You can subscribe to this list here.
2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(182) |
Nov
(548) |
Dec
(347) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2002 |
Jan
(695) |
Feb
(397) |
Mar
(438) |
Apr
(544) |
May
(367) |
Jun
(273) |
Jul
(160) |
Aug
(247) |
Sep
(204) |
Oct
(308) |
Nov
(467) |
Dec
(406) |
2003 |
Jan
(795) |
Feb
(532) |
Mar
(355) |
Apr
(375) |
May
(352) |
Jun
(312) |
Jul
(290) |
Aug
(157) |
Sep
(130) |
Oct
(261) |
Nov
(532) |
Dec
(480) |
2004 |
Jan
(445) |
Feb
(433) |
Mar
(323) |
Apr
(259) |
May
(126) |
Jun
(175) |
Jul
(219) |
Aug
(286) |
Sep
(178) |
Oct
(144) |
Nov
(141) |
Dec
(133) |
2005 |
Jan
(148) |
Feb
(263) |
Mar
(224) |
Apr
(186) |
May
(283) |
Jun
(243) |
Jul
(138) |
Aug
(124) |
Sep
(109) |
Oct
(90) |
Nov
(295) |
Dec
(164) |
2006 |
Jan
(201) |
Feb
(221) |
Mar
(221) |
Apr
(157) |
May
(87) |
Jun
(131) |
Jul
(97) |
Aug
(93) |
Sep
(174) |
Oct
(112) |
Nov
(208) |
Dec
(166) |
2007 |
Jan
(188) |
Feb
(140) |
Mar
(121) |
Apr
(105) |
May
(95) |
Jun
(131) |
Jul
(131) |
Aug
(72) |
Sep
(152) |
Oct
(116) |
Nov
(222) |
Dec
(237) |
2008 |
Jan
(160) |
Feb
(106) |
Mar
(266) |
Apr
(263) |
May
(150) |
Jun
(140) |
Jul
(224) |
Aug
(119) |
Sep
(105) |
Oct
(103) |
Nov
(71) |
Dec
(52) |
2009 |
Jan
(67) |
Feb
(39) |
Mar
(73) |
Apr
(171) |
May
(113) |
Jun
(90) |
Jul
(160) |
Aug
(129) |
Sep
(270) |
Oct
(117) |
Nov
(88) |
Dec
(97) |
2010 |
Jan
(105) |
Feb
(59) |
Mar
(219) |
Apr
(157) |
May
(128) |
Jun
(168) |
Jul
(87) |
Aug
(100) |
Sep
(95) |
Oct
(50) |
Nov
(61) |
Dec
(31) |
2011 |
Jan
(78) |
Feb
(158) |
Mar
(85) |
Apr
(95) |
May
(50) |
Jun
(37) |
Jul
(66) |
Aug
(24) |
Sep
(50) |
Oct
(79) |
Nov
(79) |
Dec
(51) |
2012 |
Jan
(75) |
Feb
(53) |
Mar
(59) |
Apr
(71) |
May
(78) |
Jun
(90) |
Jul
(83) |
Aug
(74) |
Sep
(82) |
Oct
(35) |
Nov
(10) |
Dec
(22) |
2013 |
Jan
(34) |
Feb
(35) |
Mar
(17) |
Apr
(6) |
May
(27) |
Jun
(53) |
Jul
(27) |
Aug
(19) |
Sep
(23) |
Oct
(58) |
Nov
(68) |
Dec
(24) |
2014 |
Jan
(29) |
Feb
(5) |
Mar
(22) |
Apr
(47) |
May
(73) |
Jun
(75) |
Jul
(13) |
Aug
(76) |
Sep
(40) |
Oct
(78) |
Nov
(53) |
Dec
(28) |
2015 |
Jan
(76) |
Feb
(113) |
Mar
(41) |
Apr
(94) |
May
(48) |
Jun
(41) |
Jul
(15) |
Aug
(8) |
Sep
(62) |
Oct
(67) |
Nov
(28) |
Dec
(5) |
2016 |
Jan
(20) |
Feb
(18) |
Mar
(18) |
Apr
(2) |
May
(10) |
Jun
(46) |
Jul
(11) |
Aug
(34) |
Sep
(10) |
Oct
(33) |
Nov
(30) |
Dec
(1) |
2017 |
Jan
(24) |
Feb
(1) |
Mar
(10) |
Apr
(9) |
May
(8) |
Jun
(5) |
Jul
(9) |
Aug
(17) |
Sep
(13) |
Oct
(23) |
Nov
(16) |
Dec
(14) |
2018 |
Jan
(8) |
Feb
(1) |
Mar
(7) |
Apr
(7) |
May
|
Jun
(3) |
Jul
(10) |
Aug
|
Sep
(2) |
Oct
|
Nov
|
Dec
|
2019 |
Jan
|
Feb
(2) |
Mar
|
Apr
(5) |
May
(1) |
Jun
(3) |
Jul
(5) |
Aug
(16) |
Sep
(3) |
Oct
(8) |
Nov
(3) |
Dec
(3) |
2020 |
Jan
(14) |
Feb
(4) |
Mar
(13) |
Apr
(25) |
May
(11) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(10) |
Dec
(1) |
2021 |
Jan
(5) |
Feb
|
Mar
(2) |
Apr
(11) |
May
(14) |
Jun
(17) |
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2022 |
Jan
|
Feb
|
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(11) |
2023 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
(3) |
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2024 |
Jan
|
Feb
|
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Scott H. <sha...@us...> - 2024-03-11 21:18:01
|
Yaniv, Yes, this shows the error. the "EMACS_GET_PROCESS_GROUP ()” expands to “getpgrp(0)” which is an error since getpgrp expects no arguments. It seems that the source code needs a patch to run on newer compilers that seem to be more of a stickler for arguments. Unfortunately there is no maintainer for this package. maintainer: None <fin...@li...> For all I know it may be a simple update to the patch file or a bump to the upstream version. One thing to try is in the build directory edit the line in systty.h from > #define EMACS_GET_PROCESS_GROUP() EMACS_GETPGRP (0) to > #define EMACS_GET_PROCESS_GROUP() EMACS_GETPGRP () And then see if it builds. Just a guess. -Scott > On Mar 11, 2024, at 15:32, Yaniv Almog <yal...@gm...> wrote: > > Thanks Scott > > Perhaps you need that (??) > > process-unix.c:1033:23: error: too many arguments to function call, expected 0, have 1 > > pid_t piddly = EMACS_GET_PROCESS_GROUP (); > > ^~~~~~~~~~~~~~~~~~~~~~~~~~ > > ./systty.h:286:50: note: expanded from macro 'EMACS_GET_PROCESS_GROUP' > > #define EMACS_GET_PROCESS_GROUP() EMACS_GETPGRP (0) > > ~~~~~~~~~~~~~~~^~ > > ./systty.h:282:34: note: expanded from macro 'EMACS_GETPGRP' > > #define EMACS_GETPGRP(x) getpgrp(x) > > ~~~~~~~ ^ > > /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/unistd.h:459:8: note: 'getpgrp' declared here > > pid_t getpgrp(void); > > ^ > > process-unix.c:1951:12: warning: cast to 'void *' from smaller integer type 'int' [-Wint-to-void-pointer-cast] > > *vinfd = (void*)inch; > > ^~~~~~~~~~~ > > process-unix.c:1952:13: warning: cast to 'void *' from smaller integer type 'int' [-Wint-to-void-pointer-cast] > > *voutfd = (void*)outch; > > ^~~~~~~~~~~~ > > process-unix.c:2113:12: warning: cast to 'void *' from smaller integer type 'int' [-Wint-to-void-pointer-cast] > > *vinfd = (void*)rs; > > ^~~~~~~~~ > > process-unix.c:2114:13: warning: cast to 'void *' from smaller integer type 'int' [-Wint-to-void-pointer-cast] > > *voutfd = (void*)ws; > > ^~~~~~~~~ > > > > The end of the diagnostic step follows > > > > XEmacs 21.4.24 "Standard C" configured for `i386-apple-darwin22.6.0'. > > > > > > Compilation / Installation: > > Source code location: /opt/sw.build/xemacs-21.4.24-11/xemacs-21.4.24 > > Installation prefix: /opt/sw > > Additional prefixes: /opt/sw > > Operating system description file: `s/darwin.h' > > Not using any machine description file > > Compiler: gcc -O2 -g -std=c89 > > Compiler version: Apple clang version 15.0.0 (clang-1500.1.0.2.5) > > Compiler specs file: (clang-1500.1.0.2.5) > > Relocating allocator for buffers: no > > GNU version of malloc: no > > - The GNU allocators don't work with this system configuration. > > libc: system-provided libc on i386-apple-darwin22.6.0 > > > > Window System: > > Compiling in support for the X window system: > > - X Windows headers location: /usr/X11/include > > - X Windows libraries location: /usr/X11/lib > > - Handling WM_COMMAND properly. > > Compiling in support for the Athena widget set: > > - Athena headers location: X11/Xaw3d > > - Athena library to link: Xaw3d > > Using Lucid menubars. > > Using Lucid scrollbars. > > Using Athena dialog boxes. > > Using Athena native widgets. > > > > TTY: > > Compiling in support for ncurses. > > > > Images: > > Compiling in support for GIF images (builtin). > > WARNING: ----------------------------------------------------------- > > WARNING: Compiling without XPM image support. > > WARNING: You should strongly consider installing XPM. > > WARNING: Otherwise toolbars and other graphics will look suboptimal. > > WARNING: (a copy may be found in ftp://ftp.xemacs.org/pub/xemacs/aux <ftp://ftp.xemacs.org/pub/xemacs/aux>) > > WARNING: ----------------------------------------------------------- > > Compiling in support for PNG images. > > Compiling in support for JPEG images. > > Compiling in support for TIFF images. > > Compiling in support for X-Face message headers. > > > > Sound: > > > > Databases: > > Compiling in support for GNU DBM. > > > > Internationalization: > > > > Mail: > > Compiling in support for "dot-locking" mail spool file locking method. > > > > Other Features: > > Inhibiting IPv6 canonicalization at startup. > > Compiling in support for dynamic shared object modules. > > Using the new portable dumper. > > > > I hope that this what you need. > > > > Thanks again > > Yaniv > > On Mon, Mar 11, 2024 at 8:43 PM Scott Hannahs <sha...@us... <mailto:sha...@us...>> wrote: >> Yaniv, >> >> The error occurred before your snippet. The snippet is showing the report of an error and the cleanup commands. The snippet starts with 8 warnings AND 1 ERROR. That error was above that notice. Please go back through the log and find the command (complete line) and the error message. Possibly a gcc call? >> >> -Scott >> >>> On Mar 10, 2024, at 05:17, Yaniv Almog <yal...@gm... <mailto:yal...@gm...>> wrote: >>> >>> Dear users >>> >>> I can't install xemacs on my Ventura macos. I attach the announcements I get. Any ideas for a workaround? >>> >>> Thanks, >>> Yaniv Almog >>  |
From: Scott H. <sha...@us...> - 2023-06-30 11:47:20
|
When running “fink update qt5-mac-qtbase”, I get a compiler error. Seems that “ATSFontNotify” and related functions do not exist anymore. They were deprecated a long time ago. I see that the current qt5-mac release is 5.15.2 and maybe that resolves issues with later OS versions? updated with “fink self-update” set #jobs to 1 MacOS 13.4.1 fink version 14.3.1 X86 architecture I get the following error returned: /Applications/Xcode.app/Contents/Developer/usr/bin/g++ -c -MD -isysroot /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX13.3.sdk -mmacosx-version-min=13.4 -fconstant-cfstrings -O2 -fPIC -std=c++1z -fvisibility=hidden -fvisibility-inlines-hidden -fno-exceptions -Wall -W -Wvla -Wdate-time -DQT_NO_MTDEV -DQT_NO_LIBUDEV -DQT_NO_EVDEV -DQT_NO_TSLIB -DQT_NO_LIBINPUT -DQT_NO_CAST_FROM_ASCII -DQT_BUILD_PLATFORMSUPPORT_LIB -DQT_BUILDING_QT -DQT_NO_CAST_TO_ASCII -DQT_ASCII_CAST_WARNINGS -DQT_MOC_COMPAT -DQT_USE_QSTRINGBUILDER -DQT_DEPRECATED_WARNINGS -DQT_DISABLE_DEPRECATED_BEFORE=0x050000 -DQT_NO_EXCEPTIONS -D_LARGEFILE64_SOURCE -D_LARGEFILE_SOURCE -DQT_NO_DEBUG -DQT_GUI_LIB -DQT_CORE_LIB -I. -Icglconvenience -Iaccessibility -I../../include -I../../include/QtPlatformSupport -I../../include/QtPlatformSupport/5.7.1 -I../../include/QtPlatformSupport/5.7.1/QtPlatformSupport -I../../include/QtGui/5.7.1 -I../../include/QtGui/5.7.1/QtGui -I../../include/QtCore/5.7.1 -I../../include/QtCore/5.7.1/QtCore -I../../include/QtGui -I../../include/QtCore -I.moc -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX13.3.sdk/System/Library/Frameworks/OpenGL.framework/Headers -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX13.3.sdk/System/Library/Frameworks/AGL.framework/Headers -I../../mkspecs/macx-g++ -F/opt/sw/src/fink.build/qt5-mac-qtbase-5.7.1-13/qtbase-opensource-src-5.7.1/lib -o .obj/qcoretextfontdatabase.o fontdatabases/mac/qcoretextfontdatabase.mm In file included from fontdatabases/mac/qcoretextfontdatabase.mm:51: In file included from fontdatabases/mac/qcoretextfontdatabase_p.h:58: In file included from ../../include/QtGui/5.7.1/QtGui/qpa/qplatformfontdatabase.h:1: In file included from ../../include/QtGui/5.7.1/QtGui/qpa/../../../../../src/gui/text/qplatformfontdatabase.h:60: In file included from ../../include/QtGui/5.7.1/QtGui/private/qfontengine_p.h:1: In file included from ../../include/QtGui/5.7.1/QtGui/private/../../../../../src/gui/text/qfontengine_p.h:59: In file included from ../../include/QtGui/5.7.1/QtGui/private/qtextengine_p.h:1: In file included from ../../include/QtGui/5.7.1/QtGui/private/../../../../../src/gui/text/qtextengine_p.h:58: In file included from ../../include/QtGui/qtextlayout.h:1: In file included from ../../include/QtGui/../../src/gui/text/qtextlayout.h:48: In file included from ../../include/QtGui/qevent.h:1: In file included from ../../include/QtGui/../../src/gui/kernel/qevent.h:49: In file included from ../../include/QtCore/qvariant.h:1: ….. fontdatabases/mac/qcoretextfontdatabase.mm:971:13: error: ‘ATSFontDeactivate' is unavailable ATSFontDeactivate(font.value<ATSFontContainerRef>(), 0, kATSOptionFlagsDoNotNotify); ^ /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX13.3.sdk/System/Library/Frameworks/ApplicationServices.framework/Frameworks/ATS.framework/Headers/ATSFont.h:385:1: note: 'ATSFontDeactivate' has been explicitly marked unavailable here ATSFontDeactivate( ^ fontdatabases/mac/qcoretextfontdatabase.mm:979:5: error: 'ATSFontNotify' is unavailable ATSFontNotify(kATSFontNotifyActionFontsChanged, 0); ^ /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX13.3.sdk/System/Library/Frameworks/ApplicationServices.framework/Frameworks/ATS.framework/Headers/ATSFont.h:1144:1: note: 'ATSFontNotify' has been explicitly marked unavailable here ATSFontNotify( ^ 26 warnings and 6 errors generated. make[2]: *** [.obj/qcoretextfontdatabase.o] Error 1 make[2]: Leaving directory `/opt/sw/src/fink.build/qt5-mac-qtbase-5.7.1-13/qtbase-opensource-src-5.7.1/src/platformsupport' make[1]: *** [sub-platformsupport-make_first] Error 2 make[1]: Leaving directory `/opt/sw/src/fink.build/qt5-mac-qtbase-5.7.1-13/qtbase-opensource-src-5.7.1/src' make: *** [sub-src-make_first] Error 2 make: Leaving directory `/opt/sw/src/fink.build/qt5-mac-qtbase-5.7.1-13/qtbase-opensource-src-5.7.1' ### execution of /tmp/fink.nVpiL failed, exit code 2 ### execution of /tmp/fink.N1dqr failed, exit code 2 Removing runtime build-lock... Removing build-lock package... /opt/sw/bin/dpkg-lockwait -r fink-buildlock-qt5-mac-qtbase-5.7.1-13 (Reading database ... 294161 files and directories currently installed.) Removing fink-buildlock-qt5-mac-qtbase-5.7.1-13 (2023.06.29-23.43.20) ... Failed: phase compiling: qt5-mac-qtbase-5.7.1-13 failed Also include the following system information: Package manager version: 0.45.99.git Distribution version: selfupdate-git Thu Jun 29 20:36:27 2023, 12.0, x86_64 Trees: local/injected local/main stable/main Xcode.app: 14.3.1 Xcode command-line tools: 14.3.1.0.1.1683849156 Max. Fink build jobs: 1 -Scott |
From: Ebrahim M. <eb...@ma...> - 2023-05-13 18:00:42
|
Daniel, Thank you very much for your clear, detailed explanation. Best Regards, E. > On May 13, 2023, at 7:21 PM, Daniel Macks <dm...@ne...> wrote: > > That's pretty typical for a high-level package on a machine that has not had a lot of other packages built on it. Some are obviously the direct needs of fluidsynth itself, some are infrastructure (tools used to build various other packages), and some are dependencies of those two types of packages (lots of lower-level libraries widely used by many things, recursively). I just parsed the dependency tree of fluidsynth1 and can confirm that this list looks correct. > > My first attempt was to look at a visual "dotty" graph of it, but with over 200 nodes and over 1100 edges, it's not too useful except to say "wow that's complicated!" and to see that there didn't appear to be any large disjoint sections. If you have a few certain packages in that list that are interesting or curious, let me know and I will trace them specifically. > > For the record, here's the tool I used. > > 1. Create a file 'goal' that contains a list of packages to query, one per line (your "...will be installed" set and the fluidsynth packages themselves). Order is not important, but make sure there is no other whitespace. > 2. fink -m list --format=dotty-build > pkg.dot > 3. perl -ni.bak -e 'print if /$p/; } BEGIN {open $h, "<", "goal"; chomp(@p=<$h>);close $h;$p=~s/ /|/g;$p=join "|", @p;$p=qr/"($p)" /' pkg.dot > 4. Copy the first three lines and last one line from 'pkg.dot.bak' to the start and end, respectively, of 'pkg.dot'. > > 'pkg.dot' is now a directed graph of all of the dependencies among all of the packages listed in 'goal'. > > dan > > On 5/13/23, 6:47 AM, "Ebrahim Mayat via Fink-devel" <fin...@li...> wrote: > > Good Day, > > While rebuilding the fluid synth package I got the following message: > > The following 230 additional packages will be installed: > algorithm-c3-pm app-defaults authen-ntlm-pm5303 autoconf2.6 automake1.15 automake1.15-core bison blt-dev > blt-shlibs capture-tiny-pm class-c3-pm5303 class-c3-xs-pm5303 cmake compress-raw-bzip2-pm5303 > compress-raw-zlib-pm5303 cpan-meta-pm5303 cpan-meta-requirements-pm5303 cpan-meta-yaml-pm cscope > data-optlist-pm5303 data-section-pm5303 db53-aes db53-aes-shlibs dejagnu desktop-file-utils devel-symdump-pm > digest-hmac-pm docbook-dsssl-nwalsh docbook-dtd docbook-xsl doxygen encode-locale-pm5303 encode-pm5303 expect > exporter-pm extutils-cbuilder-pm extutils-install-pm extutils-makemaker-pm5303 extutils-manifest-pm > extutils-parsexs-pm file-find-rule-pm file-listing-pm5303 file-temp-pm5303 fink-buildenv-modules > fink-package-precedence flag-sort flex fontconfig-config fontconfig2-dev fontconfig2-shlibs freetype219 > freetype219-shlibs gdbm4 gdbm4-shlibs ghostscript ghostscript-fonts glib2-dev glib2-shlibs gtk-doc > help2man-perl5303 html-parser-pm5303 html-tagset-pm http-cookiejar-pm5303 http-cookies-pm5303 http-daemon-pm5303 > http-date-pm5303 http-message-pm5303 http-negotiate-pm5303 http-tiny-pm5303 inc-latest-pm5303 info intltool40 > io-compress-pm5303 io-html-pm5303 io-socket-inet6-pm5303 io-socket-ip-pm5303 io-socket-ssl-pm5303 itstool json-c5 > json-c5-shlibs json-pp-pm ladspa ladspa-dev lame-dev lame-shlibs lcms2 lcms2-shlibs lcms2mt2 lcms2mt2-shlibs > libbrotli1 libbrotli1-shlibs libdeflate0-dev libdeflate0-shlibs libffi8 libffi8-shlibs libflac12 libflac12-dev > libidn libidn-shlibs libidn2.0-dev libidn2.0-shlibs libijs1-dev libijs1-shlibs libjbig libjbig-shlibs > libjbig2dec-dev libjbig2dec-shlibs libjpeg-bin libjpeg9 libjpeg9-shlibs libmpg123 libmpg123-shlibs libogg > libogg-shlibs libopenjp2.7 libopenjp2.7-shlibs libopus0-dev libopus0-shlibs libpaper1-dev libpaper1-shlibs > libpng16 libpng16-shlibs libsigsegv3 libsigsegv3-shlibs libsndfile1-dev libsndfile1-shlibs libtiff5 > libtiff5-shlibs libtiff6 libtiff6-shlibs libtool2 libtool2-shlibs libunistring2 libunistring2-shlibs libvorbis0 > libvorbis0-shlibs libwww-pm5303 libxml2 libxml2-bin libxml2-py27 libxml2-shlibs libxslt libxslt-bin > libxslt-shlibs libxt libxt-shlibs libzstd1-dev libzstd1-shlibs locale-gettext-pm5303 lwp-mediatypes-pm5303 > lz4-dev lz4-shlibs m4 module-build-pm5303 module-metadata-pm5303 mozilla-ca-pm mro-compat-pm5303 net-http-pm5303 > net-ssleay-pm5303 number-compare-pm openjade opensp-bin opensp5-dev opensp5-shlibs openssl openssl110-dev > openssl110-shlibs openssl300-dev openssl300-shlibs package-generator-pm par-dist-pm params-util-pm5303 > perl-ostype-pm5303 pkgconfig pkgconfig-common pod-coverage-pm pod-readme-pm pod-simple-pm podlators-pm > ppkg-config python27 python27-shlibs readline8 readline8-shlibs regexp-common-pm scalar-list-utils-pm5303 sdl2 > sdl2-shlibs sgml-entities-iso8879 shared-mime-info socket6-pm5303 software-license-pm5303 sqlite3-dev > sqlite3-shlibs sub-exporter-pm5303 sub-install-pm sub-name-pm5303 sub-uplevel-pm tcltk tcltk-dev tcltk-shlibs > test-distribution-pm5303 test-exception-pm test-failwarnings-pm test-harness-pm5303 test-leaktrace-pm5303 > test-needs-pm test-nowarnings-pm test-pod-coverage-pm test-pod-pm test-portability-files-pm test-requires-pm > test-requiresinternet-pm5303 test-simple-pm test-simple-pm5303 test-warn-pm texinfo texinfo-legacy text-glob-pm > text-template-pm try-tiny-pm uri-pm5303 version-pm5303 www-robotrules-pm5303 xdg-base xft2-dev xft2-shlibs > xml-parser-pm5303 xsloader-pm5303 yaml-pm > > But, … most of the packages listed above are simply redundant and not in any way dependencies.This is surely abnormal? > > Any insights, suggestions would be appreciated. > E > > _______________________________________________ > Fink-devel mailing list > Fin...@li... > List archive: > https://sourceforge.net/p/fink/mailman/fink-devel > Subscription management: > https://lists.sourceforge.net/lists/listinfo/fink-devel > > > |
From: Daniel M. <dm...@ne...> - 2023-05-13 17:39:33
|
That's pretty typical for a high-level package on a machine that has not had a lot of other packages built on it. Some are obviously the direct needs of fluidsynth itself, some are infrastructure (tools used to build various other packages), and some are dependencies of those two types of packages (lots of lower-level libraries widely used by many things, recursively). I just parsed the dependency tree of fluidsynth1 and can confirm that this list looks correct. My first attempt was to look at a visual "dotty" graph of it, but with over 200 nodes and over 1100 edges, it's not too useful except to say "wow that's complicated!" and to see that there didn't appear to be any large disjoint sections. If you have a few certain packages in that list that are interesting or curious, let me know and I will trace them specifically. For the record, here's the tool I used. 1. Create a file 'goal' that contains a list of packages to query, one per line (your "...will be installed" set and the fluidsynth packages themselves). Order is not important, but make sure there is no other whitespace. 2. fink -m list --format=dotty-build > pkg.dot 3. perl -ni.bak -e 'print if /$p/; } BEGIN {open $h, "<", "goal"; chomp(@p=<$h>);close $h;$p=~s/ /|/g;$p=join "|", @p;$p=qr/"($p)" /' pkg.dot 4. Copy the first three lines and last one line from 'pkg.dot.bak' to the start and end, respectively, of 'pkg.dot'. 'pkg.dot' is now a directed graph of all of the dependencies among all of the packages listed in 'goal'. dan On 5/13/23, 6:47 AM, "Ebrahim Mayat via Fink-devel" <fin...@li...> wrote: Good Day, While rebuilding the fluid synth package I got the following message: The following 230 additional packages will be installed: algorithm-c3-pm app-defaults authen-ntlm-pm5303 autoconf2.6 automake1.15 automake1.15-core bison blt-dev blt-shlibs capture-tiny-pm class-c3-pm5303 class-c3-xs-pm5303 cmake compress-raw-bzip2-pm5303 compress-raw-zlib-pm5303 cpan-meta-pm5303 cpan-meta-requirements-pm5303 cpan-meta-yaml-pm cscope data-optlist-pm5303 data-section-pm5303 db53-aes db53-aes-shlibs dejagnu desktop-file-utils devel-symdump-pm digest-hmac-pm docbook-dsssl-nwalsh docbook-dtd docbook-xsl doxygen encode-locale-pm5303 encode-pm5303 expect exporter-pm extutils-cbuilder-pm extutils-install-pm extutils-makemaker-pm5303 extutils-manifest-pm extutils-parsexs-pm file-find-rule-pm file-listing-pm5303 file-temp-pm5303 fink-buildenv-modules fink-package-precedence flag-sort flex fontconfig-config fontconfig2-dev fontconfig2-shlibs freetype219 freetype219-shlibs gdbm4 gdbm4-shlibs ghostscript ghostscript-fonts glib2-dev glib2-shlibs gtk-doc help2man-perl5303 html-parser-pm5303 html-tagset-pm http-cookiejar-pm5303 http-cookies-pm5303 http-daemon-pm5303 http-date-pm5303 http-message-pm5303 http-negotiate-pm5303 http-tiny-pm5303 inc-latest-pm5303 info intltool40 io-compress-pm5303 io-html-pm5303 io-socket-inet6-pm5303 io-socket-ip-pm5303 io-socket-ssl-pm5303 itstool json-c5 json-c5-shlibs json-pp-pm ladspa ladspa-dev lame-dev lame-shlibs lcms2 lcms2-shlibs lcms2mt2 lcms2mt2-shlibs libbrotli1 libbrotli1-shlibs libdeflate0-dev libdeflate0-shlibs libffi8 libffi8-shlibs libflac12 libflac12-dev libidn libidn-shlibs libidn2.0-dev libidn2.0-shlibs libijs1-dev libijs1-shlibs libjbig libjbig-shlibs libjbig2dec-dev libjbig2dec-shlibs libjpeg-bin libjpeg9 libjpeg9-shlibs libmpg123 libmpg123-shlibs libogg libogg-shlibs libopenjp2.7 libopenjp2.7-shlibs libopus0-dev libopus0-shlibs libpaper1-dev libpaper1-shlibs libpng16 libpng16-shlibs libsigsegv3 libsigsegv3-shlibs libsndfile1-dev libsndfile1-shlibs libtiff5 libtiff5-shlibs libtiff6 libtiff6-shlibs libtool2 libtool2-shlibs libunistring2 libunistring2-shlibs libvorbis0 libvorbis0-shlibs libwww-pm5303 libxml2 libxml2-bin libxml2-py27 libxml2-shlibs libxslt libxslt-bin libxslt-shlibs libxt libxt-shlibs libzstd1-dev libzstd1-shlibs locale-gettext-pm5303 lwp-mediatypes-pm5303 lz4-dev lz4-shlibs m4 module-build-pm5303 module-metadata-pm5303 mozilla-ca-pm mro-compat-pm5303 net-http-pm5303 net-ssleay-pm5303 number-compare-pm openjade opensp-bin opensp5-dev opensp5-shlibs openssl openssl110-dev openssl110-shlibs openssl300-dev openssl300-shlibs package-generator-pm par-dist-pm params-util-pm5303 perl-ostype-pm5303 pkgconfig pkgconfig-common pod-coverage-pm pod-readme-pm pod-simple-pm podlators-pm ppkg-config python27 python27-shlibs readline8 readline8-shlibs regexp-common-pm scalar-list-utils-pm5303 sdl2 sdl2-shlibs sgml-entities-iso8879 shared-mime-info socket6-pm5303 software-license-pm5303 sqlite3-dev sqlite3-shlibs sub-exporter-pm5303 sub-install-pm sub-name-pm5303 sub-uplevel-pm tcltk tcltk-dev tcltk-shlibs test-distribution-pm5303 test-exception-pm test-failwarnings-pm test-harness-pm5303 test-leaktrace-pm5303 test-needs-pm test-nowarnings-pm test-pod-coverage-pm test-pod-pm test-portability-files-pm test-requires-pm test-requiresinternet-pm5303 test-simple-pm test-simple-pm5303 test-warn-pm texinfo texinfo-legacy text-glob-pm text-template-pm try-tiny-pm uri-pm5303 version-pm5303 www-robotrules-pm5303 xdg-base xft2-dev xft2-shlibs xml-parser-pm5303 xsloader-pm5303 yaml-pm But, … most of the packages listed above are simply redundant and not in any way dependencies.This is surely abnormal? Any insights, suggestions would be appreciated. E _______________________________________________ Fink-devel mailing list Fin...@li... List archive: https://sourceforge.net/p/fink/mailman/fink-devel Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-devel |
From: Ebrahim M. <eb...@ma...> - 2023-05-13 10:46:49
|
Good Day, While rebuilding the fluid synth package I got the following message: The following 230 additional packages will be installed: algorithm-c3-pm app-defaults authen-ntlm-pm5303 autoconf2.6 automake1.15 automake1.15-core bison blt-dev blt-shlibs capture-tiny-pm class-c3-pm5303 class-c3-xs-pm5303 cmake compress-raw-bzip2-pm5303 compress-raw-zlib-pm5303 cpan-meta-pm5303 cpan-meta-requirements-pm5303 cpan-meta-yaml-pm cscope data-optlist-pm5303 data-section-pm5303 db53-aes db53-aes-shlibs dejagnu desktop-file-utils devel-symdump-pm digest-hmac-pm docbook-dsssl-nwalsh docbook-dtd docbook-xsl doxygen encode-locale-pm5303 encode-pm5303 expect exporter-pm extutils-cbuilder-pm extutils-install-pm extutils-makemaker-pm5303 extutils-manifest-pm extutils-parsexs-pm file-find-rule-pm file-listing-pm5303 file-temp-pm5303 fink-buildenv-modules fink-package-precedence flag-sort flex fontconfig-config fontconfig2-dev fontconfig2-shlibs freetype219 freetype219-shlibs gdbm4 gdbm4-shlibs ghostscript ghostscript-fonts glib2-dev glib2-shlibs gtk-doc help2man-perl5303 html-parser-pm5303 html-tagset-pm http-cookiejar-pm5303 http-cookies-pm5303 http-daemon-pm5303 http-date-pm5303 http-message-pm5303 http-negotiate-pm5303 http-tiny-pm5303 inc-latest-pm5303 info intltool40 io-compress-pm5303 io-html-pm5303 io-socket-inet6-pm5303 io-socket-ip-pm5303 io-socket-ssl-pm5303 itstool json-c5 json-c5-shlibs json-pp-pm ladspa ladspa-dev lame-dev lame-shlibs lcms2 lcms2-shlibs lcms2mt2 lcms2mt2-shlibs libbrotli1 libbrotli1-shlibs libdeflate0-dev libdeflate0-shlibs libffi8 libffi8-shlibs libflac12 libflac12-dev libidn libidn-shlibs libidn2.0-dev libidn2.0-shlibs libijs1-dev libijs1-shlibs libjbig libjbig-shlibs libjbig2dec-dev libjbig2dec-shlibs libjpeg-bin libjpeg9 libjpeg9-shlibs libmpg123 libmpg123-shlibs libogg libogg-shlibs libopenjp2.7 libopenjp2.7-shlibs libopus0-dev libopus0-shlibs libpaper1-dev libpaper1-shlibs libpng16 libpng16-shlibs libsigsegv3 libsigsegv3-shlibs libsndfile1-dev libsndfile1-shlibs libtiff5 libtiff5-shlibs libtiff6 libtiff6-shlibs libtool2 libtool2-shlibs libunistring2 libunistring2-shlibs libvorbis0 libvorbis0-shlibs libwww-pm5303 libxml2 libxml2-bin libxml2-py27 libxml2-shlibs libxslt libxslt-bin libxslt-shlibs libxt libxt-shlibs libzstd1-dev libzstd1-shlibs locale-gettext-pm5303 lwp-mediatypes-pm5303 lz4-dev lz4-shlibs m4 module-build-pm5303 module-metadata-pm5303 mozilla-ca-pm mro-compat-pm5303 net-http-pm5303 net-ssleay-pm5303 number-compare-pm openjade opensp-bin opensp5-dev opensp5-shlibs openssl openssl110-dev openssl110-shlibs openssl300-dev openssl300-shlibs package-generator-pm par-dist-pm params-util-pm5303 perl-ostype-pm5303 pkgconfig pkgconfig-common pod-coverage-pm pod-readme-pm pod-simple-pm podlators-pm ppkg-config python27 python27-shlibs readline8 readline8-shlibs regexp-common-pm scalar-list-utils-pm5303 sdl2 sdl2-shlibs sgml-entities-iso8879 shared-mime-info socket6-pm5303 software-license-pm5303 sqlite3-dev sqlite3-shlibs sub-exporter-pm5303 sub-install-pm sub-name-pm5303 sub-uplevel-pm tcltk tcltk-dev tcltk-shlibs test-distribution-pm5303 test-exception-pm test-failwarnings-pm test-harness-pm5303 test-leaktrace-pm5303 test-needs-pm test-nowarnings-pm test-pod-coverage-pm test-pod-pm test-portability-files-pm test-requires-pm test-requiresinternet-pm5303 test-simple-pm test-simple-pm5303 test-warn-pm texinfo texinfo-legacy text-glob-pm text-template-pm try-tiny-pm uri-pm5303 version-pm5303 www-robotrules-pm5303 xdg-base xft2-dev xft2-shlibs xml-parser-pm5303 xsloader-pm5303 yaml-pm But, … most of the packages listed above are simply redundant and not in any way dependencies.This is surely abnormal? Any insights, suggestions would be appreciated. E |
From: Scott H. <sha...@us...> - 2023-02-15 16:26:22
|
This emacs28 has lead me a long way from where I started. I got emacs28-nox to build correctly on OS 12. To build emacs28 requires a LOT of other dependencies. I have glib2 v2.74.5 that builds on OS 12. However it also requires and update to gtk+. I figured that updating from 2.18.9 to 2.24.22 would not be major changes. However, I may have had misplaced optimism! I started with just updating the current gtk+2 info with a new version number and removing a lot of the patches that seem to have been incorporated into the newer source. It seems that the pixmap subdirectory does not get configured. Is that done by the main configure routine? Do I need to tell it to work with the sub directory though some option that I haven’t found? The pixmap subdirectory has a Makefile.in but no Makefile is generated with the current settings. There are pixmap files in /opt/sw/bin but I assume they are installed as part of gtk+2. The error is configure: error: Package requirements (glib-2.0 >= 2.28.0 atk >= 1.29.2 pango >= 1.20 cairo >= 1.6 gdk-pixbuf-2.0 >= 2.21.0) were not met: Requested 'gdk-pixbuf-2.0 >= 2.21.0' but version of GdkPixbuf is 2.18.9 Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. gdk-pixbuf is not a separate package so I assume that building the one that is included with the gtk+2 package should be the correct version? Anyway to get that directory configured and then built so it can be used/installed. Any advice on this would be much appreciated. Scott |
From: Hanspeter N. <fi...@sn...> - 2022-12-18 01:38:05
|
On 12/17/22 3:44 PM, Daniel Macks wrote: > There is a distribution-restriction on go: version 1.14.6 is for OS X <= 10.11 and version 1.16.6 is for OS X 10.12-11.3. Hanspeter, it seems like your reason for the limits would allow 1.16.6 to be suitable for 12.x also. But I don't have a 12.x sandbox to test it. Unfortunately the distribution tag doesn't accept wildcards, so once there is any restriction, all allowed dists need to be enumerated explicitly. > > dan > > On 12/17/22, 4:01 PM, "Scott Hannahs via Fink-users" <fin...@li...> wrote: > > Seems there is no version of “go” to install GitHub? MacOS 12. > > -Scott > > % fink install github > Information about 13643 packages read in 1 seconds. > The package 'github' will be built and installed. > Reading build dependency for github-2.14.2-1... > WARNING: While resolving dependency "go >= 1.14.6-1" for package "github-2.14.2-1", package "go" was not found. > Can't resolve dependency "go (>= 1.14.6-1)" for package "github-2.14.2-1" (no matching packages/versions found) > Exiting with failure. I've pushed a blind change allowing go on macOS 12. Hopefully it builds, since the restriction on `_clock_gettime` shouldn't apply here. Hanspeter -- "The great tragedy of science is the slaying of a beautiful hypothesis by an ugly fact." --Thomas Henry Huxley |
From: Daniel M. <dm...@ne...> - 2022-12-17 21:45:14
|
There is a distribution-restriction on go: version 1.14.6 is for OS X <= 10.11 and version 1.16.6 is for OS X 10.12-11.3. Hanspeter, it seems like your reason for the limits would allow 1.16.6 to be suitable for 12.x also. But I don't have a 12.x sandbox to test it. Unfortunately the distribution tag doesn't accept wildcards, so once there is any restriction, all allowed dists need to be enumerated explicitly. dan On 12/17/22, 4:01 PM, "Scott Hannahs via Fink-users" <fin...@li...> wrote: Seems there is no version of “go” to install GitHub? MacOS 12. -Scott % fink install github Information about 13643 packages read in 1 seconds. The package 'github' will be built and installed. Reading build dependency for github-2.14.2-1... WARNING: While resolving dependency "go >= 1.14.6-1" for package "github-2.14.2-1", package "go" was not found. Can't resolve dependency "go (>= 1.14.6-1)" for package "github-2.14.2-1" (no matching packages/versions found) Exiting with failure. % _______________________________________________ Fink-users mailing list Fin...@li... List archive: https://sourceforge.net/p/fink/mailman/fink-users/ Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-users |
From: Scott H. <sha...@us...> - 2022-12-17 21:01:18
|
Seems there is no version of “go” to install GitHub? MacOS 12. -Scott % fink install github Information about 13643 packages read in 1 seconds. The package 'github' will be built and installed. Reading build dependency for github-2.14.2-1... WARNING: While resolving dependency "go >= 1.14.6-1" for package "github-2.14.2-1", package "go" was not found. Can't resolve dependency "go (>= 1.14.6-1)" for package "github-2.14.2-1" (no matching packages/versions found) Exiting with failure. % |
From: <fi...@sn...> - 2022-12-13 10:26:27
|
On 12/12/22 3:28 PM, Scott Hannahs via Fink-devel wrote: > > >> On Dec 12, 2022, at 16:10, fi...@sn... wrote: >> >> On 2022-12-12 14:58, Scott Hannahs via Fink-devel wrote: >>> Ok, weirdness. I could not get it to reproduce on my home machine >>> after building with single job, and it has build correctly with >>> multiple jobs enabled about 3 times after that. I have a similar >>> machine at work (ie Monterey 12.6.1) and I tried to install gcc12 on >>> that system. This gave a different error. My Xcode and command line >>> versions agree. >>>> checking for the correct version of gmp.h... yes >>>> checking for the correct version of mpfr.h... yes >>>> checking for the correct version of mpc.h... no >>>> configure: error: Building GCC requires GMP 4.2+, MPFR 3.1.0+ and >>>> MPC 0.8.0+. >>>> Try the --with-gmp, --with-mpfr and/or --with-mpc options to specify >>>> their locations. Source code for these libraries can be found at >>>> their respective hosting sites as well as at >>>> https://gcc.gnu.org/pub/gcc/infrastructure/. See also >>>> http://gcc.gnu.org/install/prerequisites.html for additional info. >>>> If >>>> you obtained GMP, MPFR and/or MPC from a vendor distribution >>>> package, >>>> make sure that you have installed both the libraries and the header >>>> files. They may be located in separate packages. >>>> ### execution of /tmp/fink.SqDfX failed, exit code 1 >>>> ### execution of /tmp/fink.3lPCk failed, exit code 1 >>>> ... >>>> Also include the following system information: >>>> Package manager version: 0.45.99.git >>>> Distribution version: selfupdate-git Mon Dec 12 10:03:17 2022, 12.0, >>>> x86_64 >>>> Trees: local/injected local/main stable/main >>>> Xcode.app: 14.1 >>>> Xcode command-line tools: 14.1.0.0.1.1666437224 >>>> Max. Fink build jobs: 12 >>> Which seems to imply that my version of MPC < 0.8 or completely >>> missing. >>> BUT, fink reports that I have libmpc3 installed. Should it be >>> libmpc2?? Is the dependency in gcc12 wrong? It claims dependency on >>> libmpc3 >= 1.0.3 which should be satisfied. >>>> % fink list libmpc2 libmpc3 >>>> Information about 13595 packages read in 1 seconds. >>>> libmpc2 0.9-2 >>>> Multi-precision complex lib >>>> libmpc2-shlibs 0.9-2 >>>> Multi-precision complex lib >>>> i libmpc3 1.3.0-1 >>>> Multi-precision complex lib >>>> i libmpc3-shlibs 1.3.0-1 >>>> Multi-precision complex lib >>> The reference file has the correct version in it as expected: >>> "/opt/sw/include/mpc.h" >>>> /* Define MPC version number */ >>>> #define MPC_VERSION_MAJOR 1 >>>> #define MPC_VERSION_MINOR 3 >>>> #define MPC_VERSION_PATCHLEVEL 0 >>>> #define MPC_VERSION_STRING "1.3.0" >> >> gcc12 has an explicit Depends on `libmpc3 (>= 1.0.3-1)`. Can you look at config.log in the build directory and see what exactly failed when checking for `mpc.h`? >> >> libmpc3 just got updated from 1.2.1 to 1.3.0. Upstream claims that 1.3.0 is backwards compatible with 1.2.1, but I wonder if the gcc12 code is expecting something else that got changed. >> >> I don't know what was going on with the multi-job issue because the gccNN build system should be multi-thread safe. It's worked for me with 4 and 8 CPU machines for a long time, and I know others have it working too. >> >> Hanspeter > > > It seems that there is an error (or undefined type FILE) in that header file? The config.log file is in the darwin_objdir directory. > > >> configure:8209: checking for the correct version of mpc.h >> configure:8226: gcc -c -g -O2 -I/opt/sw/include -I/opt/sw/include conftest.c >&5 >> In file included from conftest.c:10: >> /opt/sw/include/mpc.h:287:35: error: unknown type name 'FILE' >> __MPC_DECLSPEC void mpcr_out_str (FILE *f, mpcr_srcptr r); >> ^ >> 1 error generated. >> configure:8226: $? = 1 >> configure: failed program was: >> | /* confdefs.h */ >> | #define PACKAGE_NAME "" >> | #define PACKAGE_TARNAME "" >> | #define PACKAGE_VERSION "" >> | #define PACKAGE_STRING "" >> | #define PACKAGE_BUGREPORT "" >> | #define PACKAGE_URL "" >> | #define LT_OBJDIR ".libs/" >> | /* end confdefs.h. */ >> | #include <mpc.h> >> | int >> | main () >> | { >> | >> | #if MPC_VERSION < MPC_VERSION_NUM(0,8,0) >> | choke me >> | #endif >> | >> | ; >> | return 0; >> | } >> configure:8251: result: no >> configure:8309: error: Building GCC requires GMP 4.2+, MPFR 3.1.0+ and MPC 0.8.0+. >> Try the --with-gmp, --with-mpfr and/or --with-mpc options to specify >> their locations. Source code for these libraries can be found at >> their respective hosting sites as well as at > > > At least to me it appears that the correct flag is passed to the compiler that has the mpc libraries in the path. That should point the system to /opt/sw/include/mpc.h > >> It was created by configure, which was >> generated by GNU Autoconf 2.69. Invocation command line was >> >> $ ../gcc-12.2.0/configure --prefix=/opt/sw --prefix=/opt/sw/lib/gcc12 --mandir=/opt/sw/share/man --infodir=/opt/sw/lib/gcc12/info --enable-languages=c,c++,fortran,lto,objc,obj-c++ --with-gmp=/opt/sw --with-libiconv-prefix=/opt/sw --wi >> th-isl=/opt/sw --with-mpc=/opt/sw --with-system-zlib --program-suffix=-fsf-12 --disable-multilib --enable-stage1-checking --with-sysroot=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk --wi >> th-pkgversion=Fink gcc12 12.2.0-2 --with-bugurl=https://github.com/fink/fink-distributions/issues It's a problem with libmpc3-1.3.0. Building gcc11 and gcc12 fail if I have it installed, but work if I have libmpc3-1.2.1 installed. Upstream has a fix: https://gitlab.inria.fr/mpc/mpc/-/commit/e944aa454e60cbff8ab4e8c70dd974083398378f I've pushed an update to libmpc3 with that fix and gccNN now build. Hanspeter -- 92% of all statistics are made up. |
From: Scott H. <st...@ma...> - 2022-12-13 00:00:28
|
> On Dec 12, 2022, at 16:10, fi...@sn... wrote: > > On 2022-12-12 14:58, Scott Hannahs via Fink-devel wrote: >> Ok, weirdness. I could not get it to reproduce on my home machine >> after building with single job, and it has build correctly with >> multiple jobs enabled about 3 times after that. I have a similar >> machine at work (ie Monterey 12.6.1) and I tried to install gcc12 on >> that system. This gave a different error. My Xcode and command line >> versions agree. >>> checking for the correct version of gmp.h... yes >>> checking for the correct version of mpfr.h... yes >>> checking for the correct version of mpc.h... no >>> configure: error: Building GCC requires GMP 4.2+, MPFR 3.1.0+ and >>> MPC 0.8.0+. >>> Try the --with-gmp, --with-mpfr and/or --with-mpc options to specify >>> their locations. Source code for these libraries can be found at >>> their respective hosting sites as well as at >>> https://gcc.gnu.org/pub/gcc/infrastructure/. See also >>> http://gcc.gnu.org/install/prerequisites.html for additional info. >>> If >>> you obtained GMP, MPFR and/or MPC from a vendor distribution >>> package, >>> make sure that you have installed both the libraries and the header >>> files. They may be located in separate packages. >>> ### execution of /tmp/fink.SqDfX failed, exit code 1 >>> ### execution of /tmp/fink.3lPCk failed, exit code 1 >>> ... >>> Also include the following system information: >>> Package manager version: 0.45.99.git >>> Distribution version: selfupdate-git Mon Dec 12 10:03:17 2022, 12.0, >>> x86_64 >>> Trees: local/injected local/main stable/main >>> Xcode.app: 14.1 >>> Xcode command-line tools: 14.1.0.0.1.1666437224 >>> Max. Fink build jobs: 12 >> Which seems to imply that my version of MPC < 0.8 or completely >> missing. >> BUT, fink reports that I have libmpc3 installed. Should it be >> libmpc2?? Is the dependency in gcc12 wrong? It claims dependency on >> libmpc3 >= 1.0.3 which should be satisfied. >>> % fink list libmpc2 libmpc3 >>> Information about 13595 packages read in 1 seconds. >>> libmpc2 0.9-2 >>> Multi-precision complex lib >>> libmpc2-shlibs 0.9-2 >>> Multi-precision complex lib >>> i libmpc3 1.3.0-1 >>> Multi-precision complex lib >>> i libmpc3-shlibs 1.3.0-1 >>> Multi-precision complex lib >> The reference file has the correct version in it as expected: >> "/opt/sw/include/mpc.h" >>> /* Define MPC version number */ >>> #define MPC_VERSION_MAJOR 1 >>> #define MPC_VERSION_MINOR 3 >>> #define MPC_VERSION_PATCHLEVEL 0 >>> #define MPC_VERSION_STRING "1.3.0" > > gcc12 has an explicit Depends on `libmpc3 (>= 1.0.3-1)`. Can you look at config.log in the build directory and see what exactly failed when checking for `mpc.h`? > > libmpc3 just got updated from 1.2.1 to 1.3.0. Upstream claims that 1.3.0 is backwards compatible with 1.2.1, but I wonder if the gcc12 code is expecting something else that got changed. > > I don't know what was going on with the multi-job issue because the gccNN build system should be multi-thread safe. It's worked for me with 4 and 8 CPU machines for a long time, and I know others have it working too. > > Hanspeter It seems that there is an error (or undefined type FILE) in that header file? The config.log file is in the darwin_objdir directory. > configure:8209: checking for the correct version of mpc.h > configure:8226: gcc -c -g -O2 -I/opt/sw/include -I/opt/sw/include conftest.c >&5 > In file included from conftest.c:10: > /opt/sw/include/mpc.h:287:35: error: unknown type name 'FILE' > __MPC_DECLSPEC void mpcr_out_str (FILE *f, mpcr_srcptr r); > ^ > 1 error generated. > configure:8226: $? = 1 > configure: failed program was: > | /* confdefs.h */ > | #define PACKAGE_NAME "" > | #define PACKAGE_TARNAME "" > | #define PACKAGE_VERSION "" > | #define PACKAGE_STRING "" > | #define PACKAGE_BUGREPORT "" > | #define PACKAGE_URL "" > | #define LT_OBJDIR ".libs/" > | /* end confdefs.h. */ > | #include <mpc.h> > | int > | main () > | { > | > | #if MPC_VERSION < MPC_VERSION_NUM(0,8,0) > | choke me > | #endif > | > | ; > | return 0; > | } > configure:8251: result: no > configure:8309: error: Building GCC requires GMP 4.2+, MPFR 3.1.0+ and MPC 0.8.0+. > Try the --with-gmp, --with-mpfr and/or --with-mpc options to specify > their locations. Source code for these libraries can be found at > their respective hosting sites as well as at At least to me it appears that the correct flag is passed to the compiler that has the mpc libraries in the path. That should point the system to /opt/sw/include/mpc.h > It was created by configure, which was > generated by GNU Autoconf 2.69. Invocation command line was > > $ ../gcc-12.2.0/configure --prefix=/opt/sw --prefix=/opt/sw/lib/gcc12 --mandir=/opt/sw/share/man --infodir=/opt/sw/lib/gcc12/info --enable-languages=c,c++,fortran,lto,objc,obj-c++ --with-gmp=/opt/sw --with-libiconv-prefix=/opt/sw --wi > th-isl=/opt/sw --with-mpc=/opt/sw --with-system-zlib --program-suffix=-fsf-12 --disable-multilib --enable-stage1-checking --with-sysroot=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk --wi > th-pkgversion=Fink gcc12 12.2.0-2 --with-bugurl=https://github.com/fink/fink-distributions/issues > |
From: <fi...@sn...> - 2022-12-12 21:11:06
|
On 2022-12-12 14:58, Scott Hannahs via Fink-devel wrote: > Ok, weirdness. I could not get it to reproduce on my home machine > after building with single job, and it has build correctly with > multiple jobs enabled about 3 times after that. I have a similar > machine at work (ie Monterey 12.6.1) and I tried to install gcc12 on > that system. This gave a different error. My Xcode and command line > versions agree. > >> checking for the correct version of gmp.h... yes >> checking for the correct version of mpfr.h... yes >> checking for the correct version of mpc.h... no >> configure: error: Building GCC requires GMP 4.2+, MPFR 3.1.0+ and >> MPC 0.8.0+. >> Try the --with-gmp, --with-mpfr and/or --with-mpc options to specify >> their locations. Source code for these libraries can be found at >> their respective hosting sites as well as at >> https://gcc.gnu.org/pub/gcc/infrastructure/. See also >> http://gcc.gnu.org/install/prerequisites.html for additional info. >> If >> you obtained GMP, MPFR and/or MPC from a vendor distribution >> package, >> make sure that you have installed both the libraries and the header >> files. They may be located in separate packages. >> ### execution of /tmp/fink.SqDfX failed, exit code 1 >> ### execution of /tmp/fink.3lPCk failed, exit code 1 >> >> ... >> Also include the following system information: >> Package manager version: 0.45.99.git >> Distribution version: selfupdate-git Mon Dec 12 10:03:17 2022, 12.0, >> x86_64 >> Trees: local/injected local/main stable/main >> Xcode.app: 14.1 >> Xcode command-line tools: 14.1.0.0.1.1666437224 >> Max. Fink build jobs: 12 > > Which seems to imply that my version of MPC < 0.8 or completely > missing. > BUT, fink reports that I have libmpc3 installed. Should it be > libmpc2?? Is the dependency in gcc12 wrong? It claims dependency on > libmpc3 >= 1.0.3 which should be satisfied. > >> % fink list libmpc2 libmpc3 >> Information about 13595 packages read in 1 seconds. >> libmpc2 0.9-2 >> Multi-precision complex lib >> libmpc2-shlibs 0.9-2 >> Multi-precision complex lib >> i libmpc3 1.3.0-1 >> Multi-precision complex lib >> i libmpc3-shlibs 1.3.0-1 >> Multi-precision complex lib > > The reference file has the correct version in it as expected: > "/opt/sw/include/mpc.h" > >> /* Define MPC version number */ >> #define MPC_VERSION_MAJOR 1 >> #define MPC_VERSION_MINOR 3 >> #define MPC_VERSION_PATCHLEVEL 0 >> #define MPC_VERSION_STRING "1.3.0" gcc12 has an explicit Depends on `libmpc3 (>= 1.0.3-1)`. Can you look at config.log in the build directory and see what exactly failed when checking for `mpc.h`? libmpc3 just got updated from 1.2.1 to 1.3.0. Upstream claims that 1.3.0 is backwards compatible with 1.2.1, but I wonder if the gcc12 code is expecting something else that got changed. I don't know what was going on with the multi-job issue because the gccNN build system should be multi-thread safe. It's worked for me with 4 and 8 CPU machines for a long time, and I know others have it working too. Hanspeter |
From: Scott H. <st...@ma...> - 2022-12-12 20:59:02
|
Ok, weirdness. I could not get it to reproduce on my home machine after building with single job, and it has build correctly with multiple jobs enabled about 3 times after that. I have a similar machine at work (ie Monterey 12.6.1) and I tried to install gcc12 on that system. This gave a different error. My Xcode and command line versions agree. checking for the correct version of gmp.h... yes checking for the correct version of mpfr.h... yes checking for the correct version of mpc.h... no configure: error: Building GCC requires GMP 4.2+, MPFR 3.1.0+ and MPC 0.8.0+. Try the --with-gmp, --with-mpfr and/or --with-mpc options to specify their locations. Source code for these libraries can be found at their respective hosting sites as well as at https://gcc.gnu.org/pub/gcc/infrastructure/. See also http://gcc.gnu.org/install/prerequisites.html for additional info. If you obtained GMP, MPFR and/or MPC from a vendor distribution package, make sure that you have installed both the libraries and the header files. They may be located in separate packages. ### execution of /tmp/fink.SqDfX failed, exit code 1 ### execution of /tmp/fink.3lPCk failed, exit code 1 ... Also include the following system information: Package manager version: 0.45.99.git Distribution version: selfupdate-git Mon Dec 12 10:03:17 2022, 12.0, x86_64 Trees: local/injected local/main stable/main Xcode.app: 14.1 Xcode command-line tools: 14.1.0.0.1.1666437224 Max. Fink build jobs: 12 Which seems to imply that my version of MPC < 0.8 or completely missing. BUT, fink reports that I have libmpc3 installed. Should it be libmpc2?? Is the dependency in gcc12 wrong? It claims dependency on libmpc3 >= 1.0.3 which should be satisfied. % fink list libmpc2 libmpc3 Information about 13595 packages read in 1 seconds. libmpc2 0.9-2 Multi-precision complex lib libmpc2-shlibs 0.9-2 Multi-precision complex lib i libmpc3 1.3.0-1 Multi-precision complex lib i libmpc3-shlibs 1.3.0-1 Multi-precision complex lib The reference file has the correct version in it as expected: "/opt/sw/include/mpc.h" /* Define MPC version number */ #define MPC_VERSION_MAJOR 1 #define MPC_VERSION_MINOR 3 #define MPC_VERSION_PATCHLEVEL 0 #define MPC_VERSION_STRING "1.3.0" I haven’t tried building with a single process since that seems to permanently fix things for some bizarre reason. -Scott > On Dec 8, 2022, at 5:18 AM, fi...@sn... <mailto:fi...@sn...> wrote: > > Can you post the actual error you're getting? A gcc build should not be doing anything with `/opt/sw/bin/gawk` except running it for some string processing. And that should be thread independent. |
From: Scott H. <st...@ma...> - 2022-12-08 15:20:32
|
Sure, let me go reconfigure and rebuild and capture the output to a file. It cranks up the fans on my iMac to have those 10 threads going! :-) -Scott > On Dec 8, 2022, at 5:18 AM, fi...@sn... wrote: > > Can you post the actual error you're getting? A gcc build should not be doing anything with `/opt/sw/bin/gawk` except running it for some string processing. And that should be thread independent. |
From: <fi...@sn...> - 2022-12-08 10:19:13
|
On 12/7/22 6:17 PM, Scott Hannahs via Fink-devel wrote: > Fink is set up with the default 10 parallel jobs max. (Processor: 3.7 GHz 10-Core Intel Core i9) > > When I attempt to build gcc12 there is a race condition near the end where it tells me (falsely) that it can’t open the file or directory /opt/sw/bin/gawk. This happened twice in a row so it seems repeatable. > > If I reconfigure fink for 1 job max, it builds fine. So somewhere between 2 and 9 build jobs it fails. I think the info file can be edited to restrict the number of simultaneous build jobs? Can you post the actual error you're getting? A gcc build should not be doing anything with `/opt/sw/bin/gawk` except running it for some string processing. And that should be thread independent. Hanspeter -- I'm not a robot like you. I don't like having disks crammed into me... unless they're Oreos, and then only in the mouth. -- Fry |
From: Merle R. <mer...@ma...> - 2022-12-08 03:52:26
|
It works for me with Fink set to 8 parallel jobs. The laptop is a 2.3 GHz 8-core Intel Core i9 running macOS 12.6.1 and Xcode 14.1 Merle > On Dec 7, 2022, at 7:17 PM, Scott Hannahs via Fink-devel <fin...@li...> wrote: > > Fink is set up with the default 10 parallel jobs max. (Processor: 3.7 GHz 10-Core Intel Core i9) > > When I attempt to build gcc12 there is a race condition near the end where it tells me (falsely) that it can’t open the file or directory /opt/sw/bin/gawk. This happened twice in a row so it seems repeatable. > > If I reconfigure fink for 1 job max, it builds fine. So somewhere between 2 and 9 build jobs it fails. I think the info file can be edited to restrict the number of simultaneous build jobs? > > -Scott > > > > _______________________________________________ > Fink-devel mailing list > Fin...@li... > List archive: > https://sourceforge.net/p/fink/mailman/fink-devel > Subscription management: > https://lists.sourceforge.net/lists/listinfo/fink-devel |
From: Scott H. <st...@ma...> - 2022-12-08 01:52:01
|
Fink is set up with the default 10 parallel jobs max. (Processor: 3.7 GHz 10-Core Intel Core i9) When I attempt to build gcc12 there is a race condition near the end where it tells me (falsely) that it can’t open the file or directory /opt/sw/bin/gawk. This happened twice in a row so it seems repeatable. If I reconfigure fink for 1 job max, it builds fine. So somewhere between 2 and 9 build jobs it fails. I think the info file can be edited to restrict the number of simultaneous build jobs? -Scott |
From: Hisashi T F. <ht...@tw...> - 2022-03-28 18:22:52
|
Wow, that was 4643, not 4529. On Mon, 28 Mar 2022, Hisashi T Fujinaka via Fink-commits wrote: > Branch: refs/heads/master > Home: https://github.com/fink/fink-distributions > Commit: 5ed2fb6b2e0071e48f84956b982b68223a063728 > https://github.com/fink/fink-distributions/commit/5ed2fb6b2e0071e48f84956b982b68223a063728 > Author: Hisashi T Fujinaka <ht...@tw...> > Date: 2022-03-28 (Mon, 28 Mar 2022) > > Changed paths: > M 10.9-libcxx/stable/main/finkinfo/editors/vim.info > > Log Message: > ----------- > Vim: Welcome to Vim-8.2.4529. > > > > > _______________________________________________ > Fink-commits mailing list > Fin...@li... > http://news.gmane.org/gmane.os.apple.fink.cvs > -- Hisashi T Fujinaka - ht...@tw... BSEE + BSChem + BAEnglish + MSCS + $2.50 = coffee |
From: Alexander H. <ale...@gm...> - 2021-07-05 02:34:15
|
Hi, guys. I’ve been slammed to death at work for several years now, and I’ve been avoiding being on the computer at night unless it’s for gaming. However, I think I can scrape together a little time, particularly for focused tasks like “try this PR”. I just picked up an M1 Mini last week (and good thing, too, as my venerable MacBook hit the OS cap and won’t be getting Monterrey). I’d be happy to help test stuff out for Big Sur and later, and especially for Silicon. —akh |
From: Merle R. <mer...@ma...> - 2021-06-17 16:21:22
|
With at least Catalina, it isn't clear to me tht the Command Line Tools are actually needed to successfully build the fink packages IF the Xcode.app is installed. I have been seeing things build successfully (including gcc if I remove the dependency on the CLT). I haven't been quite as systematic in this testing as I should have been since I don't think there was anything left behind by prior macOS installs, but it may be worth looking at by others to see if perhaps Fink need to look for both, or if Xcode.app is sufficient to proceed. Merle > On Jun 17, 2021, at 10:03 AM, Scott Hannahs via Fink-devel <fin...@li...> wrote: > > Hanspeter, > > Sorry, freakin’ MS tagged this as spam in our system so I didn’t see it. > > I don’t believe that Catalina and Big Sur put those receipts in /System/Library/Receipts. Under those systems that volume is mounted read only and the App Store can not write to those directories without a boot into recovery mode for a system update. > > This leaves /private/var/db/receipts for the App store to put all the third party receipts and receipts for Apple App store software such as Numbers, Pages, Keynote and Xcode that are all there. HOWEVER the CLT does NOT go there. It may be trying to write to /System/Library/Receipts and failing. > > The pkgutil command will show that the CLT package is there, but without the actual plist and bom files it will disappear when it is rebuilt periodically. This may be a bug in the CLT package installer. Yes it is a pain that Apple keeps moving things around and tweaking the perl version. I have reported this on the Apple Bug system and see what happens! > > -Scott > >> On Jun 15, 2021, at 19:39, fi...@sn... wrote: >> >> On 6/12/21 7:26 PM, Scott Hannahs via Fink-devel wrote: >>> Both these versions of the command line tools create package info in the data base so that pkgutil can find them and they show up to fink as a virtual package. >>> HOWEVER, they do not actually create a bom and plist file in /private/var/db/receipts ! So when the database gets rebuild those packages disappear from the system and fink complains that you don’t have the command line tools installed. >>> Any work arounds for this? >> >> Unfortunately, it's been a long standing issue that any system update (and sometimes some security patches) breaks Fink's CLT detection. >> >> The receipt and bom file for the CLT are in /System/Library/Receipts but I don't know why pkgutil loses track of them. If there is a database that gets refreshed, one would think it would know about a /System location. >> >> And we can't just check for the presence of the receipt and/or bom files because you can have those installed but removed the actual CLT installation in /Library/Developer/CommandLineTools. Perhaps we could check for a few physical files in the install location... >> >> Hanspeter > > > > > > _______________________________________________ > Fink-devel mailing list > Fin...@li... > List archive: > https://sourceforge.net/p/fink/mailman/fink-devel > Subscription management: > https://lists.sourceforge.net/lists/listinfo/fink-devel |
From: Scott H. <sha...@us...> - 2021-06-17 15:37:52
|
Hanspeter, Sorry, freakin’ MS tagged this as spam in our system so I didn’t see it. I don’t believe that Catalina and Big Sur put those receipts in /System/Library/Receipts. Under those systems that volume is mounted read only and the App Store can not write to those directories without a boot into recovery mode for a system update. This leaves /private/var/db/receipts for the App store to put all the third party receipts and receipts for Apple App store software such as Numbers, Pages, Keynote and Xcode that are all there. HOWEVER the CLT does NOT go there. It may be trying to write to /System/Library/Receipts and failing. The pkgutil command will show that the CLT package is there, but without the actual plist and bom files it will disappear when it is rebuilt periodically. This may be a bug in the CLT package installer. Yes it is a pain that Apple keeps moving things around and tweaking the perl version. I have reported this on the Apple Bug system and see what happens! -Scott > On Jun 15, 2021, at 19:39, fi...@sn... wrote: > > On 6/12/21 7:26 PM, Scott Hannahs via Fink-devel wrote: >> Both these versions of the command line tools create package info in the data base so that pkgutil can find them and they show up to fink as a virtual package. >> HOWEVER, they do not actually create a bom and plist file in /private/var/db/receipts ! So when the database gets rebuild those packages disappear from the system and fink complains that you don’t have the command line tools installed. >> Any work arounds for this? > > Unfortunately, it's been a long standing issue that any system update (and sometimes some security patches) breaks Fink's CLT detection. > > The receipt and bom file for the CLT are in /System/Library/Receipts but I don't know why pkgutil loses track of them. If there is a database that gets refreshed, one would think it would know about a /System location. > > And we can't just check for the presence of the receipt and/or bom files because you can have those installed but removed the actual CLT installation in /Library/Developer/CommandLineTools. Perhaps we could check for a few physical files in the install location... > > Hanspeter |
From: <fi...@sn...> - 2021-06-15 23:39:50
|
On 6/12/21 7:26 PM, Scott Hannahs via Fink-devel wrote: > Both these versions of the command line tools create package info in the data base so that pkgutil can find them and they show up to fink as a virtual package. > > HOWEVER, they do not actually create a bom and plist file in /private/var/db/receipts ! So when the database gets rebuild those packages disappear from the system and fink complains that you don’t have the command line tools installed. > > Any work arounds for this? Unfortunately, it's been a long standing issue that any system update (and sometimes some security patches) breaks Fink's CLT detection. The receipt and bom file for the CLT are in /System/Library/Receipts but I don't know why pkgutil loses track of them. If there is a database that gets refreshed, one would think it would know about a /System location. And we can't just check for the presence of the receipt and/or bom files because you can have those installed but removed the actual CLT installation in /Library/Developer/CommandLineTools. Perhaps we could check for a few physical files in the install location... Hanspeter |
From: Scott H. <sha...@us...> - 2021-06-13 00:27:00
|
Both these versions of the command line tools create package info in the data base so that pkgutil can find them and they show up to fink as a virtual package. HOWEVER, they do not actually create a bom and plist file in /private/var/db/receipts ! So when the database gets rebuild those packages disappear from the system and fink complains that you don’t have the command line tools installed. Any work arounds for this? -Scott |
From: <fi...@sn...> - 2021-06-11 19:15:08
|
Everyone will have the po4a-XXX.info files for all distributions. Fink just uses the one that matches the Distribution set in %p/etc/fink.conf If `fink dumpinfo -finfofile` is showing po4a-11.0.info when running macOS11.4, then fink is confused about your system. 1) What does `fink dumpinfo -fallversions po4a` show? This will list all the versions that are available or installed on your system. Because each .info file for po4a for each distribution has a different Revision: value, you can tell what fink has and sees. po4a-11.0 has %r=401. po4a-11.3 has %r=501. Make sure you `fink selfupdate` first, because I just fixed something. 2) Look at the file `/opt/sw/lib/perl5/Fink/Bootstrap.pm`. Around line 272, it lists the darwin versions that get linked to `Dist: 11.3`. It should say "i386-apple-darwin20\.[4-6]". 3) To confirm, what does `uname -a` output? I think 11.4 should output darwin20.5. You might need to run `fink reinstall fink` to get your Fink to realize that you're on 11.4. In %p/etc/fink.conf, the Distribution field should show 11.3 for any current macOS11 >= 11.3. Hanspeter On 2021-06-11 13:37, Scott Hannahs via Fink-devel wrote: > Hanspeter, > > Got it, so it uses the highest distribution less than or equal to the > current OS version. (ie use of 11.3 or an 11.4 OS). > > I am confused that I have an 11.3 distribution dot-info file but my > 11.4 OS says it is using 11.0 dot-info file according to dump-info. > > I had already installed po4a under 11.3 and so it does not need an > update for 11.4, but I am then confused as to how dumpinfo is working? > If I do an install will it possibly use a different distribution than > dump-info is showing? > > -Scott > >> On Jun 11, 2021, at 11:44, fi...@sn... wrote: >> >> The possible values of the Distribution field are partially dictated >> by the system-perl version, especially when Apple changes the >> system-perl version on us in the middle of the same macOS release. >> This is why we have `Dist: 10.14` (for 10.14.0 through 10.14.4) and >> `Dist: 10.14.5` (for 10.14.5 through 10.14.6). On macOS11, `Dist: >> 11.0` is for 11.0 through 11.2.x (using /usr/bin/perl5.28). For 11.3 >> and up, Apple changed to /usr/bin/perl5.30, so 11.3 through 11.5beta >> should be using `Dist: 11.3`. If you updated to 11.4 and your Fink is >> now showing that it's on Dist:11.0, then you probably need to ./inject >> from a updated git pull of the Fink tree that knows about 11.4. >> >> https://github.com/nieder/fink/tree/dpkg1.16 >> >> Hanspeter >> >> >> On 2021-06-11 09:40, Scott Hannahs via Fink-devel wrote: >>> There is now an OS 11.3 variant for po4a, but since OS 11.4 is now >>> released, that OS reverts to using the 11.0 variant? I am confused >>> as >>> to how it picks the variant for that file. >>> OS 11.0 uses po4-11.0 >>> OS 11.1 uses ??? >>> OS 11.2 uses ??? >>> OS 11.3 uses po4-11.3 >>> OS 11.4 uses ??? >>> Can the Distribution field be >=11.0, <= 11.2 >>> and for the second one be >=11.3 >>> Does the distribution field use logical comparators on a two digit >>> version? Can it handle a 3 dot (3 digit?) version compare? >>> -Scott >>> _______________________________________________ >>> Fink-devel mailing list >>> Fin...@li... >>> List archive: >>> https://sourceforge.net/p/fink/mailman/fink-devel >>> Subscription management: >>> https://lists.sourceforge.net/lists/listinfo/fink-devel >> >> >> _______________________________________________ >> Fink-devel mailing list >> Fin...@li... >> List archive: >> https://sourceforge.net/p/fink/mailman/fink-devel >> Subscription management: >> https://lists.sourceforge.net/lists/listinfo/fink-devel > > > > _______________________________________________ > Fink-devel mailing list > Fin...@li... > List archive: > https://sourceforge.net/p/fink/mailman/fink-devel > Subscription management: > https://lists.sourceforge.net/lists/listinfo/fink-devel |
From: Scott H. <sha...@us...> - 2021-06-11 18:38:13
|
Hanspeter, Got it, so it uses the highest distribution less than or equal to the current OS version. (ie use of 11.3 or an 11.4 OS). I am confused that I have an 11.3 distribution dot-info file but my 11.4 OS says it is using 11.0 dot-info file according to dump-info. I had already installed po4a under 11.3 and so it does not need an update for 11.4, but I am then confused as to how dumpinfo is working? If I do an install will it possibly use a different distribution than dump-info is showing? -Scott > On Jun 11, 2021, at 11:44, fi...@sn... wrote: > > The possible values of the Distribution field are partially dictated by the system-perl version, especially when Apple changes the system-perl version on us in the middle of the same macOS release. This is why we have `Dist: 10.14` (for 10.14.0 through 10.14.4) and `Dist: 10.14.5` (for 10.14.5 through 10.14.6). On macOS11, `Dist: 11.0` is for 11.0 through 11.2.x (using /usr/bin/perl5.28). For 11.3 and up, Apple changed to /usr/bin/perl5.30, so 11.3 through 11.5beta should be using `Dist: 11.3`. If you updated to 11.4 and your Fink is now showing that it's on Dist:11.0, then you probably need to ./inject from a updated git pull of the Fink tree that knows about 11.4. > > https://github.com/nieder/fink/tree/dpkg1.16 > > Hanspeter > > > On 2021-06-11 09:40, Scott Hannahs via Fink-devel wrote: >> There is now an OS 11.3 variant for po4a, but since OS 11.4 is now >> released, that OS reverts to using the 11.0 variant? I am confused as >> to how it picks the variant for that file. >> OS 11.0 uses po4-11.0 >> OS 11.1 uses ??? >> OS 11.2 uses ??? >> OS 11.3 uses po4-11.3 >> OS 11.4 uses ??? >> Can the Distribution field be >=11.0, <= 11.2 >> and for the second one be >=11.3 >> Does the distribution field use logical comparators on a two digit >> version? Can it handle a 3 dot (3 digit?) version compare? >> -Scott >> _______________________________________________ >> Fink-devel mailing list >> Fin...@li... >> List archive: >> https://sourceforge.net/p/fink/mailman/fink-devel >> Subscription management: >> https://lists.sourceforge.net/lists/listinfo/fink-devel > > > _______________________________________________ > Fink-devel mailing list > Fin...@li... > List archive: > https://sourceforge.net/p/fink/mailman/fink-devel > Subscription management: > https://lists.sourceforge.net/lists/listinfo/fink-devel |