fink-devel — Discussion for Fink developers and package maintainers

Re: [Fink-devel] MirBSD and death notice

[Daniel M. <dm...@ne...>]

Thanks for being our freenode contact, even for sad details. Here's a link to some previous information about gecko:

https://sourceforge.net/p/fink/mailman/fink-devel/?viewmonth=201806&viewday=13

Our Contributors web-page is badly out of date:(

dan

On 5/11/25, 10:27 PM, "Alex Strange" <ast...@gm...> wrote:

    Due to never ever updating my IRC config, I'm the only person left in #fink on freenode. Which means it oddly falls to me to pass this sad news on.


    20:44 -!- mirabilos [~tg...@fr...1hdk.IP] has joined #fink
    20:44 < mirabilos> … huh, am I right here?
    20:46 < mirabilos> astrange: if you could please forward the following information to a suitable place, that would be nice
    20:46 < mirabilos> 1. some fink packages (at least joe/jupp in editors) pull in as distfile a file from MirBSD CVS, with an URL
                       like www.mirbsd.org/cvs.cgi/~checkout~/ <http://www.mirbsd.org/cvs.cgi/~checkout~/>…
    20:46 < mirabilos> due to AI bots hammering the main server, I had to put HTTP Basic Auth on it
    20:47 < mirabilos> user and password are currently both 'public', or change the hostname to mbsd.evolvis.org <http://mbsd.evolvis.org> which is a more
                       powerful VM that mirrors CVS and can serve the requests and currently does not have auth required
    20:47 < mirabilos> 2. Fink contributor gecko2 has been dead for a few years, unfortunately, so maybe move him from the “regular
                       contributors” list to… someplace suitable
    20:48 < mirabilos> he’s been the maintainer for these, but unfortunately I cannot prod him about this any more; he was a great
                       friend
    20:48 < mirabilos> ⇒ thanks!
    21:52 -!- mirabilos [~tg...@fr...1hdk.IP] has quit [Ping timeout: 120 seconds]


    _______________________________________________
    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] MirBSD and death notice

[Alex S. <ast...@gm...>]

Due to never ever updating my IRC config, I'm the only person left in #fink
on freenode. Which means it oddly falls to me to pass this sad news on.


> 20:44 -!- mirabilos [~tg...@fr...1hdk.IP] has joined
> #fink
> 20:44 < mirabilos> … huh, am I right here?
> 20:46 < mirabilos> astrange: if you could please forward the following
> information to a suitable place, that would be nice
> 20:46 < mirabilos> 1. some fink packages (at least joe/jupp in editors)
> pull in as distfile a file from MirBSD CVS, with an URL
>                    like www.mirbsd.org/cvs.cgi/~checkout~/…
> 20:46 < mirabilos> due to AI bots hammering the main server, I had to put
> HTTP Basic Auth on it
> 20:47 < mirabilos> user and password are currently both 'public', or
> change the hostname to mbsd.evolvis.org which is a more
>                    powerful VM that mirrors CVS and can serve the requests
> and currently does not have auth required
> 20:47 < mirabilos> 2. Fink contributor gecko2 has been dead for a few
> years, unfortunately, so maybe move him from the “regular
>                    contributors” list to… someplace suitable
> 20:48 < mirabilos> he’s been the maintainer for these, but unfortunately I
> cannot prod him about this any more; he was a great
>                    friend
> 20:48 < mirabilos> ⇒ thanks!
> 21:52 -!- mirabilos [~tg...@fr...1hdk.IP] has quit
> [Ping timeout: 120 seconds]

Re: [Fink-devel] [fink/fink-distributions] bf5494: Vim: Welcome to Vim-9.1.1238.

[Hanspeter N. <fi...@sn...>]

On 4/11/25 7:52 PM, Hisashi T Fujinaka wrote:
> I'm so bad at git. I'm not sure what I did.
> 
> On Fri, 11 Apr 2025, Hisashi T Fujinaka via Fink-commits wrote:
> 
>>  Branch: refs/heads/master
>>  Home:   https://github.com/fink/fink-distributions
>>  Commit: bf5494be3feea00968f3d8eee59d202ae641bb8c
>>      
>> https://github.com/fink/fink-distributions/commit/bf5494be3feea00968f3d8eee59d202ae641bb8c
>>  Author: Hisashi T Fujinaka <ht...@tw...>
>>  Date:   2025-04-11 (Fri, 11 Apr 2025)
>>
>>  Changed paths:
>>    M 10.9-libcxx/stable/main/finkinfo/editors/vim.info
>>
>>  Log Message:
>>  -----------
>>  Vim: Welcome to Vim-9.1.1238.
>>
>>
>>  Commit: aa80991f11e4dab7c441ed7a907380202a5d7941
>>      
>> https://github.com/fink/fink-distributions/commit/aa80991f11e4dab7c441ed7a907380202a5d7941
>>  Author: Hisashi T Fujinaka <ht...@tw...>
>>  Date:   2025-04-11 (Fri, 11 Apr 2025)
>>
>>  Changed paths:
>>    M 10.9-libcxx/stable/main/finkinfo/editors/vim.info
>>
>>  Log Message:
>>  -----------
>>  Vim: Welcome to Vim-9.1.1291.
>>
>>
>> Compare: 
>> https://github.com/fink/fink-distributions/compare/a0d7e3e8b10c...aa80991f11e4

This seems fine. Two commits to update vim to 1238 and then 1291, pushed 
at the same time, so the fink-commits mailing list shows then in one 
entry (but still two commits in the git history).

Hanspeter

-- 
There are 3 kinds of biologists: those who can count, and those who can't.

Re: [Fink-devel] [fink/fink-distributions] bf5494: Vim: Welcome to Vim-9.1.1238.

[Hisashi T F. <ht...@tw...>]

I'm so bad at git. I'm not sure what I did.

On Fri, 11 Apr 2025, Hisashi T Fujinaka via Fink-commits wrote:

>  Branch: refs/heads/master
>  Home:   https://github.com/fink/fink-distributions
>  Commit: bf5494be3feea00968f3d8eee59d202ae641bb8c
>      https://github.com/fink/fink-distributions/commit/bf5494be3feea00968f3d8eee59d202ae641bb8c
>  Author: Hisashi T Fujinaka <ht...@tw...>
>  Date:   2025-04-11 (Fri, 11 Apr 2025)
>
>  Changed paths:
>    M 10.9-libcxx/stable/main/finkinfo/editors/vim.info
>
>  Log Message:
>  -----------
>  Vim: Welcome to Vim-9.1.1238.
>
>
>  Commit: aa80991f11e4dab7c441ed7a907380202a5d7941
>      https://github.com/fink/fink-distributions/commit/aa80991f11e4dab7c441ed7a907380202a5d7941
>  Author: Hisashi T Fujinaka <ht...@tw...>
>  Date:   2025-04-11 (Fri, 11 Apr 2025)
>
>  Changed paths:
>    M 10.9-libcxx/stable/main/finkinfo/editors/vim.info
>
>  Log Message:
>  -----------
>  Vim: Welcome to Vim-9.1.1291.
>
>
> Compare: https://github.com/fink/fink-distributions/compare/a0d7e3e8b10c...aa80991f11e4
>
> To unsubscribe from these emails, change your notification settings at https://github.com/fink/fink-distributions/settings/notifications
>
>
> _______________________________________________
> 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

Re: [Fink-devel] Adding support for macOS 14.7

[Scott H. <st...@ma...>]

On Oct 5, 2024, at 8:24 PM, wren romano <win...@gm...> wrote:
> 
> I've recently switched to a new computer, and since it's currently unsupported I decided to try my hand at updating Fink to support it. With some minor adjustments to the version upper-bounds in `perlmod/Fink/{Services,Bootstrap}.pm`, the bootstrap script runs to completion and claims success. (However, despite it claiming success, it does have errors compiling libiconv-1.16-3, so it's unclear how successful it actually was.)
> 
> So my questions are:
> 
> (1) what sort of testing suite or protocol is there for checking things on a new os/arch? (i.e., before it's okay to update check_host() to say that the system is tested and supported)
> 
> (2) what sort of protocol is there for making a pull request on the github repo?
> 
> (3) I've noticed there's a lot of code with elsif-chains that could be greatly simplified (e.g., combining regexes, using a hash lookup instead, etc).  How amenable is the community to making such changes?
> 
> --
> Live well,
> ~wren

Wren,

1 - I don’t think there is a test suite for fink.  There might be someway to manually list architecture / OS versions that have been qualified in some way but that seems to be a lot of manual work.

Actually having an automated system that builds each package from a bare install would be something that could be run on a few different systems to validate all the packages.  Even if that ran once every couple of months it would be useful to catch missing dependencies.

2 - That was documented somewhere.   Basically clone the fink distributions repository into your own GitHub account.  Then branch for each package you want to update.  Then do a pull request from that branch back to the main repository.  That seems to be working for me.

3 - I will leave that up to the fink maintainers.  But if the code is easy to read and modify it should be welcomed.  The problem with complicated regex expressions is that they have unintended consequences when the input changes form a bit.  (ie going from 10.X to Y.Z or 10.9 to 10.10).

Just my $0.02 worth

-Scott

[Fink-devel] Adding support for macOS 14.7

[wren r. <win...@gm...>]

I've recently switched to a new computer, and since it's currently
unsupported I decided to try my hand at updating Fink to support it. With
some minor adjustments to the version upper-bounds in
`perlmod/Fink/{Services,Bootstrap}.pm`, the bootstrap script runs to
completion and claims success. (However, despite it claiming success, it
does have errors compiling libiconv-1.16-3, so it's unclear how successful
it actually was.)

So my questions are:

(1) what sort of testing suite or protocol is there for checking things on
a new os/arch? (i.e., before it's okay to update check_host() to say that
the system is tested and supported)

(2) what sort of protocol is there for making a pull request on the github
repo?

(3) I've noticed there's a lot of code with elsif-chains that could be
greatly simplified (e.g., combining regexes, using a hash lookup instead,
etc).  How amenable is the community to making such changes?

--
Live well,
~wren

Re: [Fink-devel] Updating pytest?

[<fi...@sn...>]

On 9/3/24 7:24 AM, Scott Hannahs via Fink-devel wrote:
> Got it.  And updating setuptools breaks too many other things?  Or setuptools-tng.  That is always the problem of a package manager is the interaction of different versions in the tool chain.

I've created a PR for an updated setuptools for testing: 
https://github.com/fink/fink-distributions/pull/1159

Hanspeter

-- 
+-------------------------------------+----------------------------+
| Hanspeter Niederstrasser            |  The Four F's of Survival: |
|                                     | Feeding, Fleeing, Fighting |
|                                     |    and ... Reproduction    |
|                                     |                            |
|                                     |          -Unknown          |
+-------------------------------------+----------------------------+

Re: [Fink-devel] Updating pytest?

[Scott H. <sha...@us...>]

Got it.  And updating setuptools breaks too many other things?  Or setuptools-tng.  That is always the problem of a package manager is the interaction of different versions in the tool chain.

Thanks,
Scott

> On Sep 3, 2024, at 7:34 AM, fi...@sn... wrote:
> 
> On 9/2/24 3:15 PM, Scott Hannahs via Fink-devel wrote:
>> Fink-experts,
>> The python build system is confusing me.  So, my package has a minor update to duplicity.  It shows some errors during testing.  I talking to the developers they of course blame the fact that ptyest is out of date with fink.  This leads to a few questions.
>> 1. So will updating pytest from version 7.4.4 to 8.3.2 break a whole bunch of stuff?  Should I create a pytest8 package?
>> 2. If I merely change the version, hash, and supported python versions, it builds but then there are no executables in /opt/sw/bin.  There are not executables to rename from pytest to pytest-py3.10.  Am I missing something obvious in this build process?  In looking at the documentation for pytest it seems it should create the executable.  If it doesn’t it will certainly break a lot of stuff, which is not good.
> 
> The problem is that newest pytest needs setuptools > 61, which we don't have. 8.0.2 is the latest that builds with our setuptools, but it's failing TestScript during the prep stage. Working to see if that can be worked around.
> 
> Hanspeter
> 
> -- 
> They that can give up essential liberty to obtain a little temporary
> safety deserve neither liberty nor safety.
>  --Benjamin Franklin
> 

Re: [Fink-devel] Updating pytest?

[<fi...@sn...>]

On 9/2/24 3:15 PM, Scott Hannahs via Fink-devel wrote:
> Fink-experts,
> 
> The python build system is confusing me.  So, my package has a minor update to duplicity.  It shows some errors during testing.  I talking to the developers they of course blame the fact that ptyest is out of date with fink.  This leads to a few questions.
> 
> 1. So will updating pytest from version 7.4.4 to 8.3.2 break a whole bunch of stuff?  Should I create a pytest8 package?
> 
> 2. If I merely change the version, hash, and supported python versions, it builds but then there are no executables in /opt/sw/bin.  There are not executables to rename from pytest to pytest-py3.10.  Am I missing something obvious in this build process?  In looking at the documentation for pytest it seems it should create the executable.  If it doesn’t it will certainly break a lot of stuff, which is not good.

The problem is that newest pytest needs setuptools > 61, which we don't 
have. 8.0.2 is the latest that builds with our setuptools, but it's 
failing TestScript during the prep stage. Working to see if that can be 
worked around.

Hanspeter

-- 
They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety.
   --Benjamin Franklin

[Fink-devel] Updating pytest?

[Scott H. <sha...@us...>]

Fink-experts,

The python build system is confusing me.  So, my package has a minor update to duplicity.  It shows some errors during testing.  I talking to the developers they of course blame the fact that ptyest is out of date with fink.  This leads to a few questions.

1. So will updating pytest from version 7.4.4 to 8.3.2 break a whole bunch of stuff?  Should I create a pytest8 package?

2. If I merely change the version, hash, and supported python versions, it builds but then there are no executables in /opt/sw/bin.  There are not executables to rename from pytest to pytest-py3.10.  Am I missing something obvious in this build process?  In looking at the documentation for pytest it seems it should create the executable.  If it doesn’t it will certainly break a lot of stuff, which is not good.

-Scott

Re: [Fink-devel] [Fink-users] Can't install xemacs

[Scott H. <sha...@us...>]

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 
>> 


[Fink-devel] qt5 does not build, missing functions?

[Scott H. <sha...@us...>]

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

Re: [Fink-devel] The following 230 additional packages will be installed: ????

[Ebrahim M. <eb...@ma...>]

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
> 
> 
> 

Re: [Fink-devel] The following 230 additional packages will be installed: ????

[Daniel M. <dm...@ne...>]

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

[Fink-devel] The following 230 additional packages will be installed: ????

[Ebrahim M. <eb...@ma...>]

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] gtk+2

[Scott H. <sha...@us...>]

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

Re: [Fink-devel] [Fink-users] install GitHub requires go?

[Hanspeter N. <fi...@sn...>]

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

Re: [Fink-devel] [Fink-users] install GitHub requires go?

[Daniel M. <dm...@ne...>]

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

[Fink-devel] install GitHub requires go?

[Scott H. <sha...@us...>]

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.
% 

Re: [Fink-devel] gcc12 will not build with multiple jobs

[<fi...@sn...>]

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.

Re: [Fink-devel] gcc12 will not build with multiple jobs

[Scott H. <st...@ma...>]

> 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
> 

Re: [Fink-devel] gcc12 will not build with multiple jobs

[<fi...@sn...>]

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

Re: [Fink-devel] gcc12 will not build with multiple jobs

[Scott H. <st...@ma...>]

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.

Re: [Fink-devel] gcc12 will not build with multiple jobs

[Scott H. <st...@ma...>]

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.

Re: [Fink-devel] gcc12 will not build with multiple jobs

[<fi...@sn...>]

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