargo BUILD_DEPENDS = textproc/mdbook - -WANTLIB += c c++ c++abi m pthread - -.include "crates.inc" +COMPILER = base-clang ports-gcc +WANTLIB += ${MODCARGO_WANTLIB} ${COMPILER_LIBCXX} m CONFIGURE_STYLE = cargo @@ -41,4 +38,5 @@ post-install: ${INSTALL_DATA_DIR} ${DOCDIR} cp -r ${WRKSRC}/html_manual/* ${DOCDIR} +.include "crates.inc" .include Index: crates.inc === RCS file: /cvs/ports/devel/difftastic/crates.inc,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 crates.inc --- crates.inc 8 Nov 2022 14:54:43 - 1.1.1.1 +++ crates.inc 29 Dec 2022 13:16:59 - @@ -67,7 +67,7 @@ MODCARGO_CRATES += terminal_size 0.1.17 MODCARGO_CRATES += terminal_size 0.2.1 # MIT OR Apache-2.0 MODCARGO_CRATES += textwrap 0.15.0 # MIT MODCARGO_CRATES += tree-sitter 0.20.9 # MIT -MODCARGO_CRATES += tree_magic_mini 3.0.3 # MIT +MODCARGO_CRATES += tree_magic_mini 3.0.3 # MIT MODCARGO_CRATES += typed-arena 2.0.1 # MIT MODCARGO_CRATES += unicode-ident 1.0.0 # MIT OR Apache-2.0 MODCARGO_CRATES += unicode-width 0.1.9 # MIT/Apache-2.0 Index: distinfo === RCS file: /cvs/ports/devel/difftastic/distinfo,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 distinfo --- distinfo 8 Nov 2022 14:54:43 - 1.1.1.1 +++ distinfo 29 Dec 2022 13:16:59 - @@ -85,7 +85,7 @@ SHA256 (cargo/windows_i686_msvc-0.36.1.t SHA256 (cargo/windows_x86_64_gnu-0.36.1.tar.gz) = Tc0XG4d2xBuXUh5doSei2GrSgBFIB9Cyqx5GK8dk2eE= SHA256 (cargo/windows_x86_64_msvc-0.36.1.tar.gz) = yBHKSoyFPvQgq9hZK6U927rJBBD6tpA7PnmXKmMfdoA= SHA256 (cargo/wu-diff-0.1.2.tar.gz) = jj5nNfzeBkMocNuNydfjqxuTcnwU6u8ymWlCYpnyiJM= -SHA256 (difftastic-0.37.0.tar.gz) = pJoym83hhXToFSsFXVu9KhCqnISSymd/v4lj+tMiCUk= +SHA256 (difftastic-0.40.0.tar.gz) = xURXP5TY0eRkaDyd/V1jAOBQL8yb6jSdLJQiz05M70E= SIZE (cargo/aho-corasick-0.7.18.tar.gz) = 112923 SIZE (cargo/ansi_term-0.12.1.tar.gz) = 24838 SIZE (cargo/atty-0.2.14.tar.gz) = 5470 @@ -173,4 +173,4 @@ SIZE (cargo/windows_i686_msvc-0.36.1.tar SIZE (cargo/windows_x86_64_gnu-0.36.1.tar.gz) = 790934 SIZE (cargo/windows_x86_64_msvc-0.36.1.tar.gz) = 661999 SIZE (cargo/wu-diff-0.1.2.tar.gz) = 33439 -SIZE (difftastic-0.37.0.tar.gz) = 28624498 +SIZE (difftastic-0.40.0.tar.gz) = 28787155 Index: patches/patch-Cargo_lock === RCS file: patches/patch-Cargo_lock diff -N patches/patch-Cargo_lock --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-Cargo_lock 29 Dec 2022 13:16:59 - @@ -0,0 +1,17 @@ +Revert upstream patch to tree_magic_mini. We cannot pull from github during build. +https://github.com/Wilfred/difftastic/commit/b662a2220b9e9eae5f69ecbc31195bcfcc5216e1 +Index: Cargo.lock +--- Cargo.lock.orig Cargo.lock +@@ -637,8 +637,9 @@ dependencies = [ + + [[package]] + name = "tree_magic_mini" +-version = "3.0.2" +-source = "git+https://github.com/Wilfred/tree_magic?branch=fix-panic-for-empty-strings#13dd6dda15c7062bd8f7dd5bc9bb5b16ce9ee613; ++version = "3.0.3" ++source = "registry+https://github.com/rust-lang/crates.io-index; ++checksum = "91adfd0607cacf6e4babdb870e9bec4037c1c4b151cfd279ccefc5e0c7feaa6d" + dependencies = [ + "bytecount", + "fnv", Index: patches/patch-Cargo_toml === RCS file: patches/patch-Cargo_toml diff -N patches/patch-Cargo_toml --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-Cargo_toml 29 Dec 2022 13:16:59 - @@ -0,0 +1,21 @@ +Revert upstream patch to tree_magic_mini. We cannot pull from github during build. +https://github.com/Wilfred/difftastic/commit/b662a2220b9e9eae5f69ecbc31195bcfcc5216e1 +Index: Cargo.toml +--- Cargo.toml.orig Cargo.toml +@@ -51,7 +51,7 @@ const_format = "0.2.22" + owo-colors = "3.3.0" +
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 06:03:41 Modified files: security/osv-scanner: Makefile Log message: mark osv-scanner as ONLY_FOR_ARCHS (the intersection of archs with Go and LP64 archs); constant in a dependency overflows 32-bit int
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 06:00:49 Modified files: sysutils/py-platformdirs: Makefile distinfo Log message: update to py3-platformdirs-2.6.2
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 06:00:48 Modified files: devel/py-tox : Makefile distinfo Log message: update to py3-tox-4.0.19
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 06:00:47 Modified files: net/librenms : Makefile distinfo net/librenms/patches: patch-LibreNMS_Config_php patch-daily_sh net/librenms/pkg: PLIST Log message: update to librenms-22.12.0
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 06:00:47 Modified files: devel/libmagic : Makefile distinfo Log message: update to libmagic-5.44
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2022/12/29 06:00:49 Modified files: sysutils/py-filelock: Makefile distinfo sysutils/py-filelock/pkg: PLIST Removed files: sysutils/py-filelock/patches: patch-pyproject_toml Log message: update to py3-filelock-3.9.0
[UPDATE] audio/ncspot to 0.12.0
Attached diff updates ncspot to 0.12.0. Changes: Maintenance - Fix: Handle explicit field from Spotify API (#844, ramsayleung/rspotify#332) - Fix: Print "Connecting to Spotify" before setting up Cursive backend by @pghvlaans (#969) - Link to Flatpak image by @thorpelawrence (#997) - Make entry point synchronous and switch to global runtime instead by @ThomasFrans (#996) - Document queue module by @ThomasFrans (#1009) Features - Add "save/unsave album" to context menu (#964) - Automatically save backtraces to cache folder by @ThomasFrans (#988) - Create IPC socket on UNIX platforms for remote control and external track display (#1018) Tested on amd64.Index: Makefile === RCS file: /cvs/ports/audio/ncspot/Makefile,v retrieving revision 1.35 diff -u -p -r1.35 Makefile --- Makefile21 Nov 2022 15:52:49 - 1.35 +++ Makefile29 Dec 2022 12:56:23 - @@ -7,7 +7,7 @@ COMMENT = ncurses Spotify client GH_ACCOUNT = hrkfdn GH_PROJECT = ncspot -GH_TAGNAME = v0.11.2 +GH_TAGNAME = v0.12.0 CATEGORIES = audio Index: crates.inc === RCS file: /cvs/ports/audio/ncspot/crates.inc,v retrieving revision 1.13 diff -u -p -r1.13 crates.inc --- crates.inc 21 Nov 2022 15:52:50 - 1.13 +++ crates.inc 29 Dec 2022 12:56:23 - @@ -1,82 +1,84 @@ -MODCARGO_CRATES += addr2line 0.17.0 # Apache-2.0 OR MIT +MODCARGO_CRATES += addr2line 0.19.0 # Apache-2.0 OR MIT MODCARGO_CRATES += adler 1.0.2 # 0BSD OR MIT OR Apache-2.0 MODCARGO_CRATES += aes 0.6.0 # MIT OR Apache-2.0 MODCARGO_CRATES += aes-ctr 0.6.0 # MIT OR Apache-2.0 MODCARGO_CRATES += aes-soft0.6.4 # MIT OR Apache-2.0 MODCARGO_CRATES += aesni 0.10.0 # MIT OR Apache-2.0 -MODCARGO_CRATES += ahash 0.8.0 # MIT OR Apache-2.0 -MODCARGO_CRATES += aho-corasick0.7.19 # Unlicense/MIT +MODCARGO_CRATES += ahash 0.8.2 # MIT OR Apache-2.0 +MODCARGO_CRATES += aho-corasick0.7.20 # Unlicense OR MIT MODCARGO_CRATES += alsa0.6.0 # Apache-2.0/MIT MODCARGO_CRATES += alsa-sys0.3.1 # MIT MODCARGO_CRATES += android_system_properties 0.1.5 # MIT/Apache-2.0 -MODCARGO_CRATES += async-trait 0.1.57 # MIT OR Apache-2.0 -MODCARGO_CRATES += atty0.2.14 # MIT +MODCARGO_CRATES += async-trait 0.1.60 # MIT OR Apache-2.0 MODCARGO_CRATES += autocfg 1.1.0 # Apache-2.0 OR MIT -MODCARGO_CRATES += backtrace 0.3.66 # MIT/Apache-2.0 -MODCARGO_CRATES += base64 0.13.0 # MIT/Apache-2.0 -MODCARGO_CRATES += bindgen 0.59.2 # BSD-3-Clause +MODCARGO_CRATES += backtrace 0.3.67 # MIT OR Apache-2.0 +MODCARGO_CRATES += base64 0.13.1 # MIT/Apache-2.0 +MODCARGO_CRATES += bindgen 0.61.0 # BSD-3-Clause MODCARGO_CRATES += bitflags1.3.2 # MIT/Apache-2.0 MODCARGO_CRATES += block 0.1.6 # MIT MODCARGO_CRATES += block-buffer0.9.0 # MIT OR Apache-2.0 MODCARGO_CRATES += block-buffer0.10.3 # MIT OR Apache-2.0 -MODCARGO_CRATES += bumpalo 3.11.0 # MIT/Apache-2.0 +MODCARGO_CRATES += bumpalo 3.11.1 # MIT/Apache-2.0 MODCARGO_CRATES += bytecount 0.6.3 # Apache-2.0/MIT MODCARGO_CRATES += byteorder 1.4.3 # Unlicense OR MIT -MODCARGO_CRATES += bytes 1.2.1 # MIT -MODCARGO_CRATES += cc 1.0.73 # MIT/Apache-2.0 +MODCARGO_CRATES += bytes 1.3.0 # MIT +MODCARGO_CRATES += cc 1.0.78 # MIT OR Apache-2.0 MODCARGO_CRATES += cesu8 1.1.0 # Apache-2.0/MIT MODCARGO_CRATES += cexpr 0.6.0 # Apache-2.0/MIT MODCARGO_CRATES += cfg-if 1.0.0 # MIT/Apache-2.0 -MODCARGO_CRATES += chrono 0.4.22 # MIT/Apache-2.0 -MODCARGO_CRATES += chunked_transfer1.4.0 # Apache-2.0 +MODCARGO_CRATES += chrono 0.4.23 # MIT/Apache-2.0 +MODCARGO_CRATES += chunked_transfer1.4.1 # MIT OR Apache-2.0 MODCARGO_CRATES += cipher 0.2.5 # MIT OR Apache-2.0 MODCARGO_CRATES += clang-sys 1.4.0 # Apache-2.0 -MODCARGO_CRATES += clap4.0.15 # MIT OR Apache-2.0 +MODCARGO_CRATES += clap4.0.32 # MIT OR Apache-2.0 MODCARGO_CRATES += clap_lex0.3.0 # MIT OR Apache-2.0 MODCARGO_CRATES += clipboard 0.5.0 # MIT / Apache-2.0 MODCARGO_CRATES += clipboard-win 2.2.0 # MIT MODCARGO_CRATES += codespan-reporting 0.11.1 # Apache-2.0 MODCARGO_CRATES += combine 4.6.6 # MIT -MODCARGO_CRATES += cookie 0.16.1 # MIT OR Apache-2.0 +MODCARGO_CRATES += cookie 0.16.2 # MIT OR Apache-2.0 MODCARGO_CRATES += cookie_store0.16.1 # MIT/Apache-2.0 MODCARGO_CRATES += core-foundation 0.9.3 # MIT / Apache-2.0 MODCARGO_CRATES += core-foundation-sys 0.8.3 # MIT / Apache-2.0 MODCARGO_CRATES +=
Re: [maintainer update] difftastic 0.39
Yesterday they released 0.40, which promises another round of performance improvements ... I'd pick up your suggestions and propose a patch to 0.40, if that works for you? On 12/29/22 13:40, Omar Polo wrote: On 2022/12/27 17:11:39 +0100, Volker Schlecht wrote: bump quickly tested with some C files, works fine and is interesting. slightly improved diff. need to set COMPILER when using c++, rearranged WANTLIBs, used GH_DISTFILE and move the include crates.inc below. ok? Index: Makefile === RCS file: /home/cvs/ports/devel/difftastic/Makefile,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 Makefile --- Makefile8 Nov 2022 14:54:43 - 1.1.1.1 +++ Makefile29 Dec 2022 12:37:12 - @@ -2,7 +2,7 @@ COMMENT = diff tool that compares files GH_ACCOUNT = Wilfred GH_PROJECT = difftastic -GH_TAGNAME = 0.37.0 +GH_TAGNAME = 0.39.0 HOMEPAGE = https://difftastic.wilfred.me.uk/ @@ -14,13 +14,13 @@ CATEGORIES = devel PERMIT_PACKAGE = Yes # as devel/cargo MODULES adds DISTFILES, GH_* didn't -DISTFILES += ${DISTNAME}${EXTRACT_SUFX} +DISTFILES += ${GH_DISTFILE} MODULES = devel/cargo BUILD_DEPENDS = textproc/mdbook -WANTLIB += c c++ c++abi m pthread +COMPILER = base-clang ports-gcc -.include "crates.inc" +WANTLIB += ${MODCARGO_WANTLIB} ${COMPILER_LIBCXX} m CONFIGURE_STYLE = cargo @@ -40,5 +40,7 @@ pre-install: post-install: ${INSTALL_DATA_DIR} ${DOCDIR} cp -r ${WRKSRC}/html_manual/* ${DOCDIR} + +.include "crates.inc" .include Index: distinfo === RCS file: /home/cvs/ports/devel/difftastic/distinfo,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 distinfo --- distinfo8 Nov 2022 14:54:43 - 1.1.1.1 +++ distinfo29 Dec 2022 12:12:11 - @@ -85,7 +85,7 @@ SHA256 (cargo/windows_i686_msvc-0.36.1.t SHA256 (cargo/windows_x86_64_gnu-0.36.1.tar.gz) = Tc0XG4d2xBuXUh5doSei2GrSgBFIB9Cyqx5GK8dk2eE= SHA256 (cargo/windows_x86_64_msvc-0.36.1.tar.gz) = yBHKSoyFPvQgq9hZK6U927rJBBD6tpA7PnmXKmMfdoA= SHA256 (cargo/wu-diff-0.1.2.tar.gz) = jj5nNfzeBkMocNuNydfjqxuTcnwU6u8ymWlCYpnyiJM= -SHA256 (difftastic-0.37.0.tar.gz) = pJoym83hhXToFSsFXVu9KhCqnISSymd/v4lj+tMiCUk= +SHA256 (difftastic-0.39.0.tar.gz) = ugVGTYpcdzdMuaB9kB7aVsZ/TVWN71FUVLxLdDHIyjg= SIZE (cargo/aho-corasick-0.7.18.tar.gz) = 112923 SIZE (cargo/ansi_term-0.12.1.tar.gz) = 24838 SIZE (cargo/atty-0.2.14.tar.gz) = 5470 @@ -173,4 +173,4 @@ SIZE (cargo/windows_i686_msvc-0.36.1.tar SIZE (cargo/windows_x86_64_gnu-0.36.1.tar.gz) = 790934 SIZE (cargo/windows_x86_64_msvc-0.36.1.tar.gz) = 661999 SIZE (cargo/wu-diff-0.1.2.tar.gz) = 33439 -SIZE (difftastic-0.37.0.tar.gz) = 28624498 +SIZE (difftastic-0.39.0.tar.gz) = 28786273 Index: patches/patch-Cargo_lock === RCS file: patches/patch-Cargo_lock diff -N patches/patch-Cargo_lock --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-Cargo_lock29 Dec 2022 12:37:28 - @@ -0,0 +1,17 @@ +Revert upstream patch to tree_magic_mini. We cannot pull from github during build. +https://github.com/Wilfred/difftastic/commit/b662a2220b9e9eae5f69ecbc31195bcfcc5216e1 +Index: Cargo.lock +--- Cargo.lock.orig Cargo.lock +@@ -637,8 +637,9 @@ dependencies = [ + + [[package]] + name = "tree_magic_mini" +-version = "3.0.2" +-source = "git+https://github.com/Wilfred/tree_magic?branch=fix-panic-for-empty-strings#13dd6dda15c7062bd8f7dd5bc9bb5b16ce9ee613; ++version = "3.0.3" ++source = "registry+https://github.com/rust-lang/crates.io-index; ++checksum = "91adfd0607cacf6e4babdb870e9bec4037c1c4b151cfd279ccefc5e0c7feaa6d" + dependencies = [ + "bytecount", + "fnv", Index: patches/patch-Cargo_toml === RCS file: patches/patch-Cargo_toml diff -N patches/patch-Cargo_toml --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-Cargo_toml29 Dec 2022 12:37:28 - @@ -0,0 +1,21 @@ +Revert upstream patch to tree_magic_mini. We cannot pull from github during build. +https://github.com/Wilfred/difftastic/commit/b662a2220b9e9eae5f69ecbc31195bcfcc5216e1 +Index: Cargo.toml +--- Cargo.toml.orig Cargo.toml +@@ -51,7 +51,7 @@ const_format = "0.2.22" + owo-colors = "3.3.0" + wu-diff = "0.1.2" + rayon = "1.5.2" +-tree_magic_mini = "3.0.2" ++tree_magic_mini = "3.0.3" + bumpalo = "3.9.1" + unicode-width = "0.1.9" + term_size = "0.3.2" +@@ -73,6 +73,3 @@ path = "src/main.rs" + + [features] + +-[patch.crates-io] +-# https://github.com/mbrubeck/tree_magic/pull/13 hasn't been merged yet. +-tree_magic_mini = { git = 'https://github.com/Wilfred/tree_magic', branch = 'fix-panic-for-empty-strings' }
Re: [maintainer update] difftastic 0.39
On 2022/12/27 17:11:39 +0100, Volker Schlecht wrote: > bump quickly tested with some C files, works fine and is interesting. slightly improved diff. need to set COMPILER when using c++, rearranged WANTLIBs, used GH_DISTFILE and move the include crates.inc below. ok? Index: Makefile === RCS file: /home/cvs/ports/devel/difftastic/Makefile,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 Makefile --- Makefile8 Nov 2022 14:54:43 - 1.1.1.1 +++ Makefile29 Dec 2022 12:37:12 - @@ -2,7 +2,7 @@ COMMENT = diff tool that compares files GH_ACCOUNT = Wilfred GH_PROJECT = difftastic -GH_TAGNAME = 0.37.0 +GH_TAGNAME = 0.39.0 HOMEPAGE = https://difftastic.wilfred.me.uk/ @@ -14,13 +14,13 @@ CATEGORIES =devel PERMIT_PACKAGE = Yes # as devel/cargo MODULES adds DISTFILES, GH_* didn't -DISTFILES += ${DISTNAME}${EXTRACT_SUFX} +DISTFILES += ${GH_DISTFILE} MODULES = devel/cargo BUILD_DEPENDS =textproc/mdbook -WANTLIB += c c++ c++abi m pthread +COMPILER = base-clang ports-gcc -.include "crates.inc" +WANTLIB += ${MODCARGO_WANTLIB} ${COMPILER_LIBCXX} m CONFIGURE_STYLE = cargo @@ -40,5 +40,7 @@ pre-install: post-install: ${INSTALL_DATA_DIR} ${DOCDIR} cp -r ${WRKSRC}/html_manual/* ${DOCDIR} + +.include "crates.inc" .include Index: distinfo === RCS file: /home/cvs/ports/devel/difftastic/distinfo,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 distinfo --- distinfo8 Nov 2022 14:54:43 - 1.1.1.1 +++ distinfo29 Dec 2022 12:12:11 - @@ -85,7 +85,7 @@ SHA256 (cargo/windows_i686_msvc-0.36.1.t SHA256 (cargo/windows_x86_64_gnu-0.36.1.tar.gz) = Tc0XG4d2xBuXUh5doSei2GrSgBFIB9Cyqx5GK8dk2eE= SHA256 (cargo/windows_x86_64_msvc-0.36.1.tar.gz) = yBHKSoyFPvQgq9hZK6U927rJBBD6tpA7PnmXKmMfdoA= SHA256 (cargo/wu-diff-0.1.2.tar.gz) = jj5nNfzeBkMocNuNydfjqxuTcnwU6u8ymWlCYpnyiJM= -SHA256 (difftastic-0.37.0.tar.gz) = pJoym83hhXToFSsFXVu9KhCqnISSymd/v4lj+tMiCUk= +SHA256 (difftastic-0.39.0.tar.gz) = ugVGTYpcdzdMuaB9kB7aVsZ/TVWN71FUVLxLdDHIyjg= SIZE (cargo/aho-corasick-0.7.18.tar.gz) = 112923 SIZE (cargo/ansi_term-0.12.1.tar.gz) = 24838 SIZE (cargo/atty-0.2.14.tar.gz) = 5470 @@ -173,4 +173,4 @@ SIZE (cargo/windows_i686_msvc-0.36.1.tar SIZE (cargo/windows_x86_64_gnu-0.36.1.tar.gz) = 790934 SIZE (cargo/windows_x86_64_msvc-0.36.1.tar.gz) = 661999 SIZE (cargo/wu-diff-0.1.2.tar.gz) = 33439 -SIZE (difftastic-0.37.0.tar.gz) = 28624498 +SIZE (difftastic-0.39.0.tar.gz) = 28786273 Index: patches/patch-Cargo_lock === RCS file: patches/patch-Cargo_lock diff -N patches/patch-Cargo_lock --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-Cargo_lock29 Dec 2022 12:37:28 - @@ -0,0 +1,17 @@ +Revert upstream patch to tree_magic_mini. We cannot pull from github during build. +https://github.com/Wilfred/difftastic/commit/b662a2220b9e9eae5f69ecbc31195bcfcc5216e1 +Index: Cargo.lock +--- Cargo.lock.orig Cargo.lock +@@ -637,8 +637,9 @@ dependencies = [ + + [[package]] + name = "tree_magic_mini" +-version = "3.0.2" +-source = "git+https://github.com/Wilfred/tree_magic?branch=fix-panic-for-empty-strings#13dd6dda15c7062bd8f7dd5bc9bb5b16ce9ee613; ++version = "3.0.3" ++source = "registry+https://github.com/rust-lang/crates.io-index; ++checksum = "91adfd0607cacf6e4babdb870e9bec4037c1c4b151cfd279ccefc5e0c7feaa6d" + dependencies = [ + "bytecount", + "fnv", Index: patches/patch-Cargo_toml === RCS file: patches/patch-Cargo_toml diff -N patches/patch-Cargo_toml --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-Cargo_toml29 Dec 2022 12:37:28 - @@ -0,0 +1,21 @@ +Revert upstream patch to tree_magic_mini. We cannot pull from github during build. +https://github.com/Wilfred/difftastic/commit/b662a2220b9e9eae5f69ecbc31195bcfcc5216e1 +Index: Cargo.toml +--- Cargo.toml.orig Cargo.toml +@@ -51,7 +51,7 @@ const_format = "0.2.22" + owo-colors = "3.3.0" + wu-diff = "0.1.2" + rayon = "1.5.2" +-tree_magic_mini = "3.0.2" ++tree_magic_mini = "3.0.3" + bumpalo = "3.9.1" + unicode-width = "0.1.9" + term_size = "0.3.2" +@@ -73,6 +73,3 @@ path = "src/main.rs" + + [features] + +-[patch.crates-io] +-# https://github.com/mbrubeck/tree_magic/pull/13 hasn't been merged yet. +-tree_magic_mini = { git = 'https://github.com/Wilfred/tree_magic', branch = 'fix-panic-for-empty-strings' }
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: o...@cvs.openbsd.org2022/12/29 05:33:49 ports/devel/difftastic/patches Update of /cvs/ports/devel/difftastic/patches In directory cvs.openbsd.org:/tmp/cvs-serv43411/patches Log Message: Directory /cvs/ports/devel/difftastic/patches added to the repository
Re: [update][proposal] sysutils/nnn add nerd font flavor
For the fonts I think I'd probably pick a couple that already exist in ports/fonts. Simplest approach for ports layout is probably like ports/fonts/nerd-fonts/{terminus,profont,noto) or similar, using the release zips rather than downloading the full repo, factoring out as much as possible to Makefile.inc. -- Sent from a phone, apologies for poor formatting. On 28 December 2022 16:27:38 Joel Carnat wrote: Le 28/12/2022 à 14:57, Martin Ziemer a écrit : Am Wed, Dec 28, 2022 at 01:23:09PM + schrieb Stuart Henderson: On 2022/12/28 13:07, Joel Carnat wrote: When using one of the Nerd Fonts in the terminal, you can get glyphs (icons) rendered. The attached patch enables a flavor for sysutils/nnn that will render mime types icons next to file listing. An example can be seen there: https://i.imgur.com/kOld6HT.gif Tested on OpenBSD/7.2-CURRENT/amd64, using xterm(1) and xfce4-terminal(1). Works here too. (Just xterm) Please add a note to DESCR describing the flavour and the requirement for the additional fonts. It would probably be better to provide ports for some of the more popular of these fonts too, rather than having to tell people to go to a website and download them.. This sounds like the best way to do it. V = 4.7 DISTNAME =nnn-v${V} PKGNAME = nnn-${V} +REVISION = 1 not really important but some people will complain ;) this normally starts at 0 Here is a patch for the corrected Makefile and modified DESCR files. I can look at fonts/msttcorefonts and do the same thing for the Nerd Fonts. But there are 53 fonts, without a popularity note. Uncompressed, they take about 4GB. I only use one font and would probably not install such a huge package. I don't know about others. Maybe using FULLPKGNAME like with x11/windowmaker-lang would be the way to go so that one font is one package and people install the one(s) they want. What do you think? Regards, Joel C.
Re: games/love: bring in multiple versions
On 2022/12/29 11:31:42 +0100, Omar Polo wrote: > On 2022/12/29 02:38:17 -0700, "Anthony J. Bentley" wrote: > > Nice to see this. > > > > Omar Polo writes: > > > The idea then is to move the current port for the 0.8.0 (assuming we > > > want to keep it) as games/love/0.8.0 and add some other versions too, > > > for the time being 0.10.4 and 11.4 (the latest.) 11.1 doesn't compile > > > out of the box, and we can always add other versions later. > > > > My understanding is that LOVE keeps API compatibility between minor > > versions, so 11.1 is obviated by 11.2 and eventually 11.4. > > I thought it was the case (that's why i haven't spent much energy in > trying to port 11.1) but wasn't sure, thanks for confirming. > > it's also probably why they switched from 0.10.X to 11.Y at some > point. > > > If that's > > the case, we should rename the packaged files and the port directories > > to 0.8, 0.10, and 11, and only package the latest of each. > > done > > > I would also like the version numbers in PLIST to use a SUFFIX variable > > (as autoconf, etc. do). > > I avoided it initially because it leads to entries such as > > @lib lib/liblove-${VERSION}.so.${LIBlove-${VERSION}_VERSION} > > but maybe it's not bad. I can't read, sorry. Here's a tarball that defines SUFFIX as autoconf does. it still has entries like @lib lib/liblove${SUFFIX}.so.${LIBlove${SUFFIX}_VERSION} i've also only used SUFFIX in plists and Makefile.inc, not in the other makefiles. should I change them xtoo? > I've also lowered the autoconf and automake version to what upstream > used (latest autoconf would warning about deprecated macros) and > adjusted Makefile.inc not to add the lang/lua category (already done > by lua.port.mk) love.tar.gz Description: GNU Zip compressed data
Re: [update][proposal] sysutils/nnn add nerd font flavor
Am Wed, Dec 28, 2022 at 05:27:23PM +0100 schrieb Joel Carnat: > Le 28/12/2022 à 14:57, Martin Ziemer a écrit : > > Am Wed, Dec 28, 2022 at 01:23:09PM + schrieb Stuart Henderson: > > > On 2022/12/28 13:07, Joel Carnat wrote: > > > > When using one of the Nerd Fonts in the terminal, you can get glyphs > > > > (icons) > > > > rendered. The attached patch enables a flavor for sysutils/nnn that will > > > > render mime types icons next to file listing. An example can be seen > > > > there: > > > > https://i.imgur.com/kOld6HT.gif > > > > > > > > Tested on OpenBSD/7.2-CURRENT/amd64, using xterm(1) and > > > > xfce4-terminal(1). > > Works here too. (Just xterm) > > > > > Please add a note to DESCR describing the flavour and the requirement > > > for the additional fonts. > > > > > > It would probably be better to provide ports for some of the more popular > > > of these fonts too, rather than having to tell people to go to a website > > > and download them.. > > This sounds like the best way to do it. > > > > > > V = 4.7 > > > > DISTNAME =nnn-v${V} > > > > PKGNAME = nnn-${V} > > > > +REVISION = 1 > > > not really important but some people will complain ;) this normally > > > starts at 0 > > > Here is a patch for the corrected Makefile and modified DESCR files. Tested successfully on my system. > I can look at fonts/msttcorefonts and do the same thing for the Nerd Fonts. > But there are 53 fonts, without a popularity note. Uncompressed, they take > about 4GB. I only use one font and would probably not install such a huge > package. I don't know about others. Maybe using FULLPKGNAME like with > x11/windowmaker-lang would be the way to go so that one font is one package > and people install the one(s) they want. > > What do you think? Sounds like a good way to go. (For *me* it would even suffice to only get one font which works fine.) > Regards, > Joel C. > Index: Makefile > === > RCS file: /cvs/ports/sysutils/nnn/Makefile,v > retrieving revision 1.22 > diff -u -p -u -p -r1.22 Makefile > --- Makefile 24 Nov 2022 10:25:40 - 1.22 > +++ Makefile 28 Dec 2022 16:03:25 - > @@ -3,6 +3,7 @@ COMMENT = the missing terminal file bro > V = 4.7 > DISTNAME = nnn-v${V} > PKGNAME =nnn-${V} > +REVISION = 0 > > CATEGORIES = sysutils > > @@ -26,6 +27,13 @@ RUN_DEPENDS = textproc/gsed > MAKE_FLAGS = CFLAGS_OPTIMIZATION= > FAKE_FLAGS = MANPREFIX="$(PREFIX)/man" \ > PREFIX="${PREFIX}" > + > +FLAVORS =nerd > +FLAVOR ?= > + > +.if ${FLAVOR:Mnerd} > +MAKE_FLAGS +=O_NERD=1 > +.endif > > USE_GMAKE = Yes > NO_TEST =Yes > Index: pkg/DESCR > === > RCS file: /cvs/ports/sysutils/nnn/pkg/DESCR,v > retrieving revision 1.1.1.1 > diff -u -p -u -p -r1.1.1.1 DESCR > --- pkg/DESCR 7 May 2018 18:29:00 - 1.1.1.1 > +++ pkg/DESCR 28 Dec 2022 16:03:25 - > @@ -18,3 +18,10 @@ Cool things you can do with nnn: > - pin a directory you may need to revisit and jump to it anytime > - lock the current terminal after a specified idle time > - change directory on exit > + > +Flavors: > + nerd- build with support for Nerd Fonts to display glyphs. > + Nerd Fonts can be browsed and downloaded from the author's > + website https://www.nerdfonts.com/font-downloads. > + The font(s) must be installed in $HOME/.local/share/fonts/ > + or in /usr/local/share/fonts/.
Re: games/love: bring in multiple versions
On 2022/12/29 02:38:17 -0700, "Anthony J. Bentley" wrote: > Nice to see this. > > Omar Polo writes: > > The idea then is to move the current port for the 0.8.0 (assuming we > > want to keep it) as games/love/0.8.0 and add some other versions too, > > for the time being 0.10.4 and 11.4 (the latest.) 11.1 doesn't compile > > out of the box, and we can always add other versions later. > > My understanding is that LOVE keeps API compatibility between minor > versions, so 11.1 is obviated by 11.2 and eventually 11.4. I thought it was the case (that's why i haven't spent much energy in trying to port 11.1) but wasn't sure, thanks for confirming. it's also probably why they switched from 0.10.X to 11.Y at some point. > If that's > the case, we should rename the packaged files and the port directories > to 0.8, 0.10, and 11, and only package the latest of each. done > I would also like the version numbers in PLIST to use a SUFFIX variable > (as autoconf, etc. do). I avoided it initially because it leads to entries such as @lib lib/liblove-${VERSION}.so.${LIBlove-${VERSION}_VERSION} but maybe it's not bad. I've also lowered the autoconf and automake version to what upstream used (latest autoconf would warning about deprecated macros) and adjusted Makefile.inc not to add the lang/lua category (already done by lua.port.mk) love.tar.gz Description: GNU Zip compressed data
Re: [new] age-plugin-yubikey
Hi Omar! On 12/29/22 10:15, Omar Polo wrote: it fails to build if security/pcsc-lite is not installed: Fixed. I applied your other suggestions as well except for +DISTFILES+=${GH_DISTNAME} s/DISTNAME/DISTFILE/ ;-) As always, thanks for your feedback! cu, Volker age-plugin-yubikey.tar.gz Description: application/gzip
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2022/12/29 02:53:10 Modified files: net/tdesktop : Makefile distinfo Removed files: net/tdesktop/patches: patch-Telegram_CMakeLists_txt patch-Telegram_SourceFiles_calls_calls_call_cpp patch-Telegram_cmake_lib_tgcalls_cmake Log message: Build libtgvoip again, 4.4.2 will need it This pulls in PulseAudio, but Qt already depends on it anyway. Most if not all calls should continue using the tgcalls.
Re: games/love: bring in multiple versions
Nice to see this. Omar Polo writes: > The idea then is to move the current port for the 0.8.0 (assuming we > want to keep it) as games/love/0.8.0 and add some other versions too, > for the time being 0.10.4 and 11.4 (the latest.) 11.1 doesn't compile > out of the box, and we can always add other versions later. My understanding is that LOVE keeps API compatibility between minor versions, so 11.1 is obviated by 11.2 and eventually 11.4. If that's the case, we should rename the packaged files and the port directories to 0.8, 0.10, and 11, and only package the latest of each. I would also like the version numbers in PLIST to use a SUFFIX variable (as autoconf, etc. do).
Re: [UPDATE] lang/node 18.12.1
> > Not sure why lib/eopenssl30/{crypto,ssl} aren't found by > > port-lib-depends-check... > > Not sure either, but that was the case for node 16 with eopenssl11 already > ... port-lib-depends-check doesn't handle libraries in lib/subdir/
Re: [UPDATE] lang/node 18.12.1
On 12/29/22 02:28, Thomas Frohwein wrote: - Build log shows a bunch of '-O3', not everywhere, but looks like CFLAGS/CXXFLAGS aren't honored consistently. True, thanks for catching that! - I'm getting some odd messages from `make port-lib-depends-check`: node-18.12.1v0(lang/node): Bogus WANTLIB: crypto.12 (/usr/local/bin/node) (NOT REACHABLE) Missing: intl.7 from gettext-runtime-0.21.1 (/usr/local/lib/node_modules/node-pledge/build/node_gyp_bins/python3) Missing: python3.10.0 from python-3.10.9p0 (/usr/local/lib/node_modules/node-pledge/build/node_gyp_bins/python3) Bogus WANTLIB: ssl.12 (/usr/local/bin/node) (NOT REACHABLE) Missing: util.16 (/usr/local/lib/node_modules/node-pledge/build/node_gyp_bins/python3) (system lib) WANTLIB += intl python3.10 util Looks to me like the libraries in the last line should be added. No, I needed to remove the python3 binary that was copied there by node-gyp as a build artifact... sorry, that escaped me. Not sure why lib/eopenssl30/{crypto,ssl} aren't found by port-lib-depends-check... Not sure either, but that was the case for node 16 with eopenssl11 already ... With the attached diff, the python binary is gone and I couldn't find an -O3 in the build log anymore.Index: Makefile === RCS file: /cvs/ports/lang/node/Makefile,v retrieving revision 1.111 diff -u -p -r1.111 Makefile --- Makefile 13 Nov 2022 15:28:44 - 1.111 +++ Makefile 29 Dec 2022 09:17:31 - @@ -5,12 +5,11 @@ USE_WXNEEDED = Yes COMMENT = JavaScript runtime built on Chrome's V8 JavaScript engine -NODE_VERSION = v16.17.1 -PLEDGE_VER = 1.1.2 +NODE_VERSION = v18.12.1 +PLEDGE_VER = 1.1.3 DISTFILES = node-pledge-{}${PLEDGE_VER}.tar.gz:0 \ ${DISTNAME}-headers.tar.gz \ ${DISTNAME}.tar.xz -REVISION = 2 DISTNAME = node-${NODE_VERSION} PKGNAME = ${DISTNAME:S/v//g} @@ -34,7 +33,7 @@ MODULES = lang/python WANTLIB += c execinfo m pthread ${COMPILER_LIBCXX} WANTLIB += z brotlienc brotlidec WANTLIB += icudata icui18n icuuc cares nghttp2 uv -WANTLIB += lib/eopenssl11/ssl lib/eopenssl11/crypto +WANTLIB += lib/eopenssl30/ssl lib/eopenssl30/crypto COMPILER = base-clang ports-gcc @@ -71,16 +70,16 @@ SUBST_VARS += WRKDIST SUBST_VARS += NODE_VERSION SUBST_VARS += EOPENSSL_LIB -# OpenSSL used: {X,Ed}25519 via EVP, SSL_CIPHER_standard_name and 5-10 other missing symbols/defines. +# uses a wide range of OpenSSL API and only really supports boring/openssl LIB_DEPENDS += archivers/brotli \ devel/libuv \ net/libcares \ textproc/icu4c \ www/nghttp2 \ - security/openssl/1.1 + security/openssl/3.0 -EOPENSSL_LIB = ${LOCALBASE}/lib/eopenssl11 -EOPENSSL_INC = ${LOCALBASE}/include/eopenssl11 +EOPENSSL_LIB = ${LOCALBASE}/lib/eopenssl30 +EOPENSSL_INC = ${LOCALBASE}/include/eopenssl30 post-extract: mv ${WRKDIR}/node-pledge-${PLEDGE_VER} \ @@ -100,8 +99,9 @@ post-install: ${INSTALL_DATA_DIR} ${PREFIX}/share/node ${INSTALL_DATA} ${FULLDISTDIR}/${DISTNAME}-headers.tar.gz \ ${PREFIX}/share/node/${DISTNAME}-headers.tar.gz - ln -s ${TRUEPREFIX}/lib/node_modules/npm/bin/node-gyp-bin/node-gyp \ + ln -fs ${TRUEPREFIX}/lib/node_modules/npm/bin/node-gyp-bin/node-gyp \ ${PREFIX}/bin/node-gyp; + rm -r ${PREFIX}/lib/node_modules/npm/node_modules/node-pledge/build/node_gyp_bins cd ${PREFIX}/lib/node_modules/npm/node_modules && for x in *; do \ mv $$x ../..; \ ln -s ../../$$x $$x; \ Index: distinfo === RCS file: /cvs/ports/lang/node/distinfo,v retrieving revision 1.65 diff -u -p -r1.65 distinfo --- distinfo 1 Nov 2022 12:01:49 - 1.65 +++ distinfo 29 Dec 2022 09:17:31 - @@ -1,6 +1,6 @@ -SHA256 (node-pledge-1.1.2.tar.gz) = zY/JcbZ32mmtqWXXNn3/9aTh7Y3F6fAAaADDA8SYwEk= -SHA256 (node-v16.17.1-headers.tar.gz) = Ncy5XK8CzaO9aA2kNQqK5dZmp6nq46/lwqGz7ymu8Qg= -SHA256 (node-v16.17.1.tar.xz) = ZyH+tBUtVtLGs1jOOXq9Wn8drwnuLiXFAhubTT+GozA= -SIZE (node-pledge-1.1.2.tar.gz) = 3155 -SIZE (node-v16.17.1-headers.tar.gz) = 568068 -SIZE (node-v16.17.1.tar.xz) = 35661452 +SHA256 (node-pledge-1.1.3.tar.gz) = fEaXvLg6hYEJ69K+mgQFizf8DiJY2/DtyFJB/pEanVU= +SHA256 (node-v18.12.1-headers.tar.gz) = nVXuByum1aFB2wks7xoPcV99P8k4KFptknodCgx0Qvc= +SHA256 (node-v18.12.1.tar.xz) = T6QGRRvFJlmikOUs/bIWKnYL1UnaS4u+vmop8pbZON8= +SIZE (node-pledge-1.1.3.tar.gz) = 3167 +SIZE (node-v18.12.1-headers.tar.gz) = 8563785 +SIZE (node-v18.12.1.tar.xz) = 38454588 Index: patches/patch-Makefile === RCS file: /cvs/ports/lang/node/patches/patch-Makefile,v retrieving revision 1.15 diff -u -p -r1.15 patch-Makefile --- patches/patch-Makefile 18 Mar 2022 19:35:16 - 1.15 +++ patches/patch-Makefile 29 Dec 2022 09:17:31 - @@ -1,7 +1,7 @@ Index: Makefile --- Makefile.orig +++ Makefile -@@ -163,7 +163,7 @@ config.gypi: configure configure.py src/node_version.h +@@ -185,7 +185,7 @@
Re: [new] age-plugin-yubikey
On 2022/12/28 20:55:21 +0100, Volker Schlecht wrote: > With the update to age 1.1.x, and free time galore, I couldn't resist > playing with age-plugin-yubikey, which to my surprise worked smoothly > with my Yubikey 5, once I re-enabled API access to the serial number. > > I'm not sure if that was something I removed myself or whether my Yubi5 > came with that setting, so I'm not sure if that's worth mentioning in > the port. > > If ```ykman list --serials``` shows your key's serial number, everything > should just work, if not you may need to use ykpersonalize to set option > serial-api-visible first. can't really tested, looked only for curiosity. it fails to build if security/pcsc-lite is not installed: --- stderr thread 'main' panicked at 'Could not find a PCSC library. For the target OS `openbsd`, I tried to use pkg-config to find libpcsclite. Do you have pkg-config and libpcsclite configured for this target?: `"pkg-config" "--libs" "--cflags" "libpcsclite" "libpcsclite >= 1"` did not exit successfully: exit status: 1 error: could not find system library 'libpcsclite' required by the 'pcsc-sys' crate --- stderr Package libpcsclite was not found in the pkg-config search path ', /home/_pobj/age-plugin-yubikey-0.3.0/age-plugin-yubikey-0.3.0/modcargo-crates/pcsc-sys-1.2.0/build.rs:30:22 you could use ${GH_DISTFILE} instead of ${DISTNAME}${EXTRACT_SUFX}, and HOMEPAGE is already defined by GH_*. (take the following with a grain of salt because i don't hack on rust ports usually.) I'd move the ".include crates.inc" at the end, just above bsd.port.mk and rearrange WANTLIB. I guess it should use ${MODCARGO_WANTLIB}, c++ was marked as unused here, and add pcsclite (could also reduce the indentation of the values, two tabs are enough usually) all together: (plus changing the order as per Makefile.template) --- Makefile.orig Thu Dec 29 10:02:10 2022 +++ MakefileThu Dec 29 10:14:24 2022 @@ -4,27 +4,26 @@ GH_PROJECT=age-plugin-yubikey GH_TAGNAME=v0.3.0 -HOMEPAGE= https://github.com/str4d/age-plugin-yubikey +CATEGORIES=security MAINTAINER=Volker Schlecht -CATEGORIES=security - # MIT or Apache 2.0 PERMIT_PACKAGE=Yes -# as devel/cargo MODULES adds DISTFILES, GH_* didn't -DISTFILES+=${DISTNAME}${EXTRACT_SUFX} -MODULES= devel/cargo +WANTLIB += ${MODCARGO_WANTLIB} m pcsclite -WANTLIB+= c c++ c++abi m pthread +DISTFILES+=${GH_DISTNAME} -.include "crates.inc" +MODULES= devel/cargo +LIB_DEPENDS+= security/pcsc-lite + CONFIGURE_STYLE= cargo post-install: ${INSTALL_DATA_DIR} ${PREFIX}/share/doc/age-plugin-yubikey ${INSTALL_DATA} ${WRKSRC}/README.md ${PREFIX}/share/doc/age-plugin-yubikey +.include "crates.inc" .include
Re: new: misc/lchat
On 2022/12/28 21:56:45 +0100, Jan Klemkow wrote: > Hi, > > This port contains the suckless tool lchat: > > lchat (line chat) is a line oriented front end for ii-like chat > programs. It handles the input from keyboard and output file in > parallel. Thus, you are able to type messages while new chat lines > are arriving. Its main focus is on usability and simplicity. > > ok? quickly tested, seems to work just fine. maybe it'll convince me to give ii another try :) could remove the comment and needs to respect the CFLAGS, otherwise OK for me. --- Makefile.orig Thu Dec 29 09:33:44 2022 +++ MakefileThu Dec 29 09:35:14 2022 @@ -8,12 +8,13 @@ # ISC PERMIT_PACKAGE=Yes -# If pledge is used, annotate with `uses pledge()' in a comment - WANTLIB += c grapheme MASTER_SITES = https://dl.suckless.org/tools/ LIB_DEPENDS = textproc/libgrapheme + +CFLAGS += -I${LOCALBASE}/include +MAKE_FLAGS += CFLAGS="${CFLAGS}" .include
Re: NEW: net/endlessh
Ping. On Wed, Dec 21, 2022 at 02:56:44PM +0100, Michael wrote: > On Tue, Dec 20, 2022 at 07:04:19PM +0100, Pascal Stumpf wrote: > > On Sun, 18 Dec 2022 15:11:22 +0100, Michael wrote: > > > On Sun, Dec 18, 2022 at 11:55:12AM +, Stuart Henderson wrote: > > > > On 2022/12/17 22:36, Michael wrote: > > > > > On Sat, Dec 17, 2022 at 11:06:32AM +0100, Omar Polo wrote: > > > > > > On 2022/12/17 10:41:26 +0100, Pascal Stumpf > > > > > > wrote: > > > > > > > On Sat, 17 Dec 2022 10:18:15 +0100, Pascal Stumpf wrote: > > > > > > > > A few tweaks: > > > > > > > > > > > > > > > > * set MAKE_FLAGS so that the build respects CFLAGS > > > > > > > > * NO_TESTS = Yes > > > > > > > > * add README and endless.rc to the port itself, reformat, turn > > > > > > > > on -s per > > > > > > > > default (syslog logging) > > > > > > > > * add dedicated _endlessh user > > > > > > > > * install a default config file into examples and @sample it > > > > > > > > > > > > > > * correct Nm in endlessh.1 > > > > > > > > > > > > there's an extra patch-Makefile.orig in the tarball and > > > > > > pkg/endlessh.rc is executable when it doesn't need to. > > > > > > > > > > > > ok for me with that fixed. > > > > > > > > > > Thanks to Pascal and you for the tweaks. > > > > > > > > > > I have tested the latest revision; works as expected and the changes > > > > > seem fine to me. I took the liberty of fixing the last problems Omar > > > > > mentioned in the attached port. > > > > > > > > > > > > > > > > > > > > > > > in pkg/README I'd say to symlink /etc/rc.d/endlessh to endlessh6 > > > > > > instead of copying it, less to worry when updating. > > > > > > > > > > > > If I'm reading it correctly, it can't directly bind to 22 because it > > > > > > doesn't start as root, it' would be nice to include an excerpt of > > > > > > the > > > > > > pf configuration to redirect the port 22 to . > > > > > > > > > > > > I'd use a patch instead of perl -pi in post-install to tweak the > > > > > > configuration, it's more verbose but it's also more resiliant to > > > > > > upstream changes to the file. > > > > > > > > > > > > Including the diff for user.list in case it comes in handy to who > > > > > > would like to test it. > > > > > > > > > > > > Index: user.list > > > > > > === > > > > > > RCS file: /home/cvs/ports/infrastructure/db/user.list,v > > > > > > retrieving revision 1.413 > > > > > > diff -u -p -r1.413 user.list > > > > > > --- user.list 14 Dec 2022 12:09:05 - 1.413 > > > > > > +++ user.list 17 Dec 2022 09:50:27 - > > > > > > @@ -395,3 +395,4 @@ id usergroup port > > > > > > 884 _iblock_iblock net/iblock > > > > > > 885 _mycorrhiza_mycorrhiza www/mycorrhiza > > > > > > 886 _eduvpn_eduvpn net/eduvpn > > > > > > +887 _endlessh _endlessh net/endlessh > > > > > > > > > > > > > > There is no 1.1 release yet, so better name it 1.1pre20210430 (date of > > > > commit). > > > > > > > > From readme: > > > > > > > > : If you want to cover both IPv4 and IPv6 you'll need to run *two* > > > > instances of > > > > : endlessh. > > > > : > > > > : - copy the endlessh rc script to ${RCDIR}/endlessh6 > > > > : - copy the config file to ${SYSCONFDIR}/endlessh/config6 > > > > : - use BindFamily 6 in config6 > > > > : - in rc.conf.local force endlessh6 to load config6 like so: > > > > : > > > > : endlessh6_flags=-s -f /etc/endlessh/config6 > > > > : endlessh_flags=-s > > > > > > > > No need to tell the user to do this, just provide an endlessh6 rc script > > > > with the required daemon_flags. I don't think you need a second config > > > > file, just use -6. > > > > > > > > : Covering more than 128 connections > > > > : == > > > > : > > > > : The defaults in OpenBSD only allow for 128 open file descriptors per > > > > process, > > > > : so regardless of the MaxClients setting in > > > > ${SYSCONFDIR}/endlessh/config > > > > : you'll end up with something like 124 clients at the most. > > > > : You can increase these limits in ${SYSCONFDIR}/login.conf for > > > > endlessh (and > > > > : endlessh6) like so: > > > > : > > > > : endlessh:\ > > > > : :openfiles=1024:\ > > > > : :tc=daemon: > > > > : endlessh6:\ > > > > : :openfiles=1024:\ > > > > : :tc=daemon: > > > > > > > > Provide pkg/endlessh.login and endless6.login files instead. 1024x2 is > > > > a bit high for the default kern.maxfiles, I would suggest not more than > > > > 512 for the installed file, users can change it if they need more. > > > > > > > > Then because it is just using standard OS mechanisms there is no more > > > > need for pkg/README. > > > > > > > > > > Thanks for the feedback. > > > > > > Attached port should contain all the suggestions above. > > > Also the pledge() comment was changed to "uses